USF Libraries
USF Digital Collections

Shyness in the context of reduced fear of negative evaluation and self-focus

MISSING IMAGE

Material Information

Title:
Shyness in the context of reduced fear of negative evaluation and self-focus a mixed methods case study
Physical Description:
Book
Language:
English
Creator:
Watson, Freda S
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Social anxiety
Social phobia
BFNE-S
Confirmatory factor analysis
Qualitative analysis
Dissertations, Academic -- Educational Measurement and Research -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: This mixed methods case study examined the effect of reduced fear of negative evaluation and self focus on behaviors related to shyness in a church environment. A sample of 239 members, regular attenders, and visitors completed a survey, consisting of the Brief Fear of Negative Evaluation-Straightforward (BFNE-S) Scale; two checklists measuring perceived acceptance and levels of comfort in situations known to be difficult for shy people; and extended response questions regarding thoughts, feelings, and behaviors in six church situations. Confirmatory factor analysis revealed the BFNE-S (General and Context-specific) had acceptable fit compared with previous studies, and descriptive statistics were similar to those of previous studies. Lower self-reported levels of fear of negative evaluation and higher levels of perceived comfort, but not acceptance, in the church setting were found to be statistically significant, although the effect size was negligible.A repeated measures ANOVA revealed no statistically significant difference for gender or race for individuals in the church setting compared to the non church setting. A multiple regression failed to reveal a statistically significant relationship between depth and breadth of involvement in church activities and reduced fear of negative evaluation. The Clark-Wells (1995) model of social phobia explained 62% of self-reported behaviors, thoughts, and feelings of individuals with high levels of shyness when in social situations in the church setting. A statistically significant difference was found between focus of attention and quality of thought scores for individuals with minimal to low levels of shyness and high levels of shyness. To explore further the validity of scores obtained with the BFNE-S, it would be useful to conduct a study in different environments and seek to understand individuals in those environments with high and low fear of negative evaluation.Future research regarding the church setting should utilize a sample with fewer long-term members and regular attenders. Additionally, future studies could probe how religious beliefs help people cope with difficult situations, in particular shyness.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2009.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Freda S. Watson.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 239 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002069243
oclc - 608046337
usfldc doi - E14-SFE0003066
usfldc handle - e14.3066
System ID:
SFS0027382:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200409Ka 4500
controlfield tag 001 002069243
005 20100420111048.0
007 cr mnu|||uuuuu
008 100420s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003066
035
(OCoLC)608046337
040
FHM
c FHM
049
FHMM
090
LB3051 (Online)
1 100
Watson, Freda S.
0 245
Shyness in the context of reduced fear of negative evaluation and self-focus :
b a mixed methods case study
h [electronic resource] /
by Freda S. Watson.
260
[Tampa, Fla] :
University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 239 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
520
ABSTRACT: This mixed methods case study examined the effect of reduced fear of negative evaluation and self focus on behaviors related to shyness in a church environment. A sample of 239 members, regular attenders, and visitors completed a survey, consisting of the Brief Fear of Negative Evaluation-Straightforward (BFNE-S) Scale; two checklists measuring perceived acceptance and levels of comfort in situations known to be difficult for shy people; and extended response questions regarding thoughts, feelings, and behaviors in six church situations. Confirmatory factor analysis revealed the BFNE-S (General and Context-specific) had acceptable fit compared with previous studies, and descriptive statistics were similar to those of previous studies. Lower self-reported levels of fear of negative evaluation and higher levels of perceived comfort, but not acceptance, in the church setting were found to be statistically significant, although the effect size was negligible.A repeated measures ANOVA revealed no statistically significant difference for gender or race for individuals in the church setting compared to the non church setting. A multiple regression failed to reveal a statistically significant relationship between depth and breadth of involvement in church activities and reduced fear of negative evaluation. The Clark-Wells (1995) model of social phobia explained 62% of self-reported behaviors, thoughts, and feelings of individuals with high levels of shyness when in social situations in the church setting. A statistically significant difference was found between focus of attention and quality of thought scores for individuals with minimal to low levels of shyness and high levels of shyness. To explore further the validity of scores obtained with the BFNE-S, it would be useful to conduct a study in different environments and seek to understand individuals in those environments with high and low fear of negative evaluation.Future research regarding the church setting should utilize a sample with fewer long-term members and regular attenders. Additionally, future studies could probe how religious beliefs help people cope with difficult situations, in particular shyness.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Co-advisor: Anthony J. Onwuegbuzie, Ph.D.
Co-advisor: Robert F. Dedrick, Ph.D.
653
Social anxiety
Social phobia
BFNE-S
Confirmatory factor analysis
Qualitative analysis
690
Dissertations, Academic
z USF
x Educational Measurement and Research
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3066



PAGE 1

Shyness in the Context of Reduced F ear of Negative Evaluation and SelfFocus: A Mixed Methods Case Study by Freda S. Watson A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Educational Measurement and Research College of Education University of South Florida Co-Major Professor: Ant hony J. Onwuegbuzie, Ph.D. Co-Major Professor: R obert F. Dedrick, Ph.D. John M. Ferron, Ph.D. Jack Darkes, Ph.D. Date of Approval: May 19, 2009 Keywords: Social Anxiety, Social Phobia, BFNE-S, Confirmatory Factor Analysis, Qualitative Analysis Copyright 2009, Freda S. Watson

PAGE 2

DEDICATION This dissertation is dedicated to my beloved Heavenly Father and my beloved daughter, who taught me how to love and how to be loved.

PAGE 3

ACKNOWLEDGEMENTS It would be impossible to menti on and adequately thank everyone who helped along this long path of higher educ ation. All I can say is I sincerely appreciate the dedication, insp iration, and assistance of all my professors and the love and support of family and friends. You have each played a very special part, without which the journey would have been impossible.

PAGE 4

i TABLE OF CONTENTS LIST OF TABLES ............................................................................................................. iv LIST OF FIGURES ........................................................................................................... vi ABSTRACT ...................................................................................................................... vii CHAPTER 1: INTRODUCTION ........................................................................................ 1 Statement of Problem ............................................................................................ 1 Background ........................................................................................................... 2 Theoretical Framework .......................................................................................... 5 Rationale of Study ................................................................................................. 9 Purpose of Study ................................................................................................. 11 Research Questions ............................................................................................ 12 Quantitative Research Questions ............................................................ 12 Mixed Methods Research Questions ....................................................... 12 Research Hypotheses ............................................................................. 13 Educational Significance ..................................................................................... 14 Definition of Terms .............................................................................................. 16 Limitations of Study ............................................................................................. 20 Delimitations of the Study .................................................................................... 26 Organization of Remaining Chapters .................................................................. 26 CHAPTER 2: REVIEW OF THE RELATED LITERATURE ............................................. 27 Overview ............................................................................................................. 27 Background ......................................................................................................... 28 Prevalence of Shyness ........................................................................................ 28 Recurrent Themes in the Literature ......................................................... 28 Traits and Temperaments ............................................................ 29 State-Trait Anxiety ....................................................................... 30 Environmental, Developmental, and Genetic Influences ............. 31 Gender Differences ...................................................................... 34 Cultural Differences ..................................................................... 36 Trends in Research Concerning Shyness ........................................................... 36 Measurement of Shyness .................................................................................... 39 Theoretical Framework ........................................................................................ 41 Cognitive Models ..................................................................................... 41 Rapee and Heimberg Model ........................................................ 41 Clark and Wells Cognitive Model ............................................... 42 Similarities and Differences between the Two Models ................ 43 Clark and Wells Cognitive Treatment Model in Depth ................. 45 Deriving an Idiosyncratic Version of the Model ........................... 45 Manipulation of Self-Focused Attention and Safety Behaviors ................................................................................ 45

PAGE 5

ii Video and Audio Feedback………………………………………….46 Shift of Attention and Interrogation of the Social Environment ................................................................................ 46 Dealing With Anticipatory and Post Event Processing ............... 47 Dealing with Assumptions ........................................................... 47 Typical Behaviors of Shy People ............................................................ 48 Key Environmental Influences .................................................... 50 Nine Situations Most Difficult For Shy People ............................ 50 Similarity to Six Activities of Church Life ..................................... 51 Two Key Environmental Influences That Maintain Shyness ................................ 53 Fear of Negative Evaluation ........................................................ 53 Self-Focused Attention ................................................................ 54 The Social Climate of the Church ........................................................... 56 Rationale ............................................................................................................ 59 Summary ............................................................................................................ 60 CHAPTER 3: METHODS ................................................................................................ 64 Selection Eligibility Criteria ................................................................................ 64 Participants ......................................................................................................... 65 Quantitative Phase ................................................................................. 65 Qualitative Phase ................................................................................... 67 Ethical Considerations ........................................................................................ 69 Instruments ......................................................................................................... 70 Brief Fear of Negative Evaluation Scale-Straightforward ...................... 70 Questions to be Excluded from Analysis ............................................... 79 Qualitative Questions ............................................................................. 79 Procedures ......................................................................................................... 80 Pragmatist Procedure ........................................................................... 80 Quantitative Procedure .......................................................................... 81 Internal Validity ................................................................................................... 84 External Validity .................................................................................................. 85 Qualitative Procedure ........................................................................................ 86 Mixed Methods Procedures ............................................................................... 89 Threats to Legitimation ........................................................................... 89 Analyses ............................................................................................................ 92 Quantitative Analyses ............................................................................ 92 Qualitative Analysis ................................................................................ 99 Combined Quantitative and Qualitative Analysis ................................. 101 CHAPTER 4: RESULTS ............................................................................................... 104 Introduction ........................................................................................................ 104 Participants ........................................................................................... 104 Research Question 1 ......................................................................................... 106 Research Hypothesis 1 ..................................................................................... 106 Reliability ............................................................................................... 109 Confirmatory Factor Analysis ................................................................. 110 Model Specification, Input Data, and Model Estimation ............. 110 One-factor Model, Fit Indices ..................................................... 112 Parameter Estimates, One-factor Model. ................................... 113 Modification Indices for One-factor Model ................................. 115

PAGE 6

iii Two-factor Model ...................................................................... 117 Two-factor Model, Measures of Fit. ........................................... 120 Parameter Estimates, Two-factor Model ................................... 120 Modification Indices for Two-factor Model ................................. 123 Comparisons across All Four Models ........................................ 126 Research Question 2 ......................................................................................... 128 Research Hypothesis 2 ..................................................................................... 128 Research Question 3 ......................................................................................... 129 Research Hypothesis 3 ..................................................................................... 129 Research Question 4 ......................................................................................... 129 Research Hypothesis 4 ..................................................................................... 130 Research Question 5 ......................................................................................... 130 Research Hypothesis 5 ..................................................................................... 130 Research Question 6 ......................................................................................... 133 Research Hypothesis 6 ..................................................................................... 133 Research Question 7 ......................................................................................... 135 Research Hypothesis 7 ..................................................................................... 136 Research Question 8 ......................................................................................... 138 Research Hypothesis 8 ..................................................................................... 139 Research Question 9 ......................................................................................... 151 Research Hypothesis 9 ..................................................................................... 152 CHAPTER 5: DISCUSSION ......................................................................................... 157 Overview of Method and Summary of Findings ................................................. 157 Research Question 1. ........................................................................................ 160 Research Question 2. ........................................................................................ 161 Research Questions 3 and 4. ............................................................................ 164 Research Question 5 ......................................................................................... 165 Research Question 6 ......................................................................................... 165 Research Question 7 ......................................................................................... 165 Research Question 8 ......................................................................................... 166 Research Question 9 ......................................................................................... 168 Comparison of Findings with Theoretical Framework ....................................... 170 Limitations ......................................................................................................... 173 Threats to Internal Validity ................................................................................. 175 Threats to External Validity ............................................................................... 179 Benefits of Mixed Methods Research ................................................................ 182 Implications for Future Research....................................................................... 183 Conclusions ....................................................................................................... 185 REFERENCES .............................................................................................................. 190 APPENDICES Appendix A. Personal Concerns and Issues Scale .......................................... 204 Appendix B. Formula for Effect Size (Dunlap, Cortina, Vaslow & Burke, 1996) ................................................................................................................ 225 Appendix C. Full Text of Write-In Responses.................................................... 223 Appendix D. Missing Data ................................................................................. 238 ABOUT THE AUTHOR ....................................................................................... End Page

PAGE 7

iv LIST OF TABLES Table 1 Beliefs about Self Typically Held by Shy Persons ........................................ 6 Table 2 Ten Situations Most Difficult for Shy People................................................. 8 Table 3 Common Vulnerabilities Underlying Shyness and Social Phobia ............... 33 Table 4 Hits for Shyness/Social Phobia Measures as Recorded in PsychINFO ..... 40 Table 5 Comparison of Clark and Wells and Rapee and Heimberg Models ........... 44 Table 6 Symptoms of shyness (Henderson & Zimbardo, 2001) .............................. 49 Table 7 Ten Most Difficult Situations for Shy People............................................... 51 Table 8 Areas of Commonality Between the Nine Situations and Five Church Life Activities ............................................................................................... 52 Table 9 Norms Reported for BFNE Scale ............................................................... 68 Table 10 Research Regarding Validity and Reliability of Scores Obtained from the BFNE-S ................................................................................................ 72 Table 11 Components of the Personal Concerns and Issues Scale ......................... 74 Table 12 BFNE-S Items ............................................................................................. 76 Table 13 Other Components of the PCI Scale .......................................................... 77 Table 14 Thoughts, Feelings, and Behaviors Questions ........................................... 79 Table 15 Primary Distinguishing Characteristics of the Pragmatist Paradigm ........... 81 Table 16 Basic Types of Designs for Case Studies................................................... 87 Table 17 Dimensions of Research Design ................................................................ 93 Table 18 Analysis Plan .............................................................................................. 96 Table 19 Map for Content Analyses of Responses to Thoughts, Feelings, and Behaviors Questions/Interviews ............................................................... 100 Table 20 Seven Stages of Mixed Analysis Process and Research Questions ........ 103 Table 21 Demographics of Respondents to Personal Concerns and Issues Survey ...................................................................................................... 105 Table 22 Descriptive Statistics for BFNE-S, General and BFNE-S, Context-specific ........................................................................................ 107 Table 23 Norms Reported for BFNE Scale (ordered by mean) ............................... 108 Table 24 Fit Indices for BFNE-S, Generaland Context-specific, Oneand TwoFactor Models ........................................................................................... 113 Table 25 Model Results for the One-Factor BFNE-S, General, and BFNE-S, Context-specific ........................................................................................ 115 Table 26 Five Highest Modification Indices One-factor BFNE-S, General and Context-specific Versions ......................................................................... 117 Table 27 Model Results for the Two-Factor BFNE-S, General, and BFNE-S, Context ..................................................................................................... 122 Table 28 Five Highest Modification Indices Two-factor BFNE-S, General and Context-specific versions .......................................................................... 125 Table 29 Comparison of Modification Indices for the Four Models in This Study .... 127 Table 30 Descriptive Statistics for Items on the Perceived Acceptance Checklist .. 132 Table 31 Descriptive Statistics for the Comfort Scale, General and Context-specific ........................................................................................ 134

PAGE 8

v Table 32 Descriptive Statistics for Context Variables (Membership, Attendance, Activities and BFNE-S, Context-specific Scores) ..................................... 138 Table 33 Standardized Multiple Regression Coefficients for Context Variables (Membership, Attendance, Activities, and Friends) .................................. 138 Table 34 Shyness Levels for Research Questions 8 and 9 ..................................... 139 Table 35 Coding Guidelines Based on the Clark and Wells (1995) Model of Social Phobia ............................................................................................ 140 Table 36 Percent of All (270) Comments Irrelevant and Relevant to Theory .......... 142 Table 37 Percent of Comments Consistent with the Clark and Wells (1995) Model by Shyness Level ........................................................................... 143 Table 38 Examples of Comments Consistent and Inconsistent With Theory (Clark and Wells’ [1995] Model of Social Phobia} .................................... 144 Table 39 Percent of Phrases Consistent with the Clark and Wells (1995) Model by Shyness Level ..................................................................................... 146 Table 40 Descriptive Statistics for Percent of Participant Comments Consistent With Theory by Level ................................................................................ 147 Table 41 Coding Results by Unit of Analysis ........................................................... 151 Table 42 Results of Coding for Focus of Attention and Thought Quality ................. 152 Table 43 Examples of Coding for Focus of Thought and Quality of Thought .......... 154 Table 44 Summary of Major Findings ...................................................................... 158 Table 45 Effect Sizes Reported for BFNE Scale (ordered by effect size) ............... 162 Table 46 Participant Ages in Previous Studies of the BFNE Scale ......................... 170

PAGE 9

vi LIST OF FIGURES Figure 1 Developing an Idiosyncratic Version of the Clark-Wells Treatment Model .......................................................................................................... 63 Figure 2 Sampling Plan ............................................................................................ 66 Figure 3 One-factor BFNE-S .................................................................................. 111 Figure 4 Two-factor BFNE-S .................................................................................. 119 Figure 5 Percent of Comments Consistent and Inconsistent with the Clark and Wells (1995) Model of Social Phobia ........................................................ 143 Figure 6 Percent of Phrases Consistent with the Clark and Wells (1995) Model by Shyness Level ..................................................................................... 146 Figure 7 Percent of Participant Comments Consistent and Inconsistent with the Clark and Wells Model .............................................................................. 148 Figure 8 Relationship Between Focus of Attention and Level of Shyness (as measured by the BFNE-S, Context-specific) ............................................ 153 Figure 9 Relationship Between Quality of Thought and Level of Shyness (as measured by the BFNE-S, Context-specific) ............................................ 153

PAGE 10

vii SHYNESS IN THE CONTEXT OF REDUCED FEAR OF NEGATIVE EVALUATION AND SELF-FOCUS: A MIXED METHODS CASE STUDY FREDA S. WATSON ABSTRACT This mixed methods case study examin ed the effect of reduced fear of negative evaluation and self focus on behavio rs related to shyness in a church environment. A sample of 239 member s, regular attenders, and visitors completed a survey, consisting of t he Brief Fear of Negative EvaluationStraightforward (BFNE-S) Scale; two che cklists measuring perceived acceptance and levels of comfort in situations k nown to be difficult for shy people; and extended response questions regarding th oughts, feelings, and behaviors in six church situations. Confirmatory factor analysis reve aled the BFNE-S (General and Contextspecific) had acceptable fit compared wit h previous studies, and descriptive statistics were similar to those of previ ous studies. Lower self-reported levels of fear of negative evaluation and higher le vels of perceived comfort, but not acceptance, in the church setting were found to be statistically significant, although the effect size was negligible A repeated measures ANOVA revealed no statistically significant difference fo r gender or race for individuals in the church setting compared to the non church setting. A multiple regression failed

PAGE 11

viii to reveal a statistically significant relationship between depth and breadth of involvement in church activities a nd reduced fear of negative evaluation. The Clark-Wells (1995) model of so cial phobia explai ned 62% of selfreported behaviors, thoughts, and feelings of individuals with high levels of shyness when in social situations in the c hurch setting. A statistically significant difference was found between focus of att ention and quality of thought scores for individuals with minimal to low levels of shyness and high levels of shyness. To explore further the valid ity of scores obtained with the BFNE-S, it would be useful to conduct a study in diffe rent environments and seek to understand individuals in those environments with high and low fear of negat ive evaluation. Future research regarding the church se tting should utilize a sample with fewer long-term members and regular attenders. Additionally future studies could probe how religious beliefs help people cope with difficult situations, in particular shyness.

PAGE 12

1 CHAPTER 1 INTRODUCTION Statement of the Problem A great deal is known about shyness, its consequences and correlates, and the effectiveness of various treatm ents (Crozier, 2001; Crozier & Alden, 2001a; Heimberg, Hope, Liebowits, & Sc hneier, 1995; Zimbardo, Pilkonis, & Norwood, 1974). Additionally, research on measures of shyness has been ongoing for decades and has resulted in a number of instrum ents, many with excellent psychometric properties and a long history of use in research (Heimberg et al., 1995; Orsillo, 2001). Mo st of the research, however, has been conducted with college and unive rsity students in laboratory or clinical settings. Scant research was found regarding how shyness manifests itself in other environments and with other populations. It appeared worth investigating how shyness affects individuals in an envir onment—that is, the church setting—where two of the most thoroughly researched correlates were presumed to occur to a lesser extent than the compet itive climate of the university. The two correlates were fear of negative evaluation and attent ional focus on the self. According to Weeks et al. (2005), fear of negative evaluation is the core feature of social anxiety disorder. Attentional focus is a key feature in research investigating shyness and social interactions (Spurr & Stopa, 2002).

PAGE 13

2 Background As Crozier and Alden (2001a., p. 4) observe, shyness is a problem for many people. It is generally believed to occur in between 40% and 50% of individuals at some point in the life span and to varying degrees (Carducci, 2000; Zimbardo et al., 1974). For some, t he problems are seve re enough that the individual thinks of his or her shyness as an illness because it interferes with one’s ability to live a normal life. As Henderson and Zimbardo (2001) indica te, shyness affects many areas of an individual’s life. According to Henderson and Zimbardo, shyness erects barriers in meeting and communicati ng with people and in becoming better acquainted. Shyness also acts as an obsta cle in functioning in small groups, in contacts with authority figures, and in asse rting oneself. A long line of research findings (e.g., Crozier, 2001, Crozier & Alden, 2001a, 2001b; Weeks et al., 2005) consistently indicate that shy people ty pically behave in an inhibited or overly restrained fashion, their approach too mu ch of life is passive, and they avoid situations that cause them discomfort or fear. Furthermore, some of their outward behaviors, such as a low speaking voice, either inhibited or excessive body movement and expression, and ot her nervous behaviors are often misinterpreted as intentional reserve or coolness (Henderson & Zimbardo, 2001). Given that shyness can affect so m any areas of an individual’s life, it seems logical that the descriptive phase of research in shyness—that is, clinical observations of medical and psychology practitioners—began as early as 1896 (Jones, Cheek, & Briggs, 1986). Since that time, a sizeable body of research

PAGE 14

3 has been constructed regarding shyness. Much of that early research was conducted with college students and in laboratory or clinical settings such as with clients who presented for mental health co unseling at clinics (Collins, Westra, Dozois, & Stewart, 2005). In the late 1980s a change in diagnostic criteria led to more research being conducted by ment al health professionals (McDaniel, 2003). Additionally, more recent research has sometimes involved clinical rather than analogue designs “. . in which hi gh and low socially anxious non-patients are compared” (Clark 2001, p. 411). The foregoing notwithstanding, search of the literatur e revealed that research still seems to be limited r egarding settings and participants. Investigating how shyness operates in previously unexplored settings and with participants atypical with respect to t he individuals usually involved in analogue and clinical designs should add to our k nowledge of shyness. It was believed that such research might help uncove r ways to reduce the difficulties that shyness causes in everyday life and in the field of education as well. In particular, and of special relevance for the current study, scant research could be found that investigated how we ll what seems to be one of the most thoroughly researched cogniti ve theories of shyness, the Clark and Wells (1995) cognitive model, performs across settings and individuals. Moreover, careful search revealed no study that has expl ored how well the Clark and Wells model explains the thoughts, feelings, and behaviors of shy individuals in settings with less potential for negative evaluation and the incentive to focus outward, rather than on the self, during social interactions.

PAGE 15

4 Briefly, the Clark and Wells model c oncerns the mental processing in which persons with social phobia engage upon entering a “feared social situation” (Clark, 2001, pp. 405-406). The mental processing can include “excessively high standards for social performance . conditional beliefs concerning the consequences. . [of cert ain actions] . and unconditional negative beliefs about the self” (Clark, 2001, pp. 405-406). The second part of the model addresses the effect s of the anxiety that typi cally occur just prior to entering a social situation and the negat ive rumination that often occurs afterwards (Clark, 2001). Another point that makes the current st udy worthwhile is that the majority of research with one of the most oft en-used measures of shyness—the Brief Fear of Negative Evaluation (BFNE) sca le—has been in typical settings (e.g., undergraduate classes or at mental health clinics), with college-age participants or clinical populations and more re cently with non-student, non-clinical populations (Duke, Krishnan, Faith, & Storch, 2006). The current research helped quantify the explanat ory power of the Clark and Wells model in a previously unexplored environment, that is the church setting. The study also evaluated the psychometric properties of the BFNE in a non-student, non-clinical sample of a previously unstudied populat ion (i.e., church members, regular attenders, and church visitors ). One weakness of the Duke et al. (2006) study was the use of a non-random convenience sa mple. Findings might help extend the generalizability of scores from the BFNE across settings.

PAGE 16

5 The methods utilized for the current st udy were appropriate for the topic. To confirm, or disconfirm, that the chur ch setting (the environment utilized in the study) holds less potential for negative eval uation, it was necessary to obtain comparison measures with the BFNE. Participants’ levels of shyness also had to be determined. Quantitative methods are suited to this purpose. To examine the extent to which the Clark and Wells model explains shy behavior in the setting used in this study it was necessary to use qualitative methods, specifically openended questions. The rich description of thoughts, feelings, and behaviors obtained thr ough open-ended questions was necessary to ascertain the extent to which t hose thoughts, feelin gs, and behaviors are explained by the theory. A similar situat ion holds with exploring the effects of attentional focus in the church setting. A mixed methods case study design was selected to fulfill both purposes. Johnson and Turner (2003, p. 299) state that “the fundamental principle of mixed methods research . [is that] . methods should be mixed in a way that has comp lementary strengths and non-overlapping weaknesses. The current study typified this principle. Theoretical Framework As mentioned earlier, there is a well-developed body of research on shyness. Relevant to the current study is the fact that existing research includes: various explanatory models (Heinrichs et al., 2006), in particular the Clark and Wells model (Clark & Wells, 1995); consensus on the situations most di fficult for shy people (Crozier, 2001); and

PAGE 17

6 the finding in a review of measures of shyness and social anxiety that scores obtained with the BFNE have very good psychometric properties (Orsillo, 2001). The Clark and Wells model was chosen for the current study because it focuses on how fear of negative evaluat ion and negative self-focused attention can maintain social phobia (Clark & We lls, 1995). The first part of the model focuses on mental processing, safety behaviors, somatic and cognitive symptoms, and processing of external social cues ex perienced by individuals with social phobia (or shy persons) upon enter ing a feared social situation. The second part of the model seeks to explai n processing before and after social situations. According to Clark and We lls, extremely shy persons accumulate three categories of beliefs about themselv es and social interaction. Table 1 (Clark, 2001) presents these three cat egories of beliefs and some typical examples. Table 1 Beliefs about Self Typically Held by Shy Persons Category Typical Examples Excessively high standards for social performance “I must not show any signs of weakness”; “I must always sound intelligent and fluent”; “I should only speak when other people pause”; “I should always have something interesting to say.”

PAGE 18

7 Conditional beliefs concerning the consequences of performing in a certain way “If I disagree with someone, they will think I am stupid/will reject me”; “If my hands shake/I blush/or show other signs of anxiety, people will think I am incompetent/odd/stupid”; “I f I am quiet, people will think I am boring”; “If peopl e get to know me, they won’t like me.” Unconditional negative beliefs about the self “I’m odd/different”; “I’m unlikable/unacceptable”; “I’m boring”; “I’m stupi d”; and “I’m different”. Another theoretical foundation of t he current study was the research finding that certain situations are particu larly difficult for shy persons. Table 2 presents the 10 situations that research has indicated are most problematic for shy individuals (Crozier, 2001) As explained below, t hese situations occur in many of the interactions in which one engages when participating in churchrelated activities, such as attending wo rship services and social functions.

PAGE 19

8 Table 2 Ten Situations Most Difficult for Shy People ** Being the focus of attention Large groups Small groups Authority figures Social situations in general New interpersonal situ ations in general Strangers Situations where asse rtiveness is required Being evaluated or co mpared with others An opposite sex group or a me mber of the opposite sex Note : Adapted from Crozier (2001). Understanding shyness : Psychological perspectives China: Palgrave. Social interactions in the church setting were believed to hold reduced potential for fear of negative evaluation and self-focused attention. The reasons for this were the cultural rules for t he church community, which are based on the teachings of the Bible. Biblical teachings include the command to love and accept one another (John 13:34-35, New York International Bible Society, 1978) and to put others first (Rom ans 12:10, New York International Bible Society, 1978).

PAGE 20

9 Another fact making the church setting ideal for the current study was that persons attending church interact in six situations that are highly similar to the situations research has indicated are diffi cult for shy people. Those situations are worship services, small group activities social events (e.g., weddings, baby showers), relationships with friends, interacting with authority, and when performing a volunteer job (e.g., taking the offering or teaching a class). Besides being valuable as a means to explore how theory performs across settings, it was also believed that the church setting should be helpful to seek further evidence for the validity of sco res from the BFNE. As stated earlier, most of the validation work for sc ores obtained with the BFNE has been conducted with college-age youn g adults and, in more recent years, with clinical populations (Duke et al., 2006). Finally, the current study was also based upon the author’s long-term experience in the church and in the hidden culture of shyness. Nineteen of the 26 years the author has been an active c hurch member have included personal observations of and discussions with other sh y church members. Many of those conversations specifically addressed t he difficulties shy individuals have in establishing social connections in the church setting. Rationale of Study As stated earlier, shyness is a signi ficant social problem, affecting between 40% and 50% of the popul ation at some time in the life span (Carducci, 2000; Zimbardo et al., 1974). Shyness c an negatively affect many areas of an individual’s life, often to a severe degree (Henderson & Zimbardo, 2001).

PAGE 21

10 Providing more evidence either confirmi ng or suggesting alternative explanations for one of the most well-re searched theories of shyne ss and further investigation of the psychometric properti es of one of the most often-used measures of shyness could assist with on-going e fforts to alleviate this problem. Research indicates shyness can be affected by many environmental conditions (Henderson & Zimbardo, 2001) Two of the most critical environmental factors are f ear of negative evaluation (Weeks et al., 2005) and self-focus (Spurr & Stopa, 2002). Shyness has been studied most often either in the highly competitive environment of college, where evaluation is central, or in clinical populations where self-focus is obviously par amount. In responding to the questions about an individual’s thoughts, feelings, and behaviors and/or in the interviews, participants reported on t heir experiences of shyness in the church setting, which was presumed to have less potential for negative evaluation and less focus upon the self. Co mparing results of the participant responses with existing theory, which is based upon data collected in traditional settings, helped measure the extent of c onvergence with theory. Those analyses helped extend the theory with respect to this previously unexplored setting. Regarding measures of shyness, the cu rrent body of literature on shyness has been constructed using measures devel oped primarily with college students and/or, in the case of clinical psycholog ists and psychiatrists, clinical populations (Duke et al., 2006). Existing theor y has been heavily influenced by these measures. Further validation work on scores from a commonly used measure (i.e., the BFNE) was needed.

PAGE 22

11 Purpose of Study The current study addressed nine purposes: 1. To evaluate the psychom etric properties of the BF NE-S in a non-student, nonclinical sample of a prev iously unstudied population. 2. To compare levels of perceived fear of negative evaluation inside and outside the church setting. 3. To compare the effect of gender on perceived fear of negative evaluation in the church setting compared to the non church setting. 4. To compare the effect of race on per ceived fear of negative evaluation in the church setting compared to the non church setting. 5. To compare levels of perceived a cceptance by people inside and outside the church setting. 6. To understand how shyness manifests itself in an environment believed to induce higher comfort levels. 7. To understand how context-specific issu es (extent of involvement in church activities) are related to self-reported f ear of negative evaluat ion in the church setting. 8. To seek confirmation of theory or alternative explanations for behaviors the theory addresses via examining the extent to which existing theory explains the thoughts, feelings, and behaviors of shy individuals in the church setting. 9. To seek confirmation of theory or alternative explanations for behaviors via examining the extent to whic h attentional focus is rela ted to self-reported levels of fear of negative evaluation.

PAGE 23

12 The first seven purposes were addressed using quantitative methods. The eighth and ninth purposes were addressed using mixed methods. Research Questions Quantitative Research Questions The following seven research questions were addressed in the quantitative portion of this study: 1. What are the psychometric properti es of the BFNE-S, General and Contextspecific, in the church setting? 2. What is the difference in perceived fear of negative evaluat ion in the church setting compared to the non church setting? 3. What is the difference in perceived fear of negative evaluat ion in the church setting compared to the non church setting for males and females? 4. What is the difference in perceived f ear of negative evaluat ion in the church setting compared to the non church setting fo r individuals of di fferent races? 5. What is the difference in per ceived acceptance between people inside and outside the church setting? 6. What is the difference in self-repor ted levels of comfort outside the church setting and inside the church setting? 7. To what extent do context-specific issu es relate to self-reported levels of fear of negative evaluation? Mixed Methods Research Questions The qualitative portion of this st udy addressed the following research questions:

PAGE 24

13 8. How do shy people typically think, feel, and behave in an environment hypothesized to have less fear of negative evaluation and self-focus? 9. To what extent is self-reported f ear of negative evaluation associated with attentional focus upon self and negative qu ality of thought in the six church situations? Research Hypotheses The current study tested nine resear ch hypotheses. These hypotheses are presented below. Research Hypothesis 1. The BFNE-S, General and Context-specific, displays psychometric properties in the sample that are similar to those demonstrated for other populations taken fr om university or clinical settings. Research Hypothesis 2. Perceived FNE is lower in the church setting compared to the non-church setting. Research Hypothesis 3. The difference in FNE between the church and non-church setting is the same for males as for females. Research Hypothesis 4. The difference in FNE between the church and non-church setting is the sa me for different races. Research Hypothesis 5. Levels of perceived acceptance by people in the church setting are higher t han are the levels of perce ived acceptance by people outside the church setting. Research Hypothesis 6. Levels of comfort perceived by people in the church setting are higher than are the levels of co mfort outside the church setting.

PAGE 25

14 Research Hypothesis 7. Greater depth and breadth of involvement in church activities are associated with reduced self-report ed fear of negative evaluation. Research Hypothesis 8. At least 75% of individuals with high levels of FNE report thoughts, feelings, and behaviors re lated to six church situations that are consistent with the Clark and Wells model and that will be at least 10 % more than those with low levels of FNE. Note: After a review of the research, it was decided to select 75% as a best esti mate because it seems likely that, for individuals with moderate to high levels of shyness, the habits of thought delineated in the Clark and Wells model will have become entrenched and that even while in an environment that is perceiv ed as more accepting, the individual will respond with their characteristic behaviors, thoughts, and feelings. It is believed that the environment will have a slight moderating effect but only for perhaps one fourth of the si tuations being analyzed. Research Hypothesis 9. Focus upon self and negat ive quality of thought related to the six church situations ar e associated with higher levels of selfreported fear of negative evaluation. Educational Si gnificance The current study contributed to the field of measurement in that it provided additional evidence regarding the generalizability of scores yielded by the BFNE across settings. It was hoped t hat such information would help inform future research in this area. As stated earlier, most of the existing research was conducted with participants who were ei ther college students or who had

PAGE 26

15 presented for psychiatric or psychological counseling. Participants in the current study were non-students. Additionally, most of the previous research was conducted in traditional settings, such as a university or in a clinic. The current study utilized the church setting, and a t horough search revealed no study that had utilized such a setting. The curr ent study has provided supplemental evidence regarding the utility of a co mmonly used measure of shyness. Additionally, education occurs in many different kinds of settings, and learning how shyness operates in a previously unexp lored setting is helpful to the field of education. Finally, the present investigation was unique in an important way. Specifically, it represents what is believed to be one of the first studies of shyness utilizing mixed methods res earch techniques. As noted by Onwuegbuzie and Leech (2004, 2005), comb ining quantitativ e and qualitative research enables researchers to be more flexible, comprehensive, holistic, and, above all, integrative in thei r investigative techniques, as they attempt to address a range of complex research questions that come to the fore. Further, by conducting mixed methods studies, resear chers are in a better position to combine empirical precision with descripti ve precision (Onwuegbuzie, 2003a). In addition, by utilizing a pragmatist lens (i.e., using both quantitative and qualitative techniques), rather than a single lens (i.e., conducting monomethod studies), researchers are in a better position to zoom in to microscopic detail or to zoom out to indefinite scope (Willems & Raush, 196 9). This flexibility in perspective

PAGE 27

16 leads to a broader understanding of the par ticipants, which is an important goal of pragmatist research (Tashakkori & Teddlie, 2003). Definition of Terms Adult. For the purpose of this study, an adul t is an individual more than 21 years of age. Attentional focus. Attentional focus, for this study, is the object upon or toward which an individual is dire cting his/her a ttention. Clark and Wells cognitive m odel of social phobia. The Clark and Wells (1995) cognitive model of social phobia em phasizes the role of the shift in attention that occurs when an individua l perceives, whether accurately or inaccurately, that he or she is about to be evaluated negatively. Continuum model. In this study, the continuum model refers to the assumption that shyness, social anxiety, and social phobia are more alike than different and that research findings relating to one construct can, with caution, be applied to the other constructs. This assumption is based on the findings of Rapee and Heimberg (1997, p. 742) that there is a “continuum from low to extreme degrees of concern over social evaluation and that shyness, social phobia, and avoidant personality disorder are on the low, middle, and upper ranges of that continuum, with a cons iderable degree of overlap.” Fear of negative evaluation. In this study, fear of negative evaluation is defined as “. . apprehension about other s’ evaluations, distress over their

PAGE 28

17 negative evaluations, avoidance of evaluativ e situations, and the expectation that others . [will] . eval uate oneself negatively” (Watso n & Friend, 1969, p. 449). Friend. A friend is “a person whom one knows, likes and trusts” (Berube, 1982, p. 534). Interacting with authority. For the purposes of this research, interacting with authority is defined as occasions when church members, regular attenders, or visitors speak to the pastor or other officially designated leaders in the church, such as Sunday School teachers, ushers, or ministers of music. Mixed methods study. A mixed methods study is “. . a type of research design in which QUAL and QUAN approache s are used in type of questions, research methods, data collection, and anal ysis procedures, and/or inferences” (Tashakkorri & Teddlie, 2003, p. 711). Self-focused attention. Self-focused attention is us ed to describe “. . an awareness of self-referent information” (Spurr & Stopa, 2002, p. 947). Shyness. Shyness is defined in this research the following way: discomfort and/or inhibition in interper sonal situations that interferes with pursuing one’s interpersonal and professional goals. It is a form of excessive self-focus, a preoccupation with one’s thoughts, feelings, and physical reactions. It may vary from mild social awkwardness to totally inhibiti ng social phobia. (Henderson & Zimbardo, 2001, p. 430) To avoid confusion regarding terminology, the reader should recall the continuum assumption defined above and bear in mi nd that one of three terms (shyness,

PAGE 29

18 social anxiety, or social phobia) is used throughout the study. When referencing research utilizing the term social phobia for example, the term social phobia is utilized. Otherwise, the term shyness is utilized, consistent with the continuum assumption that shyness, social anxiety, and social phobia exist on a continuum (Rapee & Heimberg, 1997). Shyness levels for this study. In this study, three levels of shyness (minimal to low, medium, and high) we re designated, based upon local norms provided in the present study as well as the studies listed in Chapter 3 for Research Question 6. The three levels of shyness (minimal to low, medium, and high) designated for Research Questions 8 and 9 were based upon norms reported in previous research. Small groups. The term small groups relevant to this study, means gatherings of three or more people to conduc t some activity related to the church, for example, to study t he Bible together or encour age one another in spiritual growth. These gatherings could be in the church building, a public place such as a restaurant, or a private home. Social anxiety. In this investigatio n, social anxiety is “ another aspect of shyness, being the apprehension provoked before a social situation when you want to make a good impression on a r eal or imagined audience but doubt that you can” (Carducci & Clark, 1999, p. 6). Social events In this study, social events were defined as gatherings, varying in size from small to large, in which church members and attenders interact with one another to observe or commemorate specia l occasions and/or

PAGE 30

19 to become better acquainted. Exampl es include weddings, bridal and baby showers, banquets, parties, and special m eals. These events could be held at the church building, in other public places or in private homes. For the purposes of this investigation, social events were distinguished from social situations in everyday life and in the church, as defined below. Social phobia. For the purposes of this res earch, social phobia is “a marked and persistent fear of one or more social or performance situations in which the person is exposed to unfamiliar people or to possible scrutiny by others” (American Psychiatric Associat ion, 1994, pp. 416-417). The individual with social phobia is afraid he or she will be embarrassed or humiliated by how he or she acts or that other people will perceive that he or she is anxious (American Psychiatric Association, 1994). Social situations in everyday life. Social events in everyday life are situations in which an individual is expec ted to speak with or, in other ways (e.g., smiling, nodding the head), in teract with other individu als. Examples include purchasing an item in a st ore, talking to friends and colleagues at work, or sharing an elevator. These are info rmal, unstructured situations and are distinguished from social events, which are more formal and involve interacting within a large group setting. Social situations in the church. Social situations in the church are similar to social situations in everyday life except that the situations occur in the church setting. Examples would be introduci ng oneself in a Sunday school class,

PAGE 31

20 making small talk before the worship service begins, or speaking to an acquaintance in the parking lot. Worship Services. Relevant to this study, worship services are occasions when the entire church congregation (memb ers, regular attenders, and visitors) gathers together to show reverence fo r God, to sing songs, and to hear a sermon. Limitations of the Study Limitations of this study include threats to the internal validity and external validity of the findings st emming from the quantitat ive phase of the study and threats to legitimation of the results stemming from the qualitative phase of the inquiry. Perhaps the most obvious thr eat was that of researcher bias in interpreting findings from the qualitativ e component of the st udy. This applied most particularly to the qualitative com ponents of the study as well as in the confirmatory factor analysis portion of the study, although to a lesser extent in the latter. As Onwuegbuzie (2003b) notes, a common form of researcher bias at both the data collection and t he data analysis stages of a st udy is the halo effect. This effect occurs when the resear cher has prior knowledge about the participants and allows that knowledge to influence the interpretation of findings (Onwuegbuzie, 2003b). Du ring the data analysis stag e, knowledge of the participant’s shyness level as measured by the BFNE and other characteristics could cause the researcher to perceive most or all participant responses as consistent with the known information.

PAGE 32

21 The researcher was conscientious in seeking to prevent personal bias unduly influencing the findings and interpre tations. A colleague in the field of education who had worked as a coder on tw o previous occasions was selected to act as a disinterested peer. This “disinterested peer” had no stake in the findings and interpretations and acted as “devil’s advocate” in order to keep the data interpretations as “honest” as po ssible (Lincoln & Guba, 1985, p. 308). Informal discussions of key concepts, c oding procedures and coding results were held during the training process and dur ing the coding and analysis portions of the study. Self-report measures ar e utilized under the assumption that participants can accurately reflect upon and report various aspects of their behaviors, characteristics, and so forth (Dobbs, Sl oan, & Karpinski, 2007). Regrettably, observation to confirm or disconfirm the self-report data was not feasible in the current study. However, score reliability and confirma tory factor analysis were utilized to assess the psychometric properties of the instruments. Another potential threat to the validity of the findings was that completing the BNFE scale might have sensitized participants’ answers to the open-ended questions. Completing the short-answer ques tions on the survey first might have represented a form of pretes t sensitization (Ary, Razavieh, Sorensen, Jacobs, & Sorensen, 2005), wherein data extracted from the second portion of the data collection instrument (i.e ., the extended response questions) were affected as a result of having completed the pretest or prescreening instrument (i.e., the BNFE scale). In particular, it was possible t hat some very shy individuals might have

PAGE 33

22 become anxious about supplying information about what is typically viewed as an undesirable personal problem and mi ght have declined participation. Alternatively, individuals who had low le vels of shyness might have declined participation if they believed the topic of the study to be unimportant for everyone by virtue of it being personally irrelevant for them. Either of these reasons for non-participation could have led to samp ling bias. To the greatest extent possible, the researcher attempted to minimize these threats during presentations to solicit participants. T he researcher provided reassurances that all data would be treated confidentially. The need for non-shy as well as shy individuals to participate in the study also was explained. Another way that samplin g bias might have influenced this study stems from the fact that the church selected fo r the study was an alr eady-formed group. As such, participants might have differed in important ways fr om participants from other churches or other settings in which perceived potential for negative evaluation might exist. This appeared to be unavoidable, due to study design. To the extent possible, this threat was handled by exerci sing an abundance of caution in drawing conclusions and in making generalizations. Furthermore, there seemed to be a high potential for bias in that participants were self-selected (i.e., volunt eered to participate in the study). This self-selection or “volunteer bias” (B ordens & Abbott, 2004, p. 122) was of particular relevance to the current st udy. In a comprehens ive study of the characteristics of volunteer participant s in research, Rosenthal and Rosnow (1975) reported that persons who volunt eer for research often possess the

PAGE 34

23 tendency to be more social than do nonvolun teers. It is presumed that the converse of that finding is true as well. If so, individuals who were less social, which would presumably incl ude those with moderate to hi gh levels of shyness, might have declined to participate. Accordingly, an attempt was made to remediate the potential effect s of volunteerism. During presentations to solicit participation, the researcher stressed t he critical need for participants with social anxiety. The researcher also stressed t he confidentiality with which results would be handled, as discussed previously. As stated earlier, research has indicat ed that social anxiety might affect many areas of interpersonal functi oning (Henderson & Zimbardo, 2001). Of particular relevance for the current study is that socially anxi ous individuals might have difficulty in maintaining conversati ons, they might be inhibited, and they might have speech dysfluencies or other nervous behaviors (Zimbardo et al., 1974). Each of those characteristics w ould likely have occurred numerous times in the life of a socially anxious adul t. Participants who have even a moderate degree of social anxiety might have t hat anxiety, as manifested in such characteristics, accentuated when participat ing in a study wherein the main topic is a psychological attribute about which t hey may feel embarrassed. As stated earlier, every effort was made to provi de reassurances about confidentiality. In this study, there also seemed to be a considerable potential for making misspecification errors—that is, omitti ng one or more important variables. Research has indicated shyness is a multi-faceted phenomenon (Crozier, 2001). Consequently, careful attenti on was paid to this threat to internal validity in that

PAGE 35

24 the researcher conducted a thorough revi ew of the literat ure and engaged in conversations with committee member s and peers throughout the study. Additionally, one church congregation was selected from which to draw a sample, which poses a threat to ecological validity. Accordingly, the findings of the current study might not be generalizable to churches of other denominations or to churches with widely varying char acteristics such as size, location, educational level of church attenders, and so forth. The pot ential inability to generalize findings beyond the church setti ng to the everyday world was an even larger threat to ecological validity but unavoidable due to study design. As is the case with ecological validi ty, the study design created a threat to population validity in t hat a subgroup of the initial samp le was selected. From all of the participants who completed the wr ite-in questions, 15 participants whose responses were sufficiently detailed were selected. Onwuegbuzie (2003b) specifies that “…. any kind of sub-samp ling from the data set likely decreases population validity” (p 84). Again, this threat could not be overcome with the study as designed. The final potential threat to validity was confirmation bias. Stated simply, confirmation bias is the tendency for a res earcher to find what he or she wants to find and to ignore, or misinterpret, anything else (Nickerson, 1998). As Greenwald, Pratkanis, Lei ppe, and Baumgardner (1986) noted, this type of potential bias is most often present when the aim of a study is to test, rather than create, theory. Thus, careful attent ion, through utilization of the methods

PAGE 36

25 delineated above in the discussion of res earcher bias, was paid to minimizing this threat. According to Patton (1990), triangulat ion is “. . the combination of methodologies in the study of the same phenomena or programs” (p. 187). Employing more than one met hod of investigation allows the researcher to see different aspects of the same phenomenon (Denzin, 1978). The quantitative and qualitative components of the study constituted met hodological triangulation. Data triangulation was utilized in that t he BFNE provided quantitative data, and responses to the write-in questi ons provided qualitative data. Peer debriefing was one of the most important methods employed in the current study to address threats to the legitimation of the findings. In peer debriefing, a peer questions the researcher for the purpose of “. . probing biases and clarifying interpretations” (Lincoln & Guba, 1985, p. 308). Throughout the study, the researcher discussed plans findings, and conclusions with the committee as well as with other colleagues. Another key method employed to seek maximal legitimation was negative case analysis. As Patton (1990) explai ns, “Where patterns and trends have been identified, our understanding . is incr eased by considering the instances and cases that do not fit within the pattern” (p. 463). Particular attention was given to write-in responses that seemed to indica te shyness operates differently in the church setting than in the everyday world. Throughout the study, the re searcher kept a reflexive journal, as described by Lincoln and Guba (1985). The researcher utilized journal entries to record

PAGE 37

26 relevant information about herself as well as about all methodological decisions made. Summaries of critical peer debr iefings are included in the reflexive journal. Delimitations of the Study The primary delimitati on of the current stud y concerns the age of participants. To control for developmental issues, only individuals equal to or older than 21 years of age were selected for participation. Organization of Remaining Chapters Chapter 2 is a review of research re levant to this study. Chapter 2 begins with an overview of the chapter and is fo llowed by the background for the study, which is organized under the following sub headings: key issues, measurement of shyness, a naturally occurring experim ental setting, and the social climate of the church. Chapter 2 c oncludes with the theoretical base and rationale of the study, as well as a brief summary. C hapter 3 presents the methods utilized for the study, beginning with parti cipants and ethical considerations. Next, detail is provided regarding instruments and procedu res that were utilized. Finally, analysis plans for qualitative and quantitat ive data are presented. Chapter 4 opens with a description of participants, wh ich is followed by the results of the study, presented in order of the nine res earch questions. Chapter 5 contains a summary and discussion of the findings as well as implications for future research.

PAGE 38

27 CHAPTER 2 REVIEW OF THE RELATED LITERATURE Overview Shyness affects between 40% and 50% of individuals, to varying degrees and in various situations, at some point in life (Carducci & Clark, 1999; Zimbardo et al., 1974). A substantial body of res earch on shyness has been constructed in the last several decades (Crozier & Alden, 2001a, 2001b; Weeks et al., 2005), and much research on the development of measures also has been conducted (Cheek & Briggs, 1990; Orsillo, 2001). Relev ant for the current study is the fact that no studies could be found on the topic of how shy indi viduals, feel, think, and behave in environments with less perceived potential for negative evaluation and less attention to self. Fear of negative ev aluation and attention to the self are two key elements in maintaining shy behavior (Clark, 2001). This review of literature presents the themes that are most commonly addressed in shyness research as well as a discussion of two of the most dominant cognitive models of shyness. This chapter also describes typical behaviors of shy people and environmental i ssues that research has suggested contribute to maintenance of shy behaviors. Additionally, the significant role that fear of negative evaluation and self-attenti on play in maintaining shy behaviors is addressed, as is the social climate of the church—the setting for the current study. Chapter II closes with a summary of findings pointing to the potential utility of exploring shyness in the church setting.

PAGE 39

28 Background Prevalence of Shyness In the landmark 1979 study cited most often in discussing the prevalence of shyness, Zimbardo and colleagues at St anford University (e.g., Pilkonis & Zimbardo, 1979) found that more than 40% of a sample of 470 high school and college students considered t hemselves shy and that 63% regarded shyness as a problem. A study by Carducci and Zi mbardo (1997) indicates the prevalence rate for North American adults has increased over the years to greater than 50%. One finding in the Stanford su rvey that is particularly in teresting is that a high percentage (73%) of responde nts reported they were either shy now or had been at some time in the past. Crozier ( 2001), in reviewing research using Zimbardo et al.’s (1974) survey, repor ted a median value of 84% for being currently shy or shy in the past, with little cross-cultural variation. Although Crozier’s (2001b) caveat that the increasi ng public awareness might be affecting response rates merits serious consideration, so does hi s conclusion that “a substantial number of people report that they are shy [o r have been] and that their shyness is undesirable and causes a problem for them“ (p. 3). Recurrent Themes in the Literature Beginning in 1986, several volumes, such as Understanding shyness: Psychological Perspectives (Crozier, 2001), Shyness: Perspectives on research and treatment, (Jones, et al. 1986), and Soci al phobia: Clinical and research perspectives (Stein, 1995) have reported on research trends. Other volumes

PAGE 40

29 include those by Crozier and Alden, 2001; Heimberg et al., 1995; and Leary and Kowalski, 1995. Most of those volume s include discussions of shyness as a personality trait versus a temperamen t, as well as discussions on genetic, environmental, and developmental issues, and gender differences. Research on this topic is presented in the following sections. Traits and temperaments. Some of the most influent ial efforts to identify fundamental personality traits have involved factor analysis, with three of these approaches finding “at least two fundam ental higher order dimensions: extraversion and introversion and neuroticism (or anxiety)” (Crozi er, 2001, p. 24). Introverts generally prefer and tend to be alone, act shy, and tend to withdraw during times of stress; extraverts ar e the opposite (Carver & Scheir, 1996). Neuroticism (also called emotionality) re fers to the tendency to become upset and/or distressed relatively easily and often and the tendency to be moody, anxious, and depressed (Carver & Scheir, 1996). Evidence as to whether social anxie ty is similar to introversion is inconclusive. For example, Eysen ck (1956), Crozier (1979), and Cheek and Buss (1981) concluded that t he two constructs are differ ent. In contrast, Bruch (1989) found a negative correlation of -.56 between shyness and introversion compared to a -.28 correlation repo rted by Cheek and Buss (1981). Factor analytic studies have found evi dence of a shyness factor (Crozier, 2001) but research in this area is hindered because shyness is a term taken from everyday language and there is no br oadly accepted definition of shyness

PAGE 41

30 (Crozier, 2001). Another factor making communica tion of findings among researchers difficult is that the shy population is extremel y heterogeneous. Temperament is a term used by psychol ogists to explore the effects of inheritance on personality. These effect s are observable in infants and very young children. According to Buss and Pl omin (1984), temperament differs from other personality traits in that temperament has a basis in biology, has a deeper and broader influence than do other traits, an d is stable over time, though it is subject to modification by experience. Buss and Plomin also view temperament as lying on a continuum. Kagan and Re znick (1986), however, conceptualize temperament as being categorical in natur e. These authors found evidence that “. . perhaps 15% of the normal population are born with either a very high or a very low threshold for physiological arousal and an accompanying state of uncertainty following an encounter with the unfamiliar” (p. 88). Behavioral inhibition, however, is only one of many vu lnerabilities that lead to childhood and adult shyness. State-Trait anxiety. It is customary to dist inguish between state and trait anxiety, and Spielberger’s work on trait anxiety is accepted as the standard (Reiss, 1997). The instrument used most often to measure anxiety is the StateTrait anxiety Inventory (Spielberger, C. D., Gorsuch, R.l., & Lushene, R.E., 1970). According to Spielberger et al. (1970), state anxiety is a temporary emotional state where an individual experiences tension and fear along with increased activity in the autonomonic nervous syst em (such as increased heart rate or sweating). Trait anxiety, however, is more permanent in nature in that it is a

PAGE 42

31 relatively stable tendency to feel anxious in situations where the individual feels threatened. It would be expected that indi viduals with a high level of trait anxiety would tend to be anxious in many, or mo st situations, whereas individuals with minimal to low levels of trait anxiety might be anxious mainly in situations where most people were anxious, such as giving a speech or being interviewed for a job. Environmental, developmental, and genetic influences. Table 3 depicts the conditions, or vulnerabilities, t hat Bruch and Cheek (1995) believe can eventually result in shyness or social phobi a. This conceptualization clarifies the interactive role of genetics, environm ent, and development, a finding nearly always emphasized in volumes (Crozi er, 1990, 2000a, 2000b, 2001; Crozier & Alden, 2001; Heimberg et al., 1995; Jones et al., 1986; Leary & Kowalski, 1995) and articles (Keller, Wood, McLeod, Sigm an, Wei-Chin, & Chu, 2003; Ollendick & Hirshfeld-Becker, 2002) addressing developmental issues. Specifically, consensus is that the cause of shyness and social phobia is part nature and part nurture. As Bruch and Cheek (1995) state, an individu al is born with characteristics, like behavioral inhibition, that interact with t he environment. In infancy and early childhood, the child is in fluenced most by the family, which may or may not exacerbate cert ain tendencies. By middle or late childhood, relationships with parents, peers, and the self add to the sources of possible vulnerabilities. Finally, the passage through adolescence seems to create or intensify preoccupation wit h the possibility of negat ive evaluation, and the individual enters young adulthood shy or so cially phobic. Studies involving twins

PAGE 43

32 have led some researchers (e.g., Boom sa & Plomin, 1986) to conclude that shyness has a larger component of herit ability than does any other personality trait. After summarizing si x studies that compared sco res for monozygotic and dizygotic twins on measures of inhibiti on and shyness, Crozier (2001, p. 112) supports that finding by concluding that “measures of shyness show a substantial genetic component.”

PAGE 44

33 Table 3 Common Vulnerabilities Underlying Shyness and Social Phobia Birth Early Childhood Middle and late childhood Immediate results Possible longterm consequence Inherited temperament: >Wariness >Emotionality >Behavioral inhibition Family Context: >Overcompen sation via ineffective parenting (overprotection withholding affection) >Family members not allowed to express emotion (1) Inappropriate parental child rearing attitudes (Parent & societal emphasis on traditional sex roles (2) Negative peer relations (rejection and victimization, especially if have a high need for affiliation) (3) Life experiences that disturb social facets of self-esteem (e.g., peer rejection) Conflict for shy child Preoccupation with possibility of negative evaluation Misinterpretati on of one’s physical and social acceptability Maladaptive coping styles (e.g., avoid selfdisclosure) Prone to develop shyness and social phobia symptoms Note: Adapted from Bruch, M. A, & Cheek, J. M. (199 5). Developmental factors in childhood and adolescent shyness. In R. G. Heimberg, M. R. Liebowits, D. A. Hope, & F. R. Schneier (1995) Social phobia: Diagnosis, assessment, and treatment. New York, London: The Guilford Press.

PAGE 45

34 Gender differences. Research interest regarding gender differences in shyness seems to have developed somewh at later than did research interest regarding shyness in general. In 1974, Zimbardo et al. found no gender differences in the prevalence of sh yness. That finding perhaps tended to suppress investigations concerning gender differences initially. Nonetheless, gender differences have been the focus in more relatively recent research regarding shyness (Bruch, Gorsky, Collins, & Berger, 1989; Crozier, 1990, 2001; Crozier & Alden, 2001b; Deardorff, Haywar d, Wilson, Bryson, Hamme, & Agras, 2007; Pilkonis, 1977; Pollard, & Henderso n, 1988; Rapee, 1995). The findings are often confounded with other va riables, as described below. In general, gender differences in shy ness reported in t he literature seem to coincide with what one would expec t, given a basic understanding of the variables being considered and of how shyne ss manifests itself. Deardorff et al. (2007) found no gender differences among pr epubertal youth, but pubertal girls reported more symptoms of social anx iety than did pubertal boys. This corresponds to common knowledge that adolescent girls, even more so than adolescent boys, are intensely conc erned about appearances and popularity. Gender differences in shyness also seem to be situation-specific so far as some behaviors are concerned. For ex ample, Pilkonis (1977) found more differences between shy and nonshy men than between shy and nonshy women regarding speech and eye contact. Shy m en were more hesitant to speak, and they spoke less. Shy men also engaged in less, and briefer, eye contact.

PAGE 46

35 However, Bruch et al. (1989) found no gender differences in the amount of talk in which shy men and women engaged. Another way in which gender diffe rences in shyness have been investigated perhaps points to a fundam ental difference t hat has confounded results in many studies. It also mi ght explain why results have so often been mixed or apparently contradi ctory. In one study (Rapee, 1995), it was found that individuals with social phobia presented to clinics in an approximately equal distribution regarding gender, with slightly more males than females. Rapee (1995, p. 55) believed that the finding mi ght reflect “presentation differences rather than actual diagnostic differences in that females report more anxiety disorders, including social anxiety, but males are more likely to seek treatment.” One plausible explanation for that fi nding seems to be, as Rapee (1995) observes, the influence of society—Western society in particular. In Western society, men are generally expected to in itiate romantic encounters, to be more successful in their careers, and in general to be more assertive than are women. This would logically suggest that social difficulties would cause greater problems for men than for women. Rapee based that argument, in part, on the findings by Pollard and Henderson (1988) indicating th at twice as many females as males met criteria for social phobia as specifi ed in the DSM-III. In the Pollard and Henderson study, when the criterion of ‘si gnificant distress’ was included, gender proportions became more similar. These findings possibly suggest that women tend to report more sympto ms of severe shyness than do men but that severe shyness is as troubling for men although they tend to under-report it. This

PAGE 47

36 possibility was considered during the parti cipant selection and analysis phases of the current study. Although the above argument was m ade in 1995, the social norms Rapee referenced are presumed to co ntinue exerting a powerful influence on contemporary behavior. Cultural differences. The previously mentioned finding of a 40% prevalence rate has been replicated in num erous countries, with the prevalence rate ranging between 24% for a sample of Jewish Americans to 60% for respondents in Hawaii and Japan (Pines & Zimbardo, 1978). Contemporary research has focused in particular on cultural differences in shyness for Asian populations. Possible reasons for this ar e set forth by Hsu and Alden (2007): (a) the increase in Asian immigration to North America; (b) the research finding that Chinese societies do not look on social anxiety as negatively as do other societies; and (c) the pres umption that shy behaviors are less likely to disrupt social harmony, which is more valued in Asian than in Western cultures. Furthermore, shy behaviors might actually be considered desirable in Asian societies (Hsu & Alden, 2007). Based on the foregoing, it seems logical that cultural differences in Asian populations would be an area of contemporar y research interest. A search of recent research literature did, in fact, re veal a substantial number of citations, as listed above. Trends in Research Concerning Shyness Briggs, Cheek, and Jones (1986) describe shyness research as comprising three phases, the first of wh ich is a descriptive phase based on

PAGE 48

37 clinical observations from medical and p sychological practitioners, beginning as early as 1896. They believe that as our world became increasing mobile, people were required to develop new friendships and to make new social connections beyond family and friends. In the sec ond phase, the mid and late 1970s, which Briggs et al. (1986) term the popularization of shyness, several books about shyness and how to overcome it we re written for the general public. Shyness: What it is, what to do about it ( Zimbardo 1977) was the most widely read of these books. In addition to case histor ies and interviews, social psychologist Zimbardo used data to add emphasis to his findings that shyness is a serious personal problem, that it has reached epidem ic proportions, that it is caused by living in a competitive societ y, and that it can be allevi ated. Zimbardo found that 42% of U.S. college students rated themse lves as shy and that figure rose to 73% when students were asked about past as well as current shyness. The third phase of research, accordi ng to Briggs et al. (1986), began in the early 1980s, with more traditional em pirical investigations, which have resulted in a clearer conceptualization of shyness and its relationship to other theories and models. In this phase of exploring correlates and consequences of shyness, many new scales were develo ped. Also, durin g the third phase, besides research focusing on shyness, related work on introversion, assertiveness, shame, and embarrassment has increased our understanding of how shyness affects the lives of indivi duals. For example, Crozier viewed empirical research on shyness and embarrassment up to 1990 as fitting a

PAGE 49

38 framework of either social anxiety or “the underlying em otion of shame” (Crozier, 1990, p. 53). Regarding research from the mid 1980s onwards, the focus of research interest in shyness apparently experienc ed an interdisciplinary migration from one of the two broad categories of researcher s investigating shyness (i.e., social scientists) to the other (i.e., mental health professionals). McDaniel (2003) attributes this upsurge of research intere st to two causes: the “medicalization” of shyness, which began in 1980 when the DSM-I II included “social phobia” in its diagnostic categories, and the use of drug therapies in treating shyness, beginning in the 1990s with Prozac and with Paxil. Regarding social phobia per se, social phobia had been described as early as 1970, but it was not added to the DSM until 1980. Even so, the 1980 definition was limited and remained so, wit h only 2% to 3% of the population identified as having the di sorder. By the early 1990s, however, the percentage was in the double digits. Most believed this shift was due to a 1985 article by psychiatrist Michael Liebowitz entitled “S ocial Phobia: The Neglected Anxiety Disorder.” That article sti rred interest in the medica l and research community alike. By 1987, the DMS III-Revised incl uded a general subtype of social phobia, which many researchers see as extremely close to shyness. More importantly, the DMS III-R defin ition excluded the cr iterion “compelling desire to avoid.” Before that point, even if an individual had marked distress, he or she would not be classified as social phobic, so long as he was able to endure social situations. To obtain reimbursement fo r treatment from insuranc e companies, an individual

PAGE 50

39 has to meet the definitional criteria for social phobia. With these two shifts in diagnostic criteria, in 1994 the Archives of General Psychiatry estimated a lifetime prevalence rate of social phobia in the U.S. as 13% (Cottle, 1999). Measurement of Shyness The measurement of shyness is anot her area in which the effect of discipline-specific definitions, and even more so, discipline-specific methodologies, is readily apparent. Most of the earliest questionnaires and scales, beginning with the often-cited Stanford Shyness Survey (Zimbardo, 1977), were developed and used by psychologists and social psychologists. It is interesting to note that the Stanfor d Shyness Survey, although it has so influenced research, is not actually a scale and has not been used as such. Rather, it consists of 44 questions regar ding various aspects of shyness, and the responses are not intended to be summed to derive a score (Briggs & Smith, 1986). Five of the most commonly used meas ures of shyness were reported by Briggs and Smith (1986) and we re included in a review by Crozier (2001). Four of these scales are displayed in Table 4; the fifth scale used in the study was the Morris Shyness Scale (Morris, 1982). Br iggs and Smith (1986) administered the five shyness scales to a sample of 1,213 college students from five institutions and obtained score alpha coefficients rangi ng from .82 to .92 and inter-item correlation means between .25 and .36. Besides the exceptional internal consistency, they also found that t he scales seemed to measure the same construct, even though each differed in co nceptual focus. Convergent validity

PAGE 51

40 was evidenced in that pair wise correlati ons of the five scales with each other ranged from .70 to .86. Construct-rela ted validity of these and other scales has been assessed in several studies (Cheek & Briggs, 1990). Table 4 Hits for Shyness/Social Phobia Meas ures as Recorded in PsychINFO Name of Scale Date Written Number of Items in PsychINFO *SADS (Social Avoidance and Distress) 1969 146 FNE (Fear of Negative Evaluation) 1969 83 *Social Reticence Scale 1984 13 *Shyness Scale (Cheek-Buss) 1981 57 *Interaction Anxiety 1983 27 Social Phobia and Anxiety Inventory 1989 95 Social Phobia Inventory 2000 43 Utilized by Briggs and Smith (1986). Use of these scales for shyness and social anxiety, as well as three of the most popular measures for social phobia, has been extensive, as measured by the number of researchers using them in studies (see Table 4), as reported in the PsychInfo database. If one compares the age of these scales to the number of studies using them, it is easy to conjectu re that scales developed specifically for measuring social phobia, though relatively new, are being used at a faster rate. This parallels, of course, the increasi ng interest devoted to social phobia by psychiatry subsequent to its re classification in the DSM.

PAGE 52

41 Theoretical Framework C ognitive models. The literature contains a number of models for shyness, social anxiety, and social phobia, and volumes addressing shyness typically contain at least one, usually more, chapters devoted to theoretical foundations and/or specific models (C rozier, 1990, 2000a, 2000b, 2001; Crozier & Alden, 2001b; Jones et al., 1986; Lear y & Kowalski, 1995; Stein, 1995). The two models to be utilized in the current study are described below. Rapee and Heimberg Model Rapee and Heimberg (1997) acknow ledge that their model, which is particularly helpful in conceptualizing t he mental processes of the socially anxious or phobic person, builds on earlier versions. The assumptions of the model were addressed in Chapter I. According to Rapee and Heimberg, social situations activate a series of processes that create and maintain social anxiety. When a social phobic or socially anxious person enters a social situation or merely thinks about it, the individual forms a mental representation of how he or she thinks others perceive him or her and that mental re presentation then becomes the focus of attention. The mental representati on is created from various sources of information, includi ng long-term memory (recalling one's perceived poor social performance in a similar situation), proprioceptive information (e.g., perceived heart rate increase, flushing, or blushing), and external cues (e.g., facial expressions of others). The individual's attention is then focused on what is perceived as the mo st relevant aspects of the situation, and those are generally the most negative because the individual's fear of

PAGE 53

42 negative evaluation leads him or her to focus on the worst that can happen—for example, that he/she might be unable to sa y anything at all or that others might observe that his/her hands are shaking. Possible signs of negative evaluation by others, such as a lack of interest, are also an object of attenti on as the individual monitors all potential threats. Subsequent ly, a comparison is made between the mental representation of the self as seen by other s, and, most typically, unrealistic standards of performance. The discrepancy between the actual self and the self one believes one should be cr eates a perception of negative evaluation, which causes anxiety and all its physical, mental, and behavioral manifestations. That anxiety leads the i ndividual to perceive that others evaluate him or her negatively and the cycle renew s itself (Rapee & Heimberg, 1997). Clark and Wells Cognitive Model The Clark and Wells (1995) cognitive model of social phobia emphasizes the role of the shift in attention that occurs when a social phobic client perceives, whether accurately or inaccurately that he/she is about to be evaluated negatively. Attention is focused away fr om the environment to an inward self monitoring. That self-focus causes a heightened awareness of the anxiety responses the individual fears (such as blushing or stamme ring). It also interferes with processing information about the situation, including the behavior of other people. The individual might not hear a question or might not see a smile aimed in his or her direction due to paying attention to his or her own physical and mental reactions. Besides that, social phobic clients tend to use this interoceptive information to form a negative impression of themselves, as

PAGE 54

43 seen by others. To avoid these painful consequences of social interactions, individuals with social phobia tend to us e a variety of safety behaviors, like avoiding social situations and if that is not possible, avoiding eye contact and prolonged conversational interchanges and spending a great deal of effort rehearsing what to say rather than paying attention to and responding appropriately to what is being said. Of course, such safety behaviors make it impossible for the socially phobic indivi dual to obtain evidence that disconfirms dysfunctional beliefs about the self. Furt her, these behaviors ma ke it likely that some of their fears about t heir social performances, such as stuttering, blushing, or being unable to speak, will occur. A per son who seldom smiles, who does not appear to be listening, and who seldom joins in conversations is likely to be perceived as somewhat unfriendly and wil l eventually be approached by others less often (Clark & Wells, 1995). Similarities and differences between the models. Table 5, which integrates diagrams of the Clark and Wells and the Rapee and Heimberg models, reveals their similarities, and c onfirms the view set forth by Musa and Lpine (2000) that the Clark and Wells model stresses "self focus and safety behaviors," whereas the Rapee and Heimbe rg model stresses the "discrepancy between mental representat ion of self and others' expected standards" (Rapee & Heimberg, 1997, p. 62). This is logica l if one recalls that the Clark and Wells model was developed as a model for treatm ent and to explain how social phobia is maintained, whereas the Rapee and Heimberg model was developed for heuristic purposes.

PAGE 55

44 Table 5 Comparison of Clark and Wells and Rapee and Heimberg Models Clark and Wells Model Relationship Between Models Rapee and Heimberg Model Social situation Which involves the Perceived audience Activates assumptions Which causes One of which is Preferential allocation of attentional resources Mental representation of self as seen by audience Perceived social danger The intensity of which is judged by: External indicators of negative evaluation Perceived internal cues Processing of self as a social object Which involves Comparison of mental representation of self as seen by audience with appraisal of audience’s expected standard Judgment of probability and consequence of negative evaluation from audience Safety Behaviors Behavioral symptoms of anxiety Somatic and cognitive symptoms of anxiety Cognitive and physical symptoms of anxiety Note: Adapted from Clark, D. M. (2001) A cognitive perspective on social phobia. In W. R. Crozier & L. E. Alden (Eds.) International handbook of social anxiety: Concepts, research, and interventions relating to the self and shyness (pp. 404430) New York: Wiley; and Rapee, R.M., & Heimberg, R.G. (1997). A cognitive behavioral model of anxiety in social phobia. Behavior Research Therapy 35 741-756.

PAGE 56

45 The current study employed the Clar k and Wells (1995) model as part of the data gathering and analyt ic framework because of its focus on the thoughts, feelings, behavior, and attentional focus in social situations (Clark, 2001). Specifically developed for the treatment of social phobia, the Clark and Wells model focuses on thoughts, feelings, behavio r, and attentional focus prior to, during, and after involvement in a so cial situation (Clark, 2001). The Rapee model was used primarily to provide explanatory insights. A detailed presentation of the Clark and Wells (1995) model follows. Clark and Wells Cognitive Treatment Model in Depth The aim of the treatment progr am developed by Clark, Wells, and colleagues is to reverse the processes that maintain social phobia, as specified in the model (Clark, 2001). The program se eks to modify self-focused attention, negative self-processing, and safety behaviors Doing so gives individuals with social phobia the opportunity to disconf irm their negative beliefs. The steps in the treatment program are summarized below. Deriving an idiosyncratic version of the model. Therapy begins by reviewing a typical incident of social anxi ety and fitting the particular details of the individual's experience into a personal ized diagram of the model. Figure 1 presents the simplified template of the model that is used and a hypothetical example of anxiety that a socially phobic church attender or member might experience while waiting for a wo rship service to start. Manipulation of self -focused attention and safety behaviors. After agreement is reached on the personalized model, the therapist engages the

PAGE 57

46 client in role-playing by changing, for exam ple, the client’s fo cus of attention or dropping some of the client’s safety behavio rs. Self-ratings after each role-play typically demonstrate to c lients that self-focus and sa fety behaviors tend to make them feel more anxious and that how they think they looked and performed is related to how they felt, which enables t hem to see they are using their feelings to interpret how others perceive them, rather than reality. Video and audio feedback. The purpose of this step is to help the client obtain realistic information about how they appear to others during social encounters. By viewing a video of themselv es interacting with others, clients can see, for example, that their hands did not shake visibly. This helps the client begin to alter previous beliefs. Shift of attention and interrogati on of the social environment. Next, clients are encouraged to shift their attention externally and eliminate safety behaviors during therapy session and in homework assignments. Clients are helped to engage the social events they have fear ed and avoided previously; however, the goal is not just simply to have the c lient undergo repeated ex posures to feared situations. Rather, the client is enc ouraged to predict likely outcomes and then evaluate whether those outcomes actually occurred. For example, an individual who fears engaging store clerks in conver sation might predict that the clerk would frown and make a comment such as "What a stupid thing to say!" When the client evaluates the actual situation, he or she can see that the feared event did not occur. Often, qui te the opposite happens. For example, the store clerk might smile warmly and initiate a pleasant conversation.

PAGE 58

47 Dealing with anticipatory and post-event processing. As described earlier, one of the prime characterist ics of social phobic behavior is the negative mental processing before and after social ev ents. Clients are encouraged to become aware of their habitual ways of thinking and to modify them—for example, not rehearsing a "script" before each casual encounter and afterwards focusing deliberately on what went right, rather than on what was perceived as having gone wrong. Dealing with assumptions. According to Clark (2001), three types of assumptions affect how individuals with social phobia mentally process information about social encounters: "excessively high standards for social performance. . conditional beliefs concer ning . consequences . ('if I am quiet, people will think I am boring'). . and unconditional negative beliefs about the self, e.g., 'I'm odd/different' (p 407). These are handled by "bandwidth" exercises. “Widening the bandwidth” is a term us ed by Clark and Wells (1995). To deal with the unrealistically high standar ds for social behavior that many individuals with social phobia have, the client needs to broaden his or her usual range of behaviors and needs to act in ways t hat violate self-imposed rules. For example, instead of trying to think th rough several comments before speaking, the client is encouraged to say the first thing that pops into his or her mind. Seeing that this does not cause a social ca lamity helps the client gain confidence to experiment and try other behaviors, that is, to “widen the bandwidth” (Clark & Wells, 1995, p. 424). These exercises often help clients change some of their

PAGE 59

48 self-defeating and unquestioned a ssumptions, too, such as “No one likes me” or “I can never say anything to a stranger.” However, Wells (2000) st ates that negative beliefs about oneself are often persistent because they are vague and poorl y defined. The treatment program addresses this by having the client operat ionalize negative self beliefs. Clients would list all the observable characteri stics that would support that belief and then rate themselves and others on the listed characteristics. For example, if a client believed “I am socially inept becaus e I never talk to st rangers,” the client might count how often he/she talks to st rangers in elevators, stores, and parking lots. That procedure often helps client s see that they do not possess all the characteristics they think they do and that they are generally about the same as many people. The Clark and Wells (1995) model wa s utilized to analyze thoughts, feelings, and behaviors in the study setting. Regardless of th e type of setting, however, shy individuals in general tend to exhibit certain charac teristics. These characteristics are described below. Typical Behaviors of Shy People Table 6 reveals some of the typica l thoughts, feelings, and behaviors that shy individuals experience (Henders on & Zimbardo, 2001). These thoughts, feelings, and behaviors, or symptoms frequently appear as variables in research, as the titles of the follo wing measures demonstr ate: the Social Avoidance and Distress Scale (Watson & Friend, 1969); the Fear of Negative Evaluation Scale (Watson & Friend, 1969) ; and the Interaction Anxiousness

PAGE 60

49 Table 6 Symptoms of shyness (H enderson & Zimbardo, 2001) Behavior Physiological Inhibition and passivity Gaze aversion Avoidance of f eared situations Low speaking voice Little body movement or expression or excessive nodding or smiling Speech dysfluencies Nervous behaviors, such as touching one’s hair or face Accelerated heart rate Dry mouth Trembling or shaking Sweating Feeling faint or dizzy, butterflies in the stomach or nausea Experiencing the situation or oneself as unreal or removed Fear of losing control, going crazy, or having a heart attack Cognitive Affective Negative thoughts about the self, the situation, and others Fear of negative evaluation and looking foolish to others Worry and rumination, perfectionism Self-blaming attributions, particularly after social interactions Negative beliefs about the self (weak) and others (powerful), often out of awareness Negative biases in the selfconcept e.g., “I am socially inadequate, unlovable, Embarrassment and painful selfconsciousness Shame Low self-esteem Dejection and sadness Loneliness Depression Anxiety

PAGE 61

50 unattractive” A belief that there is a “correct” protocol that the shy person must guess, rather than mutual definitions of social situations Scale (Leary, 1983). These symptoms also appear in self-diagnostic check-lists in the self-help literature, for example, in Shyness: What it is, what to do about it (Zimbardo, 1977). These characteristics of shy individuals are experienced more often in particular kinds of settings, as described below. Key Environmental Influences As stated earlier, one purpose of this st udy was to understand the effect of different environments on the manifestati ons of shyness. The environmental influences on shyness that were relevant for this study are described below. Nine situations most difficult for shy people. How shyness is manifested, in typical as well as atypical environm ents, is a recurrent theme in shyness research (Crozier, 1990, 2000a, 2000b, 2001; Crozier & Alden, 2001b; Jones et al., 1986; Leary & Kowalski, 1995; Weeks et al., 2005). In their exploratory work on shyness, Zimbardo et al. (1974) us ed the Stanford Shyness Survey to investigate the effects of di fferent situations on shy behav iors. One section of the survey asked respondents to indicate t he situations that most often elicited shyness. A replication and extension of that study conducted 20 years later (Carducci & Clark, 1999) produced similar results. For the current study, similar categories in both studies were combined in to one list, as presented in Table 7.

PAGE 62

51 Table 7 Ten Most Difficult Situations for Shy People 1. Being the focus of attention 2. Large groups 3. Small groups 4. Authority figures by virtue of knowledge (intellectual superiors, experts) or by virtue of role (po lice, teachers, s uperiors at work) 5. Social situations in general 6. New interpersonal si tuations in general 7. Strangers 8. Situations where assertiv eness is required (e.g., when complaining about faulty service in a restaurant) 9. Being evaluated or compar ed with others (e.g., when being interviewed, when being criticized) 10. An opposite sex group or a member of the opposite sex Note : Adapted from Crozier, W. R. (2001). Understanding shyness : P sychological perspectives. China: Palgrave. Similarity to six activities of church social life. These 10 situations are highly similar to the 6 social situations in church life c hosen for the current study: worship services small group gatherings (e.g., S unday School and Bible study classes) social events such as weddings, baby showers, and holiday celebrations contacts with friends and acquaintances, interaction with authority figures (e.g ., pastor, staff, group leaders) and jobs (e.g., Sunday School teacher, greeter).

PAGE 63

52 These 6 activities in which c hurch members and regular attenders routinely engage include aspects that closel y parallel the 10 situations. The cells marked with an asterisk (*) in Table 8 indi cate aspects of the six church life activities to which it seems research fi ndings could be cautiously extrapolated. Table 8 Areas of Commonality Between the Nine Si tuations and Five Church Life Activities Six church life activities Ten situations Worship Services Small groups Social events Friends, etc. Interactions with Authority Places of Service Being the focus of attention * * Large and small groups * Authority figures Social situations in general * * * New interpersonal situations in general * * * Strangers * Situations where assertiveness is required * *

PAGE 64

53 Being evaluated or compared with others * * An opposite sex group or a member of the opposite sex * * * Two key environmental influences that maintain shyness. Two key environmental influences relevant for t he current study ar e fear of negative evaluation and self focus S hy individuals report feeling shy when they fear being negatively evaluated by others and w hen they are the focus of attention. Both fear of negative evaluation as well as self-focus were cent ral in the work of several theorists researching shyne ss and social phobia (Beck, Emery, & Greenberg, 1985, 1996; Hartman, 1983; Heim berg & Barlow, 1988; Leary, 1983; Trower & Gilbert, 1989). In the Clark and Wells (1995) model, two key components that maintain social phobia ar e focusing of attention upon self and negative thoughts about the se lf. These two components of the model are addressed in detail below. Fear of negative evaluation. The effect of feared criticism has long been prominent in shyness research, as evi denced by the early and extensive use of the Fear of Negative Evaluation scale (Watson & Friend, 1969) and its subsequent revisions (Collins et al., 2005) As mentioned earlier, the FNE and BFNE are included in reviews of measures of shyness and social phobia (Antony, Orsillo, & Roemer, 2001; Orsillo, 2001). Both sc ales assess differences

PAGE 65

54 in “broad social-evaluative anxiety, such as public speaking and going on a date” (Collins et al., 2005, p. 347). The fear of being evaluated unfavorably is one of the main features that can maintain social phobia for decades ( Clark & Wells, 1995). According to the continuum assumptions described previ ously, that fear also can maintain moderate to extreme shyness for prol onged periods of time. Because of previous social difficulties, individuals with social phobi a tend to believe habitually that when they enter certain social situations, they will behave in an unacceptable manner (e.g., blush or stutte r). Individuals with social phobia also tend to believe that such behaviors will have terrible consequences, such as rejection by others and embarrassment ( Clark & Wells, 1995). Several vicious cycles are thus set in motion because the symptoms of this fear maintain and even increase the anxiety. For exampl e, stuttering can make an individual believe he or she looks foolish. Then, the individual with social phobia or the shy individual tends to become even more acutely aware of physical sensations as well as the negative thoughts and self-talk that acco mpany awareness of, for example, sweating or tremb ling. Next, as this exce ssive focus on the self continues, behavior is affected and the indivi dual can appear less friendly, which, in turn, partially confirms the shy indi vidual’s fears. Finally, the behavioral symptoms can produce more symptoms. For instance, talking quickly can lead to hyperventilation (Clark & Wells, 1995). Self-focused attention. Ingram (1990, p. 156) defines self-focused attention as “an awareness of self-refer ent, internally gener ated information.”

PAGE 66

55 Given the foregoing findings regarding fear of negative evaluation, it is not surprising that individuals with soci al phobia and shy individuals have a “narrowed attention to different elements of the social situation” (Kimble & Zehr, 1982, p. 39). Shy individuals thus ra rely obtain positive feedback about their social competence and, additionally, oft en interpret neutral social cues as evidence of negative evaluation by other s (Clark, 2001). That awareness can include information about somatic conditions or thoughts and feelings, including memories. Although the wording may differ, self -help books on shyness direct the reader to learn to think of others, rather than the self, during social interactions. For example, in Shyness, A Bold New Approach (Carducci, 1999), Carducci talks about expanding one’s comfort zone and l earning to deal with one’s tendency to be slow to warm up to new situati ons and new people (Carducci, 1999). One study on self-focused attention suggests t hat a person’s “. . representation of the self is [changed after effective treatm ent]. . in a more positive direction, primarily by decreasing the frequen cy of negative self-focused thoughts” (Hofmann, 2000, p. 722). The centrality of self-focused attenti on in research on shyness and social phobia is evident also in that a review has been written on the subject (Spurr & Stopa, 2002). Additionally, inspection of the items in measures for shyness and social phobia reveals that many of the items address the locus of attention. For example, in the Social Thoughts and Beli efs Scale (STABS), one item is “When

PAGE 67

56 other people laugh it feels as if they’re laughing at me” and another is “If there is a pause during a conversation, I feel as if I have done something wrong.” As mentioned earlier, fear of negative evaluation is held to be a central feature in shyness (Rapee, 1995) and reduc tion in fear of negative evaluation has been shown to be a good indicator of effectiveness of treatment for social phobia (Cox, Walker, Enns, & Karpinski, 200 2; Heimberg et al., 1995). Based on the previous review of lit erature, it seemed likely that an environment wherein fear of negative evaluation and self-foc used attention are diminished would be an ideal setting in which to explore furt her how these two correlates affect an individual’s behavior, thoughts, and feelings. It was believed that shy individuals woul d exhibit these characteristics to a lesser extent in the church setting. T he following section sets forth reasons why the social climate of the church wa s presumed to be more accepting. The Social Climate of the Church For evangelical churches, the Holy Bible is the absolute rule of conduct. Evangelicals believe the Bible is divinely inspired (II Timothy 3:16, NIV) and that its commands and teachings explain how Go d wants people to live here on earth. These directives from God on how to live one’s life as a believer are presented in the Biblical text in a system of commands principles, and specific applications of those commands and principles. The most fundamental level of Biblical law is the Ten Commandments (Exodu s 20:1-17, NIV). For people living before the time of Jesus, additional rules were pr ovided to explain how to apply these 10

PAGE 68

57 general laws to particular instances. For example, God’s holy people were told “. . do not go about spreading slander among your people” (Levitic us 19:16, NIV). For believers living after the time of Jesus, the New Te stament writings explain even more fully the intent of the 10 Commandm ents and the underlying principles by which believers are to liv e. Jesus summarized all of God’s laws when He said: “The whole law is summed up in this one command: “Love the Lord thy God with all your heart, with all yo ur soul, and with all your might and your neighbor as yourself ” (Deuteronomy 6:5, NIV). For another example of how the t eachings of Jesus illustrate the 10 commandments, one can consider Exodus 20:17: “Do not covet your neighbor’s house. You shall not covet your neigh bor’s wife, or his manservant or maidservant, or his ox or donkey, or any thing that belongs to your neighbor.” Jesus explained the underly ing principle of this commandment in His words recorded in the gospel of Matthew. Je sus reminded His listeners that the Old Testament included the comm and, “Do not commit adultery .” But I [Jesus] tell you that anyone who looks at a woman lustfully has already committed adultery with her in his heart.” Jesus explained that He came “. . not to abolish the Law or the Prophets [the Old Testament] but to fulfill them” (Matt hew 5:17, NIV). Besides the teachings of Jesus as re corded in the Gospels, the other New Testament writings also give specific, cl ear instructions for daily living that are consistent with God’s will, as revealed in the 10 Commandments. For example, believers are admonished to “love one anot her” (I John 3:11, NIV) and to “look after orphans and widows in their distress” (James 1:27, NIV).

PAGE 69

58 Two aspects of God’s commands for pres ent-day believers were of central importance in the current study: believers are to be accepting, com passionate, and kind to each other, believers are to be focu sed on helpin g others. Colossians 3:12 (NIV) says, “. . clot he yourselves with compassion, kindness, humility, gentleness, and patience. Bear with each other and forgive whatever grievances you may have against one another.” The believer in Jesus is to be focused on others. One of the clearest ex plications of the concern believers are to have for each other is found in the 12t h chapter of First Corinthians. All believers as a whole are compared to one body, called the “body of Christ” (First Corinthians 12:27, NIV). In the First Corinthians passage, the apostle Paul explains that believers are to be as concerned for each othe r as if one believer were an eye and another believer were a foo t, both parts of the same body. As Paul explains, each part of the body is to be concerned for ev ery other part and if one part hurts, whether the whole body hurts or feels good, the whole body feels the same way the part of the body does (First Corinthians 12:26, NIV). As described above, environmental charac teristics play an important role in how individuals manifest shyness. Three aspects of t he church environment were particularly relevant for the current study: the type of situations in which church members are often involved; the likelihood that the social world of the chur ch is perceived by church members and attenders as holding less potential for negative evaluation; and

PAGE 70

59 a focus upon others, rather than se lf, a key aim of ideal social interactions within the church setting. Rationale Shyness is a significant social probl em that can negatively affect many areas of an individual’s life (Henderson & Zimbardo, 2001). It seemed logical that contributing to existing theory of and measures for shyness also could contribute to alleviatio n of this problem. Two environmental factors that affect shyness are fear of negative evaluation (Weeks et al., 2005) and self focus (S purr & Stopa, 2002). Shyness has been studied most often in environments where these conditions are prominent. It was believed that if findings of the current study suggested that the Clark and Wells (1995) model explains how shyness operat es in a more accepting environment, research could continue with greater c onfidence in the Clark and Wells model. That is, the Clark and Wells model wo uld have more generalizability. Additionally, existing measures of sh yness have been constructed primarily utilizing college students and/ or, in the case of clinical psychologists and psychiatrists, clinical populations (Duk e et al., 2006). Existing theory has been heavily influenced by these measures. It was believed that further validation work on scores from a commonly used m easure (the BFNE) would be useful. Finally, it was believed that this study c ould identify new empirical indicators for future use in theory and in measures.

PAGE 71

60 Summary A review of the literature revealed that shyness is a common problem that affects more than 50% of the population, to varying degrees (Carducci & Zimbardo, 1997). Several themes were f ound to recur in the substantial body of research on shyness. Recurrent themes included whether shyness is a trait or inborn temperament, environmental, devel opmental and genetic influences, and gender and cultural differences (Cro zier, 2001; Crozier & Alden, 2001b; Heimberg et al., 1995; Jones et al., 1986; Leary & Kowalski, 1995; Stein, 1995). Current consensus is that behav ioral inhibition is one of many vulnerabilities that can lead to childhood and adult shyness but that genetics, the environment, and developmental issues all play an interactive role (Bruch & Cheek, 1995). Gender and cu lture have also been f ound to affect whether shyness develops in an individual. Evidence thus far regarding gender differences is inconclusive (Bruch et al., 1989; Crozier, 1990, 2001; Crozier & Alden, 2001b; Deardorff et al., 2007; Pilkonis, 1977; Pollard & Henderson, 1988; Rapee, 1995). Recent studies concerni ng cultural differences in shyness, however, have consistently found that i ndividuals from Asian cultures report higher levels of shyness than do Nort h Americans (Chen, 2000; Hsu & Alden, 2007; Pines & Zimbardo, 1978). Regarding the measurement of sh yness, most of the earliest questionnaires and scales were developed by psychologists and social psychologists. Following a change in diagnost ic criteria that resulted in a broader definition of social phobia, psychiatrists became more involved in research as

PAGE 72

61 well as in measurement. The most commonly utilized measures of shyness, social anxiety, and social phobia have been employed ext ensively in research, as indicated earlier in Table 4. According to Briggs et al. (1986), shyness research can be described as occurring in three phases: (a) descriptive studies beginning in 1986; (b) the popularization of shyness occurring in the mid to late 1970s with the publication of several books about shyness; and (c) trad itional empirical investigations that began in the early 1980s. Research from the mid 1980s onwards has been affected by the increased interest in so cial phobia and the use of drug therapies in treating shyness (McDaniel, 2003). Two cognitive models of shyness were of particular interest for the current study. The Rapee and Heimberg (1997) model focuses on how social situations activate a series of processes for a shy individual and how those processes create and maintain social anxiety. T he Clark and Wells (1995) cognitive model emphasizes the role of the shift in att ention that occurs when a social phobic perceives he/she is about to be evaluat ed negatively. The Clark and Wells model was foundational to the data gat hering and analytic fr amework of the current study because of its focus on the thoughts, feelings, behavior, and attentional focus in social situations (Clark, 2001). The Rapee and Heimberg model was utilized to provide addi tional explanatory insights. The literature review also revealed that the 10 environm ental situations most difficult for shy indivi duals (Crozier, 2001) are simila r to six basic activities of the social life of church members, as conceptualized by the author.

PAGE 73

62 Additionally, it was presumed that f ear of negative evaluation and self-focused attention would be of less intensity in the church setting. According to Biblical teachings, church members are to love (First John 3:11, NI V) and accept one another (Colossians 3:12, NI V). Based upon the previ ously stated findings and assumptions, the social clim ate of the church was believed to be an ideal setting in which to investigate the effect of reduced fear of negative evaluation on shyness. Because shyness is a significant soci al problem, it seemed logical that studies contributing to ex isting knowledge would be wort hwhile. It was believed that the current study could help confirm or disconfirm the Clark and Wells (1995) model of social phobia and the theory on whic h it is based. If shyness manifests itself in the experimental setting in the same way, it was believed that we could continue to employ the Clark and Wells model, as well as t heory, with greater confidence.

PAGE 74

63 Adapted from: Clark, D. M. (2001). A cogniti ve perspective on social phobia. In International handbook of social anxiety: Concepts, research, and interventions relating to the self and shyness (pp. 404-430) New York: Wiley. Figure 1 Developing an Idiosyncratic Versi on of the Clark-Wells Treatment Model Situation: Sitting alone, waiting for a worship service to begin Thoughts: Everyone is looking at me. They think it’s weird to be sittin g b y m y self. Focus on self Anxiety symptoms : Sweating Rapid heart rate Tremor Safety behaviors Avoid eye contact Pretend to be reading the bulletin Control body movements

PAGE 75

64 CHAPTER 3 METHODS Chapter 3 begins with a description of the participants in the study and an explanation of how they were selected for participation. The background of the quantitative instrument to be employ ed is presented next, followed by a description of the qualitativ e instrument. The procedur es section of Chapter 3 opens with an explanation of why the pr agmatic approach was selected. Next, details about how the quant itative data were gathered are provided, and threats to external validity are presented, followed by an explanation of how the qualitative data were gathered and a discu ssion regarding potential threats to legitimation of the qualitative phase. T he analysis section, which concludes the chapter, provides specifics about t he quantitative, qualitative, and mixed analyses conducted in this study as well as threats to legiti mation of the mixed methods design. Selection Eligibility Criteria The population for the quantit ative study consisted of three subgroups of individuals attending an evangelical chur ch in the Tampa Bay, Florida area: members, regular attenders, and visitors. The church was established at its current suburban location in 1997, having relocated from three other locations since its original founding in the 1950s. The church had a membership of approximately 1,000, and no de mographic information was available regarding its members. In this study, church me mbers were those individuals who have

PAGE 76

65 formally requested to join the church and to be included on the church membership list. Based on conversations the author has had with various church members over a period of 27 years, indi viduals generally attend a church for a number of weeks, often much longer, before joining. Ot her individuals, called in this study regular attenders, might be pr esent in any given worship service and might be on church mailing lists (but not t he membership list) by virtue of regular attendance at worship services and other church-sponsored gatherings. Other individuals present in any given worship service include first-time visitors and individuals who have attended at least onc e before but do not attend services on a regular basis. Sample A (the quantitat ive sample) was lim ited to one church (called Church 1 in this study) because of heterogeneity in beliefs and practices among churches. Any individual over age 21 who was present in the service or who heard about the study was eligible to participate. To control for developmental issues, only adults were selected. Participants Quantitative Phase Participation was solicited, as des cribed in the Quant itative Procedure section of this chapter. The plan for Samp le A (i.e., Phase I; Quantitative Phase) was to solicit and obtain participation fr om a minimum of 250 individuals, from Church 1. Sample A (see Figure 2) c onsisted of individuals completing the BFNE-S. Data collection was halted when 239 responses had been obtained, so the sample utilized in the current study consisted of 239 participants. One

PAGE 77

66 hundred and forty-two completed the BFNE -S electronically, and 97 utilized the paper-and-pencil format. Figure 2 Sampling Plan The sample size was selected because it represents an adequate number of participants to conduct all analyses. Specifically, a sample size of 239 was adequate to conduct the two sets of confir matory analyses of the 12-item BFNES scale (i.e., Research Question 1). According to Hatcher (1994): For the [confirmatory factor] anal yses discussed here, a minimally acceptable number of observation s would be the larger of 150 observations or 5 observations per par ameter to be estimated. Larger samples are always preferable, and if many model modifications are to be Sample A Respondents to PCI Survey (which includes the BFNES, General & Context-Specific Forms) n = 239 Sample B Responses to thoughts, feelings, and behaviors questions, n = 15 (Five each: low, medium, and hi g h

PAGE 78

67 made, substantially larger samples are required to arrive at a model that will successfully generalize to other samples. (p. 260) With 12 items, the sample size of 239 exceeded both criteria (i.e., 150 observations or 5 observations per parameter to be estimated). For the dependent samples t -tests (i.e., Research Questions 2 and 3), a sample size of 27 was needed to detect a statistically significant one-tailed difference with a moderate effect size (i.e., d = 0.5) with power = .80 and alpha = .05 (Erdfelder, Faul, & Buchner, 1996). Finally, for the analysis of variance (ANOVA), a sample size of 85 was needed to detect a statistically significant difference with a moderate effect size (i.e., f2 = 0.15) with power = .80 and alpha = .05 (Erdfelder, et al., 1996). Thus, the sample size of 239 was more than adequate for both the dependent samples t -test and the ANOVA. Qualitative Phase Appended to the BFNE-S was a stat ement requesting that participants who wished to answer more detailed questions about thoughts, feelings, and behaviors in the church setting to continue with the write-in response questions. From those who provided responses to the thoughts, feelings, and behaviors questions, 15 respondents’ comments we re analyzed, as detailed in the Quantitative Procedure and Q ualitative Procedure secti ons of this document. The decision to analyze responses to open-ended questions from 15 participants was made by the researcher and the panel of experts. The panel of experts (i.e., committee members), who have many years of research experience, determined thr ough discussions with the researcher that the sample

PAGE 79

68 space should be adequate. F ear of negative evaluation is a critical component of shyness and it was believed it would be present to some degr ee in most of the respondents identified as shy. Five of the 15 participants whose responses were selected for analysis had BFNE-S, Context-specific scores indica ting a high level of shyness, five had a medium level of shyness, and five had little or no shyness, as measured by responses to the BFNE-S, Context-specif ic, that were collected during Phase I. Standardized, rather than local norms were utilized for this part of the analysis to allow findings from the study to be ex trapolated, although with caution, to the general population. Norms based upon the general populat ion were the appropriate measure in this instance. The study revealed that shyness levels reported by church member s, regular attenders, and vi sitors of Church 1 for situations outside as well as inside the church setting were similar but not identical to the general population. Tabl e 9 contains the norms for the BFNE scale that have been reported in previ ous studies. The norms based upon the general population are the appr opriate measure in this instance, as described above. Table 9 Norms Reported for BFNE Scale Study Author(s) N Sample Description M SD Collins et al., 2005 82 Individuals with social phobia 51.50 7.30 99 Panic disorder 39.80 12.50

PAGE 80

69 30 Community sample (nonanxious) 29.20 8.20 Weeks et al., 2005 138-165 (missing data varied) Socially anxious 46.91 9.27 138-165 Non socially anxious 26.81 4.78 Duke et al., 2006 355 Shopping mall 32.30 7.34 Rodebaugh, Woods, Thissen, Heimberg, Chambless, and Rapee, 2004 1,049 Archival data (anxious and nonanxious) 29.41 7.72 Ethical Considerations During the informed consent process, t he purpose of the research, as well as costs and benefits to participants, was explained, and participants were advised that information they provided wo uld be treated confident ially. It was emphasized that withdrawal from the research was possible at any time, and contact information for psychological counseling would be provided to all participants as part of the informed consent process, although it was considered unlikely that participation in this study would cause psychological distress. The researcher described the study and solicited participation during a Sunday morning worship in June and again in July. The role of the researcher regarding participants was that of a non-participant data gatherer and analyst.

PAGE 81

70 Instruments Brief Fear of Negative Evaluat ion-Straightforward (BFNE-S). Two facts made the BFNE-S a good instrument for th is study. First, a brief scale was needed, and second, fear of negative evaluat ion is the core feature of social anxiety disorder (Weeks et al., 2005). T he purpose of the current study is to understand how shyness is manifested in a setting believed to have less potential for negative evaluation and less se lf-focus, two critical issues in shyness. It was, therefor e, essential and appropriate to have a measure of fear of negative evaluation. McNeil, Ries, and Turk (1995) described the Social Avoidance and Distress (SAD) scale and th e FNE scale as some of the most often-used measures of social anxiety. Cox et al. (2002) and Heimberg et al. (1995) found the FNE scale to be one of t he most sensitive outcome measures for social phobia treatment. Developed in 1969, concurrently with the SAD (Watson & Friend, 1969), the FNE measures apprehension, av oidance, and expectation of being negatively evaluated. The original FNE, consisting of 30 true/false items, had very good psychometric properties as report ed in a review by Orsillo (2001). Leary (1983) developed a shortened versio n, the Brief Fear of Negative Evaluation (BFNE) scale, which consisted of 12 of the origin al items (those with Pearson correlation coefficients greater than 0.50) in a 5-point Likert-type response format. Results indicated that the BFNE was sensitive to changes in social anxiety and panic disorder. Addi tionally, the 12-item BFNE correlated

PAGE 82

71 significantly with various measures of tr eatment responsiveness, such as somatic arousal and depression. The four items that were reverse-worded in the BFNE were straightforwardly worded in the current stud y. That decision is in accordance with the recommendation of Rodebaugh et al (2004) that the BFNE-S (BFNEStraightforward) be used to assess fear of negative evaluation. Rodebaugh et al. found that straightforwardly worded it ems “. . had significantly stronger relationships with theoretically related meas ures . than did the reverse-worded items” (p. 169). Three subsequent studi es (Carleton, Coll imore, & Asmundson, 2007; Duke et al., 2006; Weeks et al., 2005) supported the recommendation of Rodebaugh et al. to use t he BFNE-S, with all items worded straightforwardly for assessing fear of negative evaluation. Fu rthermore, as Weeks et al. (2005) note, “. . a sizeable body of literature demonstrates that when scales include a combination of straightforward and re verse-scored items, factor analyses frequently produce distinct factors based on th is difference in item construction” (p. 188). Recent studies have supported the score reliability as well as content-, construct-, and criterion-rela ted validity of the BFNE-S, as detailed in Table 10. Note that the studies included in T able 10 also include the BFNE, with the original reverse-worded items. The current study used the BFNE-S and it seemed reasonable to assume that resear ch findings from the BFNE apply to the BFNE-S as well. Each of the studies r eported in Table 10, ex cept for Collins et al. (2005) and Duke et al. (2006), empl oyed both the BFNE and the BFNE-S.

PAGE 83

72 Table 10 Research Regarding Validity and Reliability of Scores Obtained from the BFNE-S Validity and Reliability Study Author(s) and Findings Regarding Validity and Reliability Collins et al. (2005) n = 181 (82 social phobia, 99 panic disorder) and 30 nonanxious controls) Construct-related validity Principal components analysis ( n = 107). One factor accounted for 74% of variance, and pattern/structure coefficients ranged from .76 to .90. Concurrent validity Statistically signifi cant correlations with Social Avoidance subscale of the Fear Q uestionnaire-Social Phobia subscale and Beck Depression Inventory-II. Concurrent and discriminant validity BFNE correlated statistically significantly with social avoidance but not with agor aphobic avoidance or measures of panic or theoretically unrel ated variables of education and age. Discriminant validity Discriminant func tion analysis (DFA) revealed the BFNE differentiated significantly among groups of individuals with social phobia, panic disorder, and no psychiatric difficulties. Another DFA discriminated significantly between individuals with social phobia and panic disorder. Inter-item reliability No systemat ic differences on symptom or sociodemographic variables ( n = 107). Internal consistency Cronbach’s alpha =.97. Test-retest Two week test-re test correlation was .94, p < .001, with treatment effect size of 0.63. Criterion-related validity BFNE change scores correlated statistically significantly and positively with changes on t he Beck Anxiety Inventory, Anxiety Sensitivity Index, Be ck Depression Inventory-II, and both the Social Avoidance and Agoraphobic Avoidance subscales of the Fear Questionnaire. Carleton et al. (2007) Construct-related validity Confirmed unitary factor structure of BFNE-S. Convergent validity Demonstrated conver gent validity with measures related to social anxiety (Social subsca le of the Anxiety Sensitivity Index) though not as strong as expected. That was believed to be due to sample characteristics (i.e., the sample consisted of undergraduate students with no diagnosed mental or em otional disorders) Divergent validity Found divergent correla tions with measures of illness and

PAGE 84

73 injury. Weeks et al. (2005) n = 138-165 (missing data varied) Construct-related validity n = 165 Confirmatory factor analysis (CFA) supported one factor (and confirmed reve rse-worded items formed a second factor, affirming findings of Rodebaugh et al., 2004). Convergent validity BFNE-S statistica lly significantly correlated with all measures of social anxiety (L iebowitz Social Anxiety Scale .59, Social Interaction Anxiet y Scale .46, Social Phobia Scale .40, and Fear Ques tionnaire-Social Phobia Subscale .40.) Discriminant validity Lower correlations were found with Anxiety Sensitivity Scale, Penn State Worry questionnaire, and Beck Depression Inventory. Internal consistency Excellent in patient, control, and overall sample (Cronbach’s alpha =.92, 90, and .96 respectively). Rodebaugh et al. (2004) Construct-related validity Suggested a two-factor model but the most parsimonious explanation was method variance of reverse-worded items. Item Response Theory Analysis Both FNE and BFNE had good discrimination but BFNE discriminated across a wider range of the underlying construct Convergent validity Straightforwardly wo rded items had signif icantly stronger relationships with theoretica lly-based measures than did the reverse-worded items. Duke et al. (2006) n = 355, shopping mall, nonclinical, nonstudent sample Construct-related validity (CFA) Supported two-factor solution corresponding to straightforwardly and reverse-worded items Internal consistency Cronbach’s alpha = .94 Convergent validity Correlated significant ly in expected directions with Beck Depression Inventory and UCLA Loneliness Scale. The BFNE-S is an appropriate instru ment for the current study, as evidenced by the research reporting its use in large samples (Collins et al., 2005; Duke et al., 2006; Rodebaugh et al., 2004; W eeks et al., 2005). The author, who

PAGE 85

74 has experience administering surveys and who is a member of Church 1, administered the survey. The aut hor announced the survey, explained the purpose and advised how individuals wishi ng to do so could participate. Electronic participation was possible by usin g the Internet link that was printed in a flyer distributed with the weekly bulle tin. Paper and pencil copies could be picked up and returned to the Media Cent er in the lobby area. To avoid reactivity, the survey was entitled “Per sonal Concerns and Issues” (PCI) rather than The Brief Fear of Negat ive Evaluation Scale. Care was taken, however, to use the name of the BFNE-S scale in all other documentati on involved in the research. The PCI survey consisted of the components lis ted in Table 11. Table 11 Components of the Personal Concerns and Issues Scale Component Total No. of Items Format and Description BFNE-S, contextspecific and general 24 12 Likert-format context-specific items and 12 Likert-format general items Perceived acceptance 13 Rating of perceived acceptance by various people and in various situations Comfort in nine situations, contextspecific and general 18 Ten ratings (answered twice; once for context-specific and once for general situation) of comfort level in the ten situations known to be difficult for shy people Involvement in and connections with church members 8 Rating of depth of involvement in churchrelated activities and connections with church attenders Other items 8 Age, gender, race/ethnicity, and 5 ratingscale items from Zimbardo survey (Zimbardo, 1974) Total = 71 Note : 1-2 open-ended questions, specific to the local church and not included in the dissertation analysis, were included a fter the Zimbardo survey questions. See Questions to be Excluded from Analysis section below

PAGE 86

75 The BFNE-S was used both as a general and as a context-specific measure as was the checklist for perceiv ed comfort in the 10 situations. The instructions directed the participants to respond to the items a first time while thinking about how they felt during chur ch-related activities and a second time when thinking about how they felt in social situations in general. Score reliability, as measured by Cronbach’s alpha, was co mputed for the sample in the current study. The second component was a list of five items assessing the level of perceived acceptance the respondent had for various individuals and in various situations inside and outside the church se tting. The list of situations also included two additional situations (a t place of work/business and with one’s family at home) where only one answer wa s required, rather than one answer for inside the church and one answer for outside the church. The third component of the PCI Scale was a checklist for respondents to indicate their levels of comfort in the 10 situations, in general and in the context of the church. The fourth group of items concerned the level of involvement in the church and church activities. Cronbach’s alpha was computed to ascertain t he level of internal consistency for the second, third, and fourth components of the PCI Scale scores. The final component of the PCI Scale was three items eliciting participants’ age, gender, and race/ethnicity and five items from Zi mbardo’s (1977) shyness survey. Table 12 presents items for the BFNE-S.

PAGE 87

76 Table 12 BFNE-S Items Instructions: Read each of the following statements carefully and indicate how characteristic it is of you according to the following scale. 1 = Not at all characteristic of me 2 = Slightly characteristic of me 3 = Moderately characteristic of me 4 = Very characteristic of me 5 = Extremely characteristic of me 1. I worry about what other people will thin k of me even when I know it does not make any difference. 2. I am concerned if I know people are forming an unfavorable impression of me. 3. I am frequently afra id of other people noticing my shortcomings. 4. I often worry about what kind of impression I am making on someone. 5. I am afraid others will not approve of me. 6. I am afraid that peopl e will find fault with me. 7. Other people’s opinions of me bother me. 8. When I am talking to someone, I worry about what they may be thinking about me. 9. I am usually worried about what kind of impression I make. 10. If I know someone is judging me, it has a big effect on me. 11. Sometimes I think I am too concer ned with what other people think of me. 12. I often worry that I will say or do the wrong things. Table 13 presents the remaining component s of the PCI Scale. For items in the Ten Situations, respondents were asked to specify their levels of comfort using a 5-point Likert-format scale. A 5-poi nt Likert-format scale also was utilized for rating perceived acceptance. The items concerning extent of involvement in

PAGE 88

77 the church, age, sex, race/ethnicity had fi xed-response options. Four of the five items from Zimbardo’ s (1977) shyness survey had fixed-response options, with the last item being open-ended in format. Table 13 Other Components of the PCI Scale Perceived Acceptance (respondent to indi cate how accepted he/she generally feels as an individual: 1. by the people in this church in general 2. by friends and acquaintances you have in this church 3. by the pastor and other leaders in this church 4. when you meet someone at church you do not know 5. by friends and acquaintances you have who do not attend this church 6. when you meet someone you do not know outside this church setting 7. by yourself while in this church 8. by God while in church 9. outside your home in general (in a store, at a t heme park, etc.) 10. at your place of work/business ( If you are not employed outside the home, please respond based on how you feel when you go into a relatively formal setting, like renewing your driver’s license) 11. with your family at home 12. by yourself outside church 13. by God outside church Level of Comfort in the Ten Situations Known to be Difficult for Shy People 1. Being the focus of attention 2. Large groups 3. Small groups 4. Authority figures by virtue of know ledge (intellectual superiors, experts) or

PAGE 89

78 by virtue of role (police, teachers, superiors at work) 5. Social situations in general 6. New interpersonal situations in general 7. Strangers 8. Situations where assertiveness is required (e.g., when complaining about faulty service in a restaurant) 9. Being evaluated or compared with others (e.g., when being interviewed, when being criticized) 10. An opposite sex group or a member of the opposite sex (* Adapted from Crozier, W. R. (2001). Understanding shyness : P sychological perspectives. China: Palgrave.) Extent of Involvement in Church Activities 1. How long have you been a member of this church? 2. For how many years of your life hav e you been a member of any church? 3. Do you have a designated ta sk or job in this church? 4. Approximately how many times a m onth do you participate in church activities outside worship services? 5. Approximately how many times a m onth do you attend worship services? 6. Approximately how many close fr iends do you have at this church? 7. Approximately how many acquaintanc es do you have at this church? Demographic Variables Age (1 item), gender (1 item), race and /ethnicity (1 item) Self-rating of shyness 1. Do you consider yourself to be a shy person? (If no, skip to #4) 2. If yes, have you always been shy? 3. If you are currently shy, is that in most or only in some situations? 4. Was there ever a prior time in your life when you were shy? 5. How desirable is it for you to be shy?

PAGE 90

79 Questions to be Excluded from Analysis As a professional courtesy to C hurch 1, two open-ended questions specifically about the church were added after the PCI scale. The pastor selected the topic of the questions, and t he researcher worked with the pastor to ensure the questions were appropriately worded. Responses were analyzed using inductive constant comparison anal ysis. Findings were presented to the pastor in a separate report but are not included in the curr ent study. Additionally, a copy of the completed dissertation wil l be presented to the pastor. Extreme caution was exercised throughout the analysi s to maintain confidentiality of all respondents and to ensure that no individual was identifiable in the report on the church-specific question(s) or in the completed dissertation. Qualitative Questions Written responses were obtained to the thoughts, feelings, and behaviors questions listed in Table 14 for each of the six church situations. The responses were obtained via writein items on the electronic SurveyMonkey form (http://www.surveymonkey.com) or in paper /pencil format for those respondents who preferred that mode. One hundred and forty-two participants utilized the electronic format, and 97 chose the paper/pencil format. Table 14 Thoughts, Feelings, and Behaviors Questions Thoughts, Feelings, and Behaviors Questions (Behavior) “If I were with you in a typical c hurch service, what would I probably see you do?”

PAGE 91

80 (Thoughts) “What are some of the t houghts you might be having?” (Feelings) “Please describe how you would be feeling.” Procedures Pragmatist procedure. Table 15 presents the distinguishing features of the research paradigm known as pragmat ism. The pragmatic approach was considered appropriate for the research que stions. The current study is a mixed methods study in that it combined qua litative and quantitativ e approaches in a single study and employed a sequential, equi valent status design, specifically QUAN/QUAL (Tashakkori & Teddlie, 1998) The instrument used for the regression analysis and selection of Sa mple B was quantitative in nature, whereas the thoughts, feelings, and behav ior questions were qualitative in nature. The BFNE-S was used compar atively, being administered and analyzed as a general measure and as a contextspecific measure, and compared to norms in existing literature. The questions about thoughts, feelings, and behaviors in the six church situations were exploratory and were analyzed thematically as well as quantitized and linked to existing theory and instrumentation. A mixed methods design was appropria te because the research questions were both quantitative and qualitative (Yin, 2003). Additionally, from a measurement perspective, it was believe d that mixed methods could reveal manifestations of shyness that would ot herwise remain hidden. Also, use of multiple methods avoided a grave shor tcoming inherent in monomethod studies

PAGE 92

81 (Campbell & Fiske, 1959). “Triangulation of methods” was the initial impetus to greater utilization of mixed methods (T ashakkori & Teddlie, 1998, p. 41). Campbell and Fiske (1959) proposed using multiple measurement methods to help assure that results were attr ibutable to the ac tual phenomenon being studied rather than the method being empl oyed. The mix of quantitative and qualitative questions provided methodolog ical triangulation (Tashakkori & Teddlie, 1998). Table 15 Primary Distinguishing Characterist ics of the Prag matist Paradigm Paradigmatic Element Pragmatist Distinctive Methods Quantitative and qualitative Logic Deductive and inductive Epistemology Both objective and subjective points of view Axiology Values play a large role in interpreting results. Ontology Accept external reality. Chose explanations that best produce desired outcomes. Causal linkages There may be causal relationships but we will never be able to pin them down. Tashakkori and Teddlie, 1998, p. 23. Quantitative Procedure The quantitative component utilized descr iptive, correlational, and causalcomparative designs. For participants in Sample A, descriptive statistics were presented for scores from the BFNE-S, General and Context-specific, and for items on the checklist of perceived accept ance in situations inside and outside

PAGE 93

82 the church setting. Also for participants in Sample A, a correlational design was utilized to examine the relationships among fear of negative evaluation and length of membership, regularity of a ttendance, and frequency of interpersonal contact (i.e., Research Ques tion 7). For participants in Sample B, a correlational design was utilized to measure the extent to which self-reported shyness was associated with attentional focus upon se lf and negative quality of thought in the six church situations under inve stigation in the current study. A causal-comparative design is one in which groups known to have differed in the past “. . either in the dependent variable or the independent variable. . ” are compared retroactively in order to infer “. . relationships (especially tentative causal ones” (Tashakkori & Teddlie, 2003, p. 704). The current study also employed a causal-com parative design in that it involved a comparison of (a) perceived fear of negativ e evaluation in the church setting and perceived fear of negative evaluation out side the church setting; (b) perceived levels of acceptance in the church se tting and perceived levels of acceptance outside the church setting; and (c) perceiv ed levels of comfort in the church setting and perceived levels of comf ort outside the church setting. Participation was solicited in the follo wing manner in the church where the data were collected. On the Sunday t hat data were collected, a flyer was included in the bulletin t hat every individual receiv ed upon entering the worship service. The flyer gave a brief explanati on of the purpose of the research and included a link to the website where t he Personal Concerns and Issues Survey

PAGE 94

83 could be accessed. It was emphasized to participants that the electronic survey software does not record the res pondent’s name or em ail address. In addition to the potential partici pants being informed via the flyer that was distributed prior to the worship serv ice, the author gave a brief presentation during the worship service. She announced that paper versions of the survey were available in the Media Center imm ediately after service for individuals who wished to pick up a paper/pencil version and that the completed surveys could be returned to the same place. Ninetyseven individuals chose the paper/pencil option. The author of the current study was t he data collector, who is experienced in administering surveys, and analyzing qua ntitative and qualitative data. Data collection procedures were systematic in that standard responses to potential questions participants might ask were pr epared and utilized by the researcher. For example, the researcher was prepared to respond to the question, “Why are you doing this survey?” with the reply “For my dissertation and to help the church.” The author updated her training, as r equired by the University of South Florida Institutional Review Board, and obtained approval for the study before any data were collected. All data were handled confidentially, with participants so advised, and data were kept in a lo cked file cabinet in the author’s home office. Electronic data files were k ept on the home office computer. As mentioned before, a referral for psychological counseling was provided, although

PAGE 95

84 it was believed unlikely that participating in the study would cause psychological distress. No formal pilot study was conduct ed; however, friends, acquaintances, and colleagues of the author provided informal feedback on instruments and the process of using the SurveyMonkey so ftware in the proposed manner. The purpose of the informal pilot work was to verify that the in struments would not require too much time, that the thoughts, feelings, and behavior questions would elicit in-depth responses, and that the author communicated clearly to participants. Internal validity According to Trochim (2006), internal validity is “the approximate truth about inferences r egarding cause-effect or causal relationships” or as Babbie (2004, p. 230) ex plains, “the threat to internal validity is present whenever anything other than th e experimental stimulus can affect the dependent variable.” Regarding this study, in ternal validity would tend to be low because this is a non-experimental study; however, the instrum ent being utilized, the BFNE-S, is a psychometrically sound in strument. In the current study, the original version of the quant itative instrument to be utilized, the BFNE-S, has a long history of use in research, and vari ous studies have explored psychometric properties of the instrument (Ors illo, 2001; Rodebaugh et al., 2004). Psychometric properties obtai ned with the instru ment as a context-specific and as a general measure were compared to available norms, although it is known that most of the norms were obtained using college student samples and, more recently and on only a few occasions, with clinically anxious samples and, even

PAGE 96

85 more recently, the general population. The author recognized that using norms thus obtained constitutes a threat to in ternal validity; however, the BFNE-S seems to be the instrument most suited fo r use in the current study. Particular attention was paid to psychometric properti es of the BFNE-S, both as a general and as a context-specific measure. External validity Issues relating to external validity were carefully considered. Selection was a threat to external validity because individuals who elected to participate likely possessed attr ibutes different fr om those declining involvement in the study (Tashakkori & Teddlie, 1998). The BFNE-S was used to explore whether the same symptoms that predict shyness in a general setting also predict shyness in the church setti ng. The BFNE-S also was used as a screening device to identify responses to the write-in questions from shy and non-shy individuals. It was recogniz ed that the convenience sampling used for this study limits generalizability to other populations and constitutes a threat to population validity. The quantitative and qualitative components of this study have equivalent status in this study in that “. . bot h types of methods ar e given equal weight (Tashakkori & Teddlie, 2003, p. 285). As Tashakkori and Teddlie (2003) observe, this type of design is often employed in dissertation research in educational settings. The first sample consisted of responses from 239 individuals, and the second, purposive sa mple of 15 participants was selected from the first sample. The second sa mple of 15 participants was utilized to extend existing knowledge. There was no treatment group in the study. The

PAGE 97

86 quantitative instrument was administered as a general and as a context specific measure. Additionally, scores from the BFNE-S were used to identify participants having self-reported high, m edium, and low to minimal levels of shyness. Responses from participants for the 15 se ts of write-in comments that were qualitatively analyzed were selected based on scores from the BFNE-S, Contextspecific. The research paradigm, as stated ear lier, was pragmatism. Pragmatism embraces aspects of both post-positiv ism and constructivism (Tashakkori & Teddlie, 1998). Thus, the research design was appropriate for the proposed study. As Tashakkori and Teddlie (2003) state, Pragmatism supports the use of both qualitative and quantitative research methods in the same study. . [and] Pragmatist researchers consider the research question to be more important than either the method . or the paradigm that underlies the method. (p. 21) Findings have been shared with members of the dissertation committee and with the leadership of Church 1 wher ein the data were gathered. Qualitative Procedure The qualitative component utilized a multiple case study, embedded research design, with multiple units of analysis (Yin, 2003, p. 40). Table 16 presents Yin’s classification scheme for t he characteristics of research design.

PAGE 98

87 Table 16 Basic Types of Designs for Case Studies Single-Case Designs Multiple-Case Designs Holistic (single unit of analysis) Type I Type 3 Embedded (multiple units of analysis) Type 2 Type 4 Yin (2003, p. 40). It is helpful to bear in mind that case study research relies on “analytical generalization . [wherein] . the in vestigator is striving to generalize a particular set of results to some br oader theory” (Yin, 1 984, pp. 43-44). Replication, rather than sampling logic, is employed, in that the same results are predicted for the number of cases availa ble within time and financial constraints (Yin, 2003). Each individual case is analogous to a single experiment and the analysis follows “cross-experimental rat her than within-exper imental design and logic” (Yin, 1984, p. 53). The researcher was not the sole vo ice representing the participant. Results were discussed with committee mem bers as well as other professional colleagues. The 16 detailed responses were assessed for interpretive validity. A disinterested peer, a colle ague in the field of education who had worked as a coder on two previous occasions, worked as a second coder. The researcher specifically chose a second coder who had very minimal experience with attending a church. After training, the second coder reviewed a sample of coded

PAGE 99

88 responses, explanatory discussion with t he researcher ensued, and then the second coder reviewed each of the 16 se ts of responses to the write-in questions, attaining a satisfactory rate of agreement (90%). The second coder thus affirmed that the interpretations of the author stemmed directly from the findings and thus exhibited interpreti ve consistency (Collins, Onwuegbuzie, & Jiao, 2007). Rich data were collected because t he thoughts, feelings, and behaviors questions contained ideas of different individuals about the issues under investigation. Although interviews were not utilized, the desired richness of response was achieved through the ext ended response questions (Rubin & Rubin, 2005). Sets of comments from five individuals were selected in each category of minimal, low, medium, and high shyne ss, based on scores from the BFNE-S, Context-specific. To gain entry into the research context, the researcher secured permission from the local pas tor. Church members and attenders were invited to participate either during a br ief presentation at a worship service or via mail or internet, depending on what permission was given. Verification included utilizing the ext ensive experience of the researcher. The researcher has had long-term exper ience with being shy and has conducted numerous personal conversations for mo re than a decade regarding shyness and specifically the effects of shyness in the church setting. The researcher also has been a church member for 26 years. Terminology and expressions utilized by the participants were thus easil y understandable. Traditional member

PAGE 100

89 checking (Lincoln & Guba, 1985) was not possible because interviews were not utilized. Data triangulation was employed in that both quantitativ e and qualitative data were utilized. The aforementioned notwi thstanding, it is recognized that the quantitative instrument and responses to the thoughts, feelings, and behaviors questions were both subject to self-repor t bias (Maxwell, 2005). Legitimation threats were addressed by searchi ng for discrepant evidence and negative cases, those “instances and cases that do not fit within the pattern” (Patton, 1990, p. 463). In addition to testing for rival explanati ons, findings were discussed with committee members and co lleagues. Researcher bias was controlled for to the extent possible by the author maintaining awareness that she is motivated to help shy individuals, which might have affected analysis of the data. In all aspects of the study, the re searcher endeavored to achieve intersubjectivity, by deliberately shifting bet ween objectivity and subjectivity (Morgan, 2007; Onwuegbuzie & Johnson, 2006 ). To minimize bias during data analysis, “analyst triangulation” (Patton, 1990, p. 464) was employed in that a second rater reviewed the codings of responses to the thoughts, feelings, and behaviors questions. Mixed Methods Procedures Threats to legitimation. Threats to legitimation of findings stemming from the qualitative and mixed methods procedur es in the following discussion are reviewed within the framework of the “pr oblem of legitimation” (Onwuegbuzie &

PAGE 101

90 Johnson, 2006), which “refers to the difficulty in obtaining findings and/or making inferences that are credible, trus tworthy, dependable, transferable, and/or confirmable” (p. 52). The fi rst type of legitimation to be employed in the current study was sample integration legitimati on. Care was taken when integrating inferences from the quantitat ive data collected from the larger sample (Sample A) and inferences from the smaller, qualit ative sample (Sample B). It was recognized that making “meta-inferences” (Tashakkori & Teddlie, 2003, p. 686) by combining inferences from the quant itative and qualitative phases might not be justified. According to Onwuegbuzie and Johnson (2006), “Inside-outside legitimation [concerns]…the ex tent to which the researcher accurately presents and appropriately utilizes the insider’s view and the observer’s views for purposes such as description and explanat ion” (p. 55). Peer review, through discussions with committee members, was utilized to obtain an outsider’s view that was as accurate as possible, and t he extensive experience of the researcher with the church setting and with shyness were utilized to obtain an accurate insider’s view. As discussed earlier in the Qualitative Procedure section, peer review involved having a disinterested outsi der work as a second coder. Finally, attempts were made to enhance “weakness minimization” (Onwuegbuzie & Johnson, 2006). The ri chness of the qualitative data, most specifically the in-depth responses to the thoughts, feelings, and behaviors questions, helped compensate for the inabilit y of the quantitative data to explain

PAGE 102

91 why individuals did or did not repor t feeling shy in the environment under investigation. As stated earlier, the current stud y was a mixed methods study that employed a sequential, equiva lent status design, specifically QUAN/QUAL (Tashakkori & Teddlie, 1998). Deciding to utilize mixed methods was an iterative process as the researcher considered the purposes of the research and the specific questions to be addressed. Tas hakkorri and Teddlie (2003) indicate that when the purpose of the research is to hav e a “social [or] institutional impact,” mixed methods research is appropriate in that the “research can be used to test hypotheses related to values idiosyncratic to the contex t” (p. 186). Examination of Table 18 (Analysis Plan) reveals that the specific research questions were all focused upon increasing knowledge regar ding how shyness operates in a setting—the church setting—where two of the most import ant environmental variables affecting shyness were presumed to be substantially different. The two central purposes of the current study were to confirm existing theory and provide alternative explanations of behaviors t he theory addresses and to provide further validation work on a frequently used meas ure of shyness. Fulfilling those purposes may help advance the understanding and treatment of shyness in all settings and specifically in educationa l settings. A mixed method research design was therefore considered appropriate. Early in the study, the researcher decided upon the dimensions of “ paradigm emphasis (deciding whether to give the quantitative and qualitative components of a mixed study equal status or to give one paradigm the dominant

PAGE 103

92 status) . time ordering of qualitative and quantitative components” as well as the degree to which research methods would be mixed and in what temporal order (Onwuegbuzie & Johnson, 2004, p. 19). As the design evolved through di scussions with committee members, decisions were made regarding other dimens ions of research design. Table 17 presents other dimensions of the final research design that was developed in concert with committee members. It is im portant to note that for purposes of visual presentation, some components of Table 17 are summarized. Table 18 (Analysis Plan) provides more detail regarding analyses that were conducted. Analyses Quantitative Analysis A variety of quantitative analyses were undertaken to address the quantitative research questions, as depicted in Table 18. Statistical tests were conducted at the .05 level of statistical significance, using SPSS statistical software (SPSS Inc., 1998).

PAGE 104

93 Table 17 Dimensions of Research Design Research Question Instrument (Components of the PCIS) Analysis Nature of Analysis Li nkage to Theory Phase of Study, Data Type and Source 1. What are the psychometric properties of the BFNE-S, general and context specific? What are the psychometric properties of the BFNE-S, General and Context-specific, i.e., the church? (A) Descriptive statistics and Cronbach alpha (B) Confirmatory factor analysis Confirmatory and exploratory Validation work on existing instrument Phase I, Quantitative data, Sample A (n = 239) 2. What is the difference in perceived fear of negative evaluation in the church setting compared to the non church setting? What is the difference in perceived fear of negative evaluation in the church setting compared to the non-church setting? A dependent samples t tests to examine differences in means Confirmatory Validation work on existing instrument 3. What is the difference in perceived fear of negative evaluation between the genders in the church setting compared to the non church setting? What is the difference in perceived fear of negative evaluation in the church setting compared to the non church setting for males and females? A repeated measures ANOVA. Confirmatory and exploratory Exploratory 4. What is the difference in perceived fear of negative evaluation among the races in the church What is the difference in perceived fear of negative evaluation in the church setting A repeated measures ANOVA. Confirmatory and exploratory Exploratory

PAGE 105

94 setting compared to the non church setting? compared to the non church setting for individuals of different races? 5. What is the difference in perceived acceptance between people inside and outside the church setting? What is the difference in perceived acceptance between people inside and outside the church setting? (A) Descriptive statistics (B) dependent samples t -test Exploratory and confirmatory Exploratory and confirmatory 6. What is the difference in selfreported levels of comfort outside the church setting and inside the church? What is the difference in selfreported levels of comfort outside the church setting and inside the church setting? A dependent samples t test of the difference between mean levels of comfort in the general setting and in the Context-specific setting Exploratory and confirmatory Exploratory and confirmation of theory or alternative explanations of behavior theory addresses 7. To what extent do context-specific issues relate to self-reported levels of fear of negative evaluation? To what extent do Context-specific issues relate to selfreported levels of fear of negative evaluation? An analysis of variance for effect on fear of negative on: length of membership, formal place of service, regularity of attendance, and frequency of interpersonal contact. Exploratory Exploratory 8. How do shy people typically think, feel, and behave in an environment with less fear of negative How do shy people typically think, feel, and behave in an environment hypothesized to (A) Frequencies for thoughts, feelings, and behavior consistent with the model versus those that are Content analysis (classical, manifest, latent and inductive constant comparison) Confirmation of theory or alternative explanations of behavior theory addresses Phase II, Qualitative data, Sample B (n = 15)

PAGE 106

95 evaluation and selffocus? have less fear of negative evaluation and self-focus? inconsistent. (B) Thematic analysis of unaccounted-for comments 9. To what extent is self-reported shyness associated with attentional focus upon self and negative quality of thought in the six church situations? To what extent is self-reported fear of negative evaluation associated with attentional focus upon self and negative quality of thought in the six church situations? (A) Pearson correlations between mean attentional focus score and scores from BFNE-S, context specific (B) Pearson correlations between thought quality scores and scores from BFNES, Context-specific. Latent content analysis For purposes of visual presentation, some components of Tabl e 17 are summarized. Table 18 (Analysis Plan) provides more deta il regarding analyses to be conducted.

PAGE 107

96 Table 18 Analysis Plan Purpose Research Question Research Hypothesis Data Collection Instrument (Components of the PCIS) Analysis 1. To evaluate the psychometric properties of the BFNE-S in a nonstudent, non-clinical randomly selected sample of a previously unstudied population. What are the psychometric properties of the BFNE-S, General and Contextspecific, i.e., the church? The BFNE-S, General and Context-specific, displays psychometric properties in the sample that are similar to those demonstrated for other populations taken from university or clinical settings. BFNE-S, general and context-specific versions, and demographic questions (A) Descriptive statistics and Cronbach alpha for scores from the BFNE-S, General and Context-specific. (B) Confirmatory factor analysis for scores from the BFNE-S, General and Context-specific. 2. To compare levels of perceived fear of negative evaluation inside and outside the church setting. What is the difference in perceived fear of negative evaluation in the church setting compared to the non-church setting? Perceived FNE is lower in the church setting compared to the nonchurch setting. BFNE-S, general and context-specific versions A dependent samples t -test to examine differences in means 3. To compare levels of perceived fear of negative evaluation between the genders inside and outside the church setting What is the difference in perceived fear of negative evaluation in the church setting compared to the non church setting for males and females? The difference in FNE between the church and non-church setting is the same for males as for females. BFNE-S, general and context-specific versions A repeated measures ANOVA to compare perceived FNE for males and females in the church setting compared to the non church setting. 4. To compare levels of perceived fear of negative What is the difference in The difference in FNE between the church and BFNE-S, general and context-specific versions A repeated measures ANOVA to compare perceived FNE for

PAGE 108

97 evaluation among the races inside and outside the church setting perceived fear of negative evaluation in the church setting compared to the non church setting for individuals of different races? non-church setting is the same for different races. Caucasians, African Americans, Hispanics, and persons of multiracial background in the church setting compared to the non church setting. 5. To compare the levels of perceived acceptance by people inside and outside the church setting. What is the difference in perceived acceptance between people inside and outside the church setting? Levels of perceived acceptance by people in the church setting are higher than the levels of perceived acceptance by people outside the church setting. Checklist of perceived acceptance in situations inside and outside the church setting. (A) Descriptive statistics for the items on the perceived acceptance checklist, outside and inside the church setting. (B) A dependent samples t -test for the difference in levels of perceived acceptance by people inside and outside the church setting. 6. To understand how shyness manifests itself in an environment believed to induce higher comfort levels. What is the difference in selfreported levels of comfort outside the church setting and inside the church setting? Levels of comfort perceived by people in the church setting are higher than the levels of comfort outside the church setting. Checklist of 10 situations known to be difficult for shy persons, with respondents to indicate level of self-reported comfort in the 10 situations, inside and outside church setting. A dependent samples t -test of the difference between mean levels of comfort in the general setting and in the Contextspecific setting. 7. To understand how context-specific issues (extent of involvement in church activities) moderate self-reported fear of negative evaluation in the church setting To what extent do Context-specific issues relate to self-reported levels of fear of negative evaluation? Greater depth and breadth of involvement in church activities are associated with reduced self-reported fear of negative evaluation. Eight questions regarding extent of involvement in church activities. A multiple regression for fear of negative evaluation using length of membership, regularity of attendance, number of activities participated in per month, and number of close friends as predictor variables 8. To seek confirmation or disconfirmation of theory How do shy people typically think, feel, At least 75% percent of individuals with high 15 in-depth responses* to write-in questions, with the Percentage of thoughts, feelings, and behaviors for

PAGE 109

98 via examining the extent to which existing theory explains the thoughts, feelings, and behaviors of shy individuals in the church setting. and behave in an environment hypothesized to have less fear of negative evaluation and self-focus? levels of FNE will report thoughts, feelings, and behaviors related to six church situations that are consistent with the Clark and Wells model and that will be at least 10 & more than those with low levels of FNE. 15 individuals to have selfreported minimal-to-low, medium and high levels of shyness (five from each category) as measured by the BFNE-S. individuals with high levels of shyness in the church setting that were consistent with the Clark and Wells (1995) model. 9. To seek confirmation or disconfirmation of theory via examining the extent to which attentional focus is related to self-reported levels of fear of negative evaluation. To what extent is self-reported fear of negative evaluation associated with attentional focus upon self and negative quality of thought in the six church situations? Focus upon self and negative quality of thought related to the six church situations are associated with higher levels of selfreported fear of negative evaluation. 15 in-depth responses described above. (A) A Pearson correlation between mean attentional focus score and scores from BFNE-S, Context-specific (B) A Pearson correlation between thought quality scores and scores from BFNE-S, Context-specific

PAGE 110

99 Qualitative Analysis Any comments that do not fit the t heory-derived categories were reviewed for emergent themes. A variable-oriented analysis was undertaken. Analyses were both exploratory and confirmatory. The analyses were expl oratory in that the research was conducted to underst and how well existing theory explains shyness in a previously unexplored setting and confirmatory in the sense that it was believed the theory would explain most of the behaviors of shy individuals who manifest symptoms of shyne ss in the church setting. Three types of qualitative analyses were employed: manifest content analysis, manifest and latent content analysis, and inductive constant comparison, also called emergent themes analysis. Manifest content analysis, which Boyatzis (1998) defines as “. . the analysis of t he visible or apparent content of something” (p. 16) was used to categorize responses of individuals with high, medium, and low-to-minimal le vels of shyness to the thoughts, feelings, and behaviors questions. The manifest content analysis also was classical in the sense that the researcher counts “. . the number of times each code is utilized” (Leech & Onwuegbuzie, 20 07, p. 569). Comments of individuals that could not be categorized into the components of the Clark and Wells model were analyzed using the inductive form of constant comparison in that codes were allowed to emerge from t he data (Glaser & Strauss, 1967). Boyatzis (1998, p.16) defines latent content analysis as “. . looking at the underlying aspects of the phenomenon under investigation. It is more interpretive than manifest content anal ysis.” Latent content analysis was utilized

PAGE 111

100 for the 16 sets of responses selected for Phase II of the study. The researcher attempted to infer what seemed to be the point of attentional focus and the negative or positive quality of thought. Latent content analysis is appropriate because it often focuses upon “. . im portant (although hidden) aspects of individual and social cognition underlyi ng behaviors rather than assessing the behaviors that are easily observable ( Leech & Onwuegbuzie, 2007, p. 12). Table 19 Map for Content Analyses of Responses to Thoughts, Feelings, and Behaviors Questions/Interviews Content to be Analyzed Content Analysis Type Responses for individuals with high, medium, and minimal-to-low levels of shyness Features of Clark and Wells cognitive model (safety behaviors, high standards, and conditional beliefs) Manifest content analysis Unaccounted for comments In ductive constant comparison For all respondents – attentional focus scoring sheet Latent content analysis No qualitative software was necessary because of the relatively small amount of data and the fa ct that most of the coding em ployed a priori categories. Responses to thoughts, feelings, and behaviors questions for participants with high and medium levels of shyness were mapped to the components of the Clark and Wells (1995) model of social phobia, as depicted in Table 20.

PAGE 112

101 Responses were entered into an Excel spreadsheet and tallied. The unit of analysis for the qualitative data was each response to each question. In other words, for each participant, there were 18 units of analysis (i.e., responses to thoughts, feelings, and behaviors in each of the six church situations). Equal weight was given to the key issue—that is, the proportion of the comments in each situation that the Clark and Wells model explained or did not explain. An example of a comment about a behavio r that would fit the model under the label of safety behaviors would be if a participant mentioned arriving late, keeping one’s eyes averted, or leaving the building as soon as the service is concluded. As another example, if a re spondent stated “I felt bad at the wedding reception because when you meet a str anger, you should always smile and look poised and make a comment that is just right for that person and I just turned away,” that comment woul d fit the model because it is an excessively high standard. Inductive constant com parison analysis was conducted of all comments that did not fit components of the model. Attentional focus scores were also tallied using an Excel spread sheet. Combined Qualitative and Quant itative Data Analysis A sequential analysis of both quantitativ e and qualitative data (i.e., mixed analysis) was undertaken, with qualitative data being quantitized. A sequential mixed analysis was used in that “. . one set of data [the quant itative data, or surveys] was analyzed prior to analyzing the other dataset” (Onwuegbuzie, Slate, Leech, & Collins, 2007, p. 6) [the qualitative data, or open-ended responses to the thoughts, feelings, and behavior questi ons]. The mixed analysis gave

PAGE 113

102 approximately equal weight to the quantit ative and qualitative portions of the study and thus was an equivalent status design. Specifically, responses to survey questions will be utilized to addre ss five of the nine research questions whereas checklist data and write-in res ponses to the thoughts, feelings, and behaviors questions were utilized for four research questions. Using the typology developed by Onwuegbuzie et al. ( 2007), the study was an equal-status sequential multitype mi xed analysis. The mixed analysis included the seven stages of mixed analysis explicated by Onwuegbuzie and Teddlie (2003): (a) data reduction, (b) data display, (c) data transformation, (d) data co rrelation, (e) data consolidation, (f) data comparison, and (g) data integration. Table 20 pr esents definitions of the seven stages of the mixed analysis process and indicates which stage was employed for each of the nine research questions in the current study. Classical, manifest and latent content analyses, as well inductive constant comparison analyses were used for res ponses to the thoughts, feelings, and behaviors questions as described previously in the Qualitative Analysis Section. Descriptive statistics and graphical displays were used to present the results of both quantitative and qual itative analyses.

PAGE 114

103 Table 20 Seven Stages of Mixed Analysis Process and Research Questions Stage of Mixed Analysis Definition* Research Question(s) Data reduction Reducing dimensionality of quantitative data and qualitative data 1, 5, 8, and 9 Data display Describing visually the quantitative data 1-9 Data transformation Data are quant itized and/or qualitized 8 and 9 Data correlation Involves qualitative data being correlated with quantitized data or quantitative data being correlated with qualitized data. 9 Data consolidation Both quantitative and qualitative data are combined to create new or consolidated codes, variables, or data sets. 8 and 9 Data comparison Involves comparing the findings from the qualitative and quantitative data sources or analysis 8 and 9 Data integration Both qualitative and quantitative findings are integrated into either a coherent whole or two separate sets (i.e., qualitative and quantitative) of coherent wholes 1-9 *Definitions quoted from Onwuegbuzie et al. (2007, pp. 15-16).

PAGE 115

104 CHAPTER 4 RESULTS Introduction Chapter 4 presents the results from this study by research question. This chapter comprises nine sections, one for each research question. Results of the data analysis are presented in descriptive text, tables, and/or figures. This chapter concludes with a summary of findings. Two forms of the 12-item BFNE-S were utilized, the BFNE-S, General and the BFNE-S, Context-specific. Respondent s were first asked to complete the BFNE-S, General, when thinking about situat ions in general outside the church setting. Respondents were next asked to complete the BFNE-S, Contextspecific, which consisted of the same 12 items, when thinking about situations inside the church. Comparisons are made between responses to the two different versions of the BFNE-S. The church setting was believed to hold less potential for fear of negative evaluati on. For all analyses, missing data were minimal. Appendix E contains a table indicating the number of missing data for each analysis that wa s conducted. Participants. Participants were 239 members, regular attenders, and visitors of Church A. As depicted in Table 21, a large percentage of survey respondents were older (51% were 50 or more years of age, with 77% being 40 years of age or more) and were long-ti me members of Church A (46% had been members for five or more years).

PAGE 116

105 Table 21 Demographics of Respondents to Personal Concerns and Issues Survey Gender Male Female 73 (32%) 157 (68%) Age 21-30 31-40 41-50 51-60 61-70 70 + 23 (10%) 31 (13%) 60 (26%) 66 (28%) 37 (16%) 17(7%) Race/ ethnicity Caucasian African American Hispanic Multiracial 144 (64%) 48 (21%) 26 (11%) 10 (4%) Length of Membership at Church A Not a member Less than 1 year 2 to 3 years 4 to 5 years 6 to 10 years 11 to 20 years Over 20 years 61 (27%) 20 (9%) 23 (10%) 19 (8%) 41 (18%) 35 (15%) 39 (13%) Length of Membership at Any Church Not a member of any church Less than 1 year 2 to 3 years 4 to 5 years 6 to 10 years 11 to 20 years Over 20 years 15 (6%) 6 (3%) 8 (3%) 12 (5%) 25 (11%) 50 (21%) 119 (51%) Number of Worship Services Attended Per Month 1 to 2 3 to 4 5 to 8 9 to 12 20 (9%) 92 (40%) 69 (30%) 47 (21%) Number of Activities Participated in Per Month 0 1 to 2 3 to 4 5 to 10 68 (29%) 95 (41%) 40 (17%) 30 (13%) Number of Close Friends at Church A 0 1 to 2 3 to 4 5 to 10 10+ 46 (19%) 55 (24%) 46 (20%) 36 (16%) 50 (21%) Number of Acquaintances at Church A 0 1 to 2 3 to 4 5 to 10 10+ 11 (5%) 19 (8%) 39 (17%) 44 (18%) 120 (52%)

PAGE 117

106 Research Question 1 What are the psychometric properties of the BFNE-S, General and Context-specific, in the church setting? Research Hypothesis 1 The BFNE-S, General and Context-specific, displays psychometric properties in the sample that are si milar to those demon strated for other populations taken from university or clinical settings. To answer Research Question 1, de scriptive statistics and Cronbach alphas were calculated and compared wit h results found in other studies. SPSS Version 16.0 (SPSS, Inc., Chicago, IL) was utilized for all analyses in the current study except the confirmatory factor analysis portion of Research Question 1. Table 22 presents descriptive statis tics for the items on the BFNE-S, General and the BFNE-S, C ontext-specific, as well as values for skewness and kurtosis. Data screening revealed the scores for both instruments were approximately normally distri buted. Inspection of t he skewness and kurtosis values revealed that for the BFNE-S, General, one value each for skewness and kurtosis was greater than the absolute value of 1. For the BFNE-S, Context-specific, se ven skewness values were greater than the absolute value of 1; however, si x of the skewness values greater than the absolute value of 1 were 1.33 or less; the largest skewness value was 1.61. Only three kurtosis values were great er than the absolute value of 1 (-1.219, 1.103, and 2.00).

PAGE 118

107 Table 22 Descriptive Statistics for BFNE-S, G eneral and BFNE-S, Context-specific BFNE-S, General BFNE-S, Context-specific Item N M SD SkewnessKurtosis N M SD SkewnessKurtosis I worry about what othe r people will think of me even when I know it does not make any difference. 2382.371.160.40-0.732392.271.230.70-0.51 I am concerned if I know people are forming an unfavorable impression of me. 2382.861.300.08-1.132392.801.370.16-1.22 I am frequently afraid of other people noticing my shortcomings. 2391.991.091.000.192391.851.121.250.67 I often worry about what kind of impression I am making on someone. 2392.501.220.41-0.862392.371.300.59-0.81 I am afraid others will not approve of me 2391.941.070.90-0.152391.861.121.170.44 I am afraid that people will find fault with me. 2361.971.040.83-0.102371.831.091.331.10 Other people’s opinions of me bother me. 2392.181.080.71-0.252392.011.131.020.26 When I am talking to someone, I worry about what they may be thinking about me. 2381.820.991.160.782391.691.021.611.99 I am usually worried about what kind of impression I make. 2392.151.070.800.132392.031.101.070.54 If I know someone is judging me, it has a big effect on me. 2392.351.240.60-0.582392.231.290.75-0.52 Sometimes I think I am too concerned with what other people think of me. 2372.131.260.90-0.312392.001.321.140.04 I often worry that I will say or do the wrong things. 2362.301.170.64-0.432372.151.230.94-0.11 Note: The minimum for all items was 1 ( Not at all like me ) and the maximum was 5 ( Extremely like me ).

PAGE 119

108 For both versions of the BFNE-S, sco res from the 12 items were summed to produce a score, which could range from 12 to 60. Listwise deletion was utilized for missing values. The BF NE-S, General, mean and standard deviation (26.50 and 10.39, respectively) and the BFNE-S, Context-specific mean and standard deviation (25.22 and 11.09, respec tively) were similar to values obtained in previous research. The means in the current study were slightly lower than in previous research, and t he standard deviations were larger than all but one of the previous studies. Review of four previous studies, as displayed in Table 25, revealed that the means in previous studies r anged from 26.81 (for a non-anxious community sample) to 51.50 (for a social phobic sample). Standard deviations ranged from 4.78 (f or a non-socially anxious community sample) to 12.50 (for a panic disorder sample). Table 23 Norms Reported for BFNE Scale (ordered by mean) Study Author(s) N Sample Description M SD Cronbach Alpha Watson, 2009 232 Members, regular attenders, and visitors in a large evangelical church – BFNE-S, Context-specific 25.22 11.09 .94 Watson, 2009 226 Members, regular attenders, and visitors in a large evangelical church – BFNE-S, General 26.50 10.39 .93 Weeks et al., 2005a 1,385 Non socially anxious 26.81 4.78 .90 Collins et al. 2005b 30 Community sample (non-anxious) 29.20 8.20 .97

PAGE 120

109 Rodebaugh, et al. 1,049 Archival data (anxious and nonanxious) 29.41 7.72 Not reported Duke et al., 2006 355 Individuals in a shopping mall 32.30 7.34 .94 Collins et al., 2005b 99 Individuals with panic disorder 39.80 12.50 .97 Weeks et al., 2005a 138-165 (missing data varied) Individuals with social anxiety 46.91 9.27 .92 Collins et al., 2005b 82 Individuals with social phobia 51.50 7.30 .97 a The Weeks et al. (2005) study consisted of two samples. b The Collins et al. (2005) study consisted of three samples. Inter-item reliability was assessed with a subsample ( n = 107). Reliability Internal consistency reliability of the BFNE-S, General, and BFNE-Context-specific scores was assessed for the sample of 239 participants. Cronbach’s alpha coefficient revealed that scores for both scales had exceptional internal consistency (Nunnally, 1994). Cronbach’s alpha for the BFNE-S, General, scores was .93 (95% confidence inte rval [CI] = .92, .95), with a value of .94 (95% CI = .93, 95) for the BFNE-S, Context-specific scores. These values are consistent with resu lts of previous studies. The itemtotal correlation for the BFNE -S, General, ranged from .64 to .79, and from .64 to .80 for the BFNE-S, Contextspecific. The lowest item-total correlation for both versions of the scale was for Item 4 (“I often worry about what kind of impression I am making on someone”).

PAGE 121

110 Confirmatory Factor Analysis This section, addressing Research Questi on 1, consists of results for four different confirmatory fact or analyses (CFAs). Oneand two-factor CFA models were conducted for both versions of the BFNE-S (the General and the Contextspecific versions). The two-factor model was investigated because some researchers (Duke et al., 2006; Rodebaugh et al., 2004; Weeks et al., 2005) had utilized a two-factor model of the BFNE, consisting of four items that were originally worded negatively. Result s for the one-factor model, General and Context-specific, are present ed, followed by results fo r the two-factor model, General and Context-specific. Model Specification, Input Data, and Model Estimation. Figure 3 presents the one-factor model utilized for the BFNE-S General and Context-specific. The model was identified by fixing the first fact or loading (i.e., patte rn coefficient) to 1.0 (Brown, 2006).

PAGE 122

111 Figure 3 One-factor BFNE-S Shyness X2 – unfavorable impression X3 – noticing my shortcomings X4 – worry about kind of impression X5 – afraid others will not approve X6 – afraid others will find fault X7 – others’ opinions bother me X8 – when talking, I worry . X9 – usually worried about impression X10 – judging has a big effect on me X11 – too concerned with others’ thoughts X12 – say or do wrong things X1 – worry what others think

PAGE 123

112 A CFA was conducted to evaluate t he unidimensional model of shyness reported in the literature (Crozier, 2001). The BFNE -S, General and Contextspecific, was administered to 239 chur ch members, regul ar attenders, and visitors. The CFA was performed with the Mplus program 4.1 (L. K. Muthen & B. O. Muthen, 2001). Listwise del etion was utilized for miss ing data, resulting in a final sample of 199, which was adequate for the analysis. Hatcher (1994, p. 260) states that for confirmatory factor anal yses “. . a minimally acceptable number of observations would be the larger of 150 or 5 observat ions per parameter to be estimated.” Descriptive statistics were computed for the 12 observed variables of both versions of the BFNE-S (General and C ontext-specific). As explained in the Results for Question 1 section, item scores from the BFNE-S, General and Context-specific, were examined and found to be approximately normally distributed. For both versions of the BFNE-S, a covariance matrix of 12 observed variables was analyzed. The esti mation method employed was maximum likelihood (Brown, 2006) One-factor model, fit indices. Table 24 presents fit indices for the BFNES, General and Context-specific versions The chi square values indicated a statistically significant amount of misf it; specifically, the BFNE-S, General, yielded a statistically signifi cant chi-square statistic, 2(54, N = 199) = 148.83, p < .001, as did the BFNE-S, Context-specific, 2(54, N = 199) = 183.87, p < .001. It should be noted, however, that the chi-square statistic is sensitive to sample size (Bollen, 1989). Distefano and Hess (2005) s uggests the following cut-offs for

PAGE 124

113 Table 24 Fit Indices for BFNEa-S, Generaland Context-specif ic, Oneand Two-Factor Modelsb One-Factor Model Two-Factor Model d Instrument BFNE-S General BFNE-S Contextspecific BFNE-SGeneral BFNE-S Contextspecific Fit Indices Cutoffc 2 148.83 183.87 143.42 179.19 df 54 54 53 53 CFI >.90 .93 .92 .93 .92 RMSEA <.08 .09 .11 .09 .11 SRMR <.08 .05 .05 .05 .05 TLI >.95 .91 .90 .92 .90 a Brief Fear of Negative Evaluation-Straightforward b n = 199 c Distefano (2005). d The second factor was composed of Items 2, 4, 7, and 10. acceptable alternative fit indices in ev aluating CFAs: CFI > .90; RMSEA < .08; SRMR < .08; and TLI > .95. As present ed in Table 26, results are somewhat mixed for the one-factor model. The CF I and SRMR indicated acceptable levels of fit, whereas the other fit indice s (RMSEA and TLI) indicated less-thanacceptable levels of fit. The fit is almost the same, with very s lightly better fit for the BFNE-S, General. Parameter estimates, one-factor model. Table 27 presents standardized and unstandardized pattern coeffi cients. All of the obtained t values for the standardized factor coefficients for the one-factor model were statistically significant ( p < .001), with t values greater than 3.192 (Hatcher, 1994). Standardized factor coefficients ranged from .608 to .834 for the BFNE-S,

PAGE 125

114 General and from .615 to 827 for the BFNE-S, Context-specific, indicating that all coefficients were moderatel y large (Hatcher, 1994). The highest standardized factor pattern c oefficient for both versions of the BFNES was .83 for Item 11 (“Sometimes I think I am too concerned with what other people think of me”). The lowest st andardized factor pattern coefficient for the BFNE-S, General was Item 2 (“I am concerned if I know people are forming an unfavorable impression of me”), wher eas the lowest standardized factor pattern coefficient for the BFNE-S, Context-specific, wa s .62 for Item 4 (“I often worry about what kind of impre ssion I am making on someone”). Unstandardized factor pattern coeffi cients for the BFNE-S, General, ranged from 0.86 to 1.48 and for the BFNE -S, Context-specific, from 0.96 to 1.37. For both versions of the BFNE-S, the highest unstandardized factor pattern coefficient was for Item 11 (“Sometimes I think I am too concerned with what other people think of me”) and the lowest unstandardized pattern coefficient was for Item 8 (“When I am talking with so meone, I worry about what they may be thinking about me”). Residual variance estimates ranged from 0.37 to 0.70 for the BFNE-S, General and from 0.38 to 0. 68 for the BFNE-S, Context-specific. This suggests that the indicators were reliable indicato rs of the construct of shyness (Brown, 2006). All residual variance estimates were statistically significantly different from zero. To review unstandardized estimates, please refer to Table 25.

PAGE 126

115 Table 25 Model Results for the One-Factor BFNE-S, General, and BFNE-S, Contextspecific a BFNE-S, General BFNE-S, Contextspecific Unstandardize Pattern coefficients Standard Error Standardized Pattern coefficients Unstandardized Pattern Coefficients Standard Error Standardized Pattern coefficients Item 1. I worry about what other people will think of me even when I know it does not make any difference. 1.00 0.00 0.62 1.00 0.00 0.66 2. I am concerned if I know people are forming an unfavorable impression of me. 1.09 0.15 0.61 1.08 0.13 0.65 3. I am frequently afraid of other people noticing my shortcomings. 1.04 0.13 0.71 0.99 0.11 0.71 4. I often worry about what kind of impression I am making on someone. 1.03 0.14 0.63 0.99 0.13 0.62 5. I am afraid others will not approve of me. 1.15 0.13 0.79 1.06 0.11 0.80 6. I am afraid that people will find fault with me. 1.12 0.13 0.78 1.11 0.11 0.82 7. Other people’s opinions of me bother me. 1.12 0.13 0.76 1.08 0.11 0.80 8. When I am talking to someone, I worry about what they may be thinking about me. 0.86 0.11 0.65 0.96 0.10 0.76 9. I am usually worried about what kind of impression I make. 1.02 0.12 0.71 1.03 0.11 0.79 10. If I know someone is judging me, it has a big effect on me. 1.15 0.14 0.69 1.09 0.12 0.70 11. Sometimes I think I am too concerned with what other people think of me. 1.48 0.16 0.83 1.37 0.13 0.83 12. I often worry that I will say or do the wrong things. 1.29 0.14 0.79 1.18 0.12 0.76 a All estimates were statistically significantly different from zero. Note: For the first item, the pattern coe fficient was constrained to zero for model identification purposes. Modification Indices for One-factor Model. After overall goodness of fit was evaluated, modification indices were examined. In a onefactor model, the only potential source of misfit involves correlations between pairs of error terms. The following portion of the discussion review s sources of misfit for the BFNE-S, General, followed by sources of misfit for the BFNE-S, Context-specific. Next,

PAGE 127

116 similarities and differences between sources of misfit for the tw o versions of the one-factor model are highlighted. Modification indices for the one-fact or model of the BFNE-S, General, revealed a few localized areas of misfit in the model, with 11 indices greater than 3.84, as depicted in Table 26. The critical value of 3.84 was chosen because it is a statistically significant source of mi sfit for one degree of freedom at the .05 level. Only two modification indices were greater than 10. Re view of the four items involved (Items 1, 2, 4, and 8) rev eals that for the item pair 1 and 2, Item 2 was originally worded negatively. Items 4 and 8 both have the word “worry” in common, which may have caused the erro rs associated with these items to be correlated. Modification indices for the one-fact or model of the BFNE-S, Contextspecific, revealed 17 indices greater t han 3.84 and 4 indices greater than 10. Review of the items involved in modifica tion indices greater than 10 revealed two of the same item pairs as for the BF NE-S, General (Items 1 and 2 and Items 4 and 8) as well as Items 5 and 6 and Items 4 and 9. Items 5 and 6 share the word “afraid” and Items 4 and 9 share the word “worry.” As stated earlier, these similarities in wording may have caused the correlated error for these item pairs. Regarding similarities and differences between the two versions, the largest modification index for both versi ons was for the correlation between the error variance associated with Items 1 and 2. The second largest was for Items 4 and 8. More localized areas of misf it were found for th e BFNE-S, Contextspecific, than for the BFNE-S, General, four for the former and tw o for the latter.

PAGE 128

117 Additional item pairs showing misfit were Items 5 and 6 and Items 4 and 9. Areas of misfit were similar across the tw o version of the BFNE-S. Table 28 lists the five highest modification indices for the one-factor BFNE-S, General and Context-specific models. Table 26 Five Highest Modification Indices O ne-factor BFNE-S, General and Contextspecific Versions Item Numbers and Text of Item Pairs Modification Index Modification Index BFNE-S, General BFNE-S, Contextspecific 1. I worry about what ot her people will think of me even when I know it does not make any difference. 2. I am concerned if I k now people are forming an unfavorable impression of me. 22.02 24.23 4. I often worry about what kind of impression I am making on someone. 8. When I am talking to someone, I worry about what they may be thinking about me. 17.26 13.91 2. I am concerned if I know people are forming an unfavorable impression of me. 10. If I know someone is judging me, it has a big effect on me. 11.72 13.08 2. I am concerned if I know people are forming an unfavorable impression of me. 5. I am afraid others will not approve of me. 8.45 12.37 4. I often worry about what kind of impression I am making on someone. 6. I am afraid that peopl e will find fault with me. 6.85 9.27 Two-factor model As stated earlier, this ana lysis includes evaluation of the two-factor model for both versions of the BFNE-S. Other researchers have investigated a two-factor model consis ting of negatively worded items (Duke et al., 2006; Rodebaugh et al., 2004; Weeks et al., 2005). A two-factor CFA was

PAGE 129

118 conducted, with the second factor compos ed of the four item s that had been negatively worded in the original BFNE-S. All the items in both versions of the BFNE-S were positively worded. One patte rn coefficient in each set of factors was set to 1.0 to ident ify the model.

PAGE 130

119 Figure 4 Two-factor BFNE-S Shyness X1 – worry what others think X3 – noticing my shortcomings X5 – afraid others will not approve X6 – afraid others will find fault X7 – others’ opinions bother me X9 – usually worried about impression X11 – too concerned with others ’ thoughts X12 – say or do wrong things X2 – unfavorable impression X4 – worry about kind of impression X7 – others’ opinions bother me X10 – judging has a big effect on me Formerly reverseworded items

PAGE 131

120 Two-factor model, measures of fit. The two-factor model yielded fit indices nearly identical to those of the one-fact or model, as presented in Table 29. The two-factor BFNE-S, General, yielded a statistically signif icant chi-square statistic, 2 (53, N = 199) = 143.42, p < .001, and the two-factor BFNE-S, Context-specific, also yielded a statistically si gnificant chi-square statistic, 2(53, N = 199) = 179.19, p < .001. The statistically signifi cant chi square indicated a poor fit (Brown, 2006). It should be kept in mi nd that, as mentioned earlier, the chisquare statistic is sensitive to sample si ze (Bollen, 1989). As was the case for both versions of the one-factor model, for both versions of the two-factor model, acceptable levels of fit were found ut ilizing the CFI and the SRMR, and less-thanacceptable levels of fit were found using the RMSEA and the TLI, based upon the criterion selected (Diste fano & Hess, 2005). Parameter estimates, two-factor model Table 29 presents standardized and unstandardized pattern coefficients for the two-factor model. Model parameters were evaluated for the two-fact or model as well. As was the case with the one-factor model all of the obtained t values for the standardized pattern coefficients were statistically significant ( p < .001), with values greater than 3.19 (Hatcher, 1994). For the tw o-factor model, general, standardized factor pattern coefficients ranged from .62 to .84 for Fact or 1 (Shyness) and from .64 to .78 for Factor 2 (Reverse-worded Items). For t he two-factor BFNE-S, Context-specific model, standardized factor pattern coefficien ts ranged from .30 to .61 for Factor 1 and from .39 to .58 for Fa ctor 2. The highest st andardized factor pattern coefficient for the two-factor BFNE-S, G eneral, was .84 for Factor 1 with Item 11

PAGE 132

121 (“Sometimes I think I am too concerned wit h what other people think of me”), and the lowest was for Factor 1 with Item 1 (“I worry about what other people think of me even when I know it does not make any difference”). The highest standardized coefficient for the two-factor BFNE-S, Context-specific, was .87 for Factor 1 with Item 1 (“I worry about what other people think of me even when I know it does not make any difference”) and for Factor 1 with Item 5 (“I am afraid others will not approv e of me”). Unstandardized pattern coefficient estimates suggested that the indicators were reliable indicators of the constr uct of shyness (Brown, 2006, p. 156). Unstandardized factor coefficient estimate s for the two-factor BFNE-S, General, ranged from 0.87 to 1.49 for Factor 1 and from 0.91 to 1.02 for Factor 2. Unstandardized factor pattern coefficient estimates for the tw o-factor BFNE-S, Context-specific, ranged from 0.43 to 0.87 fo r Factor 1 and from 0.45 to 1.02 for Factor 2. The largest unstandardized fact or coefficient for the two-factor BFNES, General, was 1.49 for Factor 1 with Ite m 11 (“Sometimes I think I am too concerned with what other people think of me”) and the smallest was 0.87 for Factor 1 with Item 8 (“W hen I am talking to someone, I worry about what they may be thinking about me”). The larges t unstandardized factor coefficients for the two-factor BFNE-S, Context-specific, was 1.38, for Factor 1 with Item 11, and the smallest was 0.97, for Factor 1 with Item 8. Residual variance estimates ranged fr om 0.87 to 1.49 fo r the two-factor BFNE-S, General and from 0.43 to 1.02 for the BFNE-S, Context-specific, suggesting that the indicators were reliable indicators of the construct of shyness

PAGE 133

122 (Brown, 2006). All residual variance esti mates were statistically significantly different from zero. To review unstandar dized estimates, please refer to Table 29. Table 27 Model Results for the Two-Factor BFNE-S, General, and BFNE-S, Context a BFNE-S, General BFNE-S, Contextspecific Unstandardized Pattern coefficients Standard Error Standardized Pattern coefficients Unstandardized Pattern coefficients Standard Error Standardized Factor loadings Factor 1 by Item: 1. I worry about what other people will think of me even when I know it does not make any difference. 1.00 0.00 0.62 1.00 0.00 0.66 3. I am frequently afraid of other people noticing my shortcomings. 1.04 0.12 0.71 0.93 0.11 0.71 5. I am afraid others will not approve of me. 1.15 0.13 0.79 1.07 0.11 0.80 6. I am afraid that people will find fault with me. 1.13 0.12 0.78 1.12 0.11 0.82 8. When I am talking to someone, I worry about what they may be thinking about me. 0.87 0.11 0.65 0.97 0.10 0.80 4. I often worry about what kind of impression I am making on someone. 1.02 0.12 0.71 1.03 0.11 0.79 11. Sometimes I think I am too concerned with what other people think of me. 1.49 0.16 0.84 1.38 0.14 0.83 12. I often worry that I will say or do the wrong things. 1.29 0.14 0.79 1.18 0.13 0.76 Factor 2 by Item: 2. I am concerned if I know people are forming an unfavorable impression of me. 1.00 0.0000.65 1.00 0.00 0.68 9. I am usually worried about what kind of impression I make. 0.91 0.12 0.64 .89 0.110.62 7. Other people’s opinions of me bother me. 1.00 0.11 0.78 0.99 0.10 0.83 10. If I know someone is judging me, it has a big effect on me. 1.02 0.12 0.71 0.99 0.11 0.71 a All estimates were statistically significantly different from zero.

PAGE 134

123 The correlation between factors for t he two-factor BFNE-S, General was .95 and .96 for the BFNE-S, Context-specif ic. This high correlation between the two factors provides support for the one-fa ctor solution originally presented for the instrument in previous research. The high correlation also meant that the second factor was not contributing mu ch unique information. Nonetheless, modification indices for the two-factor model were also reviewed to identify areas of misfit in the model. Modification Indices fo r two-factor model. With the BFNE-S two-factor model, there are two sources of potentia l misfit: correlated error and secondary pattern coefficients. Modifi cation indices for the two-fa ctor model of the BFNE-S (General) revealed three indices great er than 10 and 16 indices greater than 3.84, as portrayed in Table 30. Of the three indices greater than 10, two involved correlated error and one was a secondary loading. For the BFNE-S, Contextspecific, there were 6 indices greater than 10 and 16 greater than 3.84. Of the six indices greater than 10, 5 involv ed correlated error and one was a secondary loading. The misfit was very simila r across both versions (General and Contextspecific) of the two-factor model BFNE-S. In comparing modifi cation indices for the two-factor model, there were more modification indices greater than 10 for the Context-specific than fo r the General version of the BFNE-S. The largest source of misfit for both versions of t he two-factor model was correlated error. For the BFNE-S, General, the first of the three item pairs involved in modification indices greater than 10 was Ite ms 1 and 2. Items 4 and 9 have the word “worry” in common, which may have caused error variances to correlate.

PAGE 135

124 Another source of misfit was the secondary loading of Item 2 on Factor 1 (modification index greater than 10). Review of modification indices for the two-factor model of the BFNE-S, Context-specific, revealed 17 indices gr eater than 3.84 and 6 gr eater than 10. The largest modification index represented a correlated error involving Items 1 and 2. The second largest modification index was a secondary loading for Item 8 on Factor 2; Factor 2 consisted of all t he items that had originally been worded negatively. Careful consideration discl osed no plausible explanation for this result. Items 4 and 9 had the second largest modification index for both versions in the two-factor general model. Other item pairs having modification indices greater than 10 included Items 4 and 10 (w hich were both worded negatively originally), Items 5 and 6 (which share the word afraid), and Items 1 and 10 (with Item 10 having been worded negatively). Regarding similarities and differences between the two-factor models of the BFNE-S, General, and the BFNE-S, Cont ext-specific, modification indices for the two-factor model of the BFNE-S, G eneral, revealed slightly more localized areas of misfit than in the one-factor model, with 16 indices greater than 3.84 and 3 greater than 10. The largest modifi cation index for the two-factor model Context-specific model was for the correla ted errors involving Items 1 and 2, as was the case with the General model Table 28 lists the five highest modifica tion indices for the two-factor BFNES, General and Context-specific versions.

PAGE 136

125 Table 28 Five Highest Modification Indices Tw o-factor BFNE-S, General and Contextspecific versions Item Numbers and Text of Item Pairs Modification Index BFNE-S, General 1. I worry about what other people wil l think of me even when I know it does not make any difference. 2. I am concerned if I know people are forming an unfavorable impression of me. 25.717 4. I often worry about what ki nd of impression I am making on someone. 8. When I am talking to someone I worry about what they may be thinking about me. 16.578 Factor 1 (Shyness) 2. I am concerned if I know people are forming an unfavorable impression of me. 9.734 2. I am concerned if I know peop le are forming an unfavorable impression of me. 10. If I know someone is judging me, it has a big effect on me. 8.281 Factor 2 (Originally Reversed Items) 8. When I am talking to someone, I worry about what they may be thinking about me. 7.824 BFNE-S, Context-specific 1. I worry about what other people wil l think of me even when I know it does not make any difference. 2. I am concerned if I know people are forming an unfavorable impression of me. 25.867 Factor 2 (Originally Reversed Items) 8. When I am talking to someone I worry about what they may be thinking about me. 16.512 4. I often worry about what kind of impression I am making on someone. 9. I am usually worried about what kind of impression I make. 14.127 4. I often worry about what kind of impression I am making on someone. 10. If I know someone is judging me, it has a big effect on me. 13.278 5. I am afraid others will not approve of me. 6. I am afraid that peopl e will find fault with me. 12.938

PAGE 137

126 Comparisons across all four models. Confirmatory factor analyses of the BFNE-S, General and Context-specific, we re conducted to investigate whether the data suggested utilizing a two-factor model, rather than the one-factor model on which most of previous literature was based. The data revealed nearly identical fit indices for bot h versions of the BFNE-S and highly similar parameter estimates and modification indices. The individual items most often listed as indicating misfit were Items 2, 4, 1, 8, and 10. It is interesting that these are almost the same f our items (Items 2, 4, 7, and 10) that were formerly negatively wo rded. It is possibl e that the reason for this is in how the scale has been dev eloped. The original Fear of Negative Evaluation Scale (which had 22 items in dichotomous, true-false format) was written at a time when only exploratory fa ctor analysis was available; thus, some items might have been redundant. Table 29 presents modification indices for the four CFA models in this study (General and Context-specific versions of the one-factor as well as the twofactor model). As can be seen, the most problematic item pair was Items 1 and 2, which had the largest modification index across all four models. Other item pairs showing misfit were Items 4 and 8, Items 5 and 6, and Items 4 and 9. The number of modification indices greater t han the statistically significant value of 3.84 (with one degree of freedom) is lowest for the one-factor, general version. The higher number of statisti cally significant modificati on indices for the Contextspecific version of the oneand two-fact or models may possibly be due to order

PAGE 138

127 effects. All participants in the st udy completed the BFNE-S, General, immediately followed by the BFNE-S, Context-specific. Table 29 Comparison of Modification Indices fo r the Four Models in This Study One-Factor Model Two-Factor Model General Contextspecific General Contextspecific Largest M.I.a Item Pair 22.024 1,2 24.234 1,2 25.717 1,2 25.867 1,2 M.I.s >3.84 b 11 17 16 17 Number M.I.s>10 2 4 3 6 Item Pairs with M.I.s>10 1,2 1,2 1,2 1,2 4,8 4,8 4,9 4, 9 -5,6 -4, 10 -4,9 -5, 6 ---4,8 ---1,10 Items Loading on More Than One Factorc --Factor 1, Item 2 Factor 2, Item 8 a Modification Index b Statistically significant, with 1 degree of freedom, at the .05 level c Not relevant for one-factor model. To the extent possible, comparisons were made across all four models evaluated in this study. It is essentia l to bear in mind that with a one-factor model, the only source of mi sfit is correlated error, whereas a two-factor model had two sources of misfit: secondary as we ll as correlated error. Consistent with parsimony, the one-factor model was utilized in the current study. Based on descriptive statistics, Cronbach’s alpha, and a confirmatory factor analysis, BFNE-S, General and Cont ext-specific, displayed psychometric properties that are similar but not identical to those in previous research. The means and standard deviations of both vers ions of the BFNE-S were generally

PAGE 139

128 similar to values obtained in previous re search with similar samples; however, the means for the BFNE-S, General and the BFNE-S, Context-specific were lower than those in previous research with similar samples, and the standard deviations for both versions of the BFNE-S were larger than the standard deviations that had been found in previous research. Internal consistency reliability, as measured by Cronbach’s al pha, was consistent with previous studies, and confirmatory factor analysi s also revealed findings similar to previous studies. Research Question 2 What is the difference in perceived fear of negative evaluation in the church setting compar ed to the non church setting? A paired sample t test was conducted to answer this question because each participant provided two scores (Hatcher & Stepanski, 1994). Research Hypothesis 2 Perceived fear of negat ive evaluation is lower in the church setting compared to the non church setting. Listwise deletion was utilized for missi ng data. Review of box plots for scores from the BFNE-S, General and Contex t-specific revealed two outliers for the BFNE-S, Context-specific. A score wa s considered an outlier if it was farther than 1.5 times the interquartile range away from the median. Removal of the two outliers did not affect statistical signi ficance and will not be discussed further. The outliers were included in the data utilized in all analyses. The paired-samples t test revealed a statistically significant difference between mean levels of fear of negative evaluation in the general setting and in the Context-specific setting, t (220) = 4.03; p < .001. The mean score on the

PAGE 140

129 BFNE-S, General, was 26.50 ( SD = 10.39) and the mean for the BFNE-S, Context-specific, was 25.22 ( SD = 11.09). The effect si ze, utilizing the Dunlap, Cortina, Vaslow, and Burke (1996) formula for a paired t test, was negligible (0.01). Research Question 3 What is the effect of gender on perceived fear of negative evaluation for males and females in the church setting compared to the non church setting? Research Hypothesis 3 The difference in perceived fear of negative evaluation in the church setting compared to the non church setting is the same for males and females. Levene’s test for gender in the general setting and in the Context-specific setting indicated no evidence of heter ogeneity of variances. A repeated measures ANOVA was conducted to test the effect of gender on self-reported fear of negative evaluation as measured by the BFNE-S, General. A statistically significant within-subjects main effect for setting was found, F (1, 212) = 13.87, p < .01. The effect si ze of .06, utilizing 2, was negligible. No statistically significant within-subjects interaction effect was observed between context and gender, F(1, 212) = .01, p > .05. No statistically significant between-subjects main effect was observed for gender, F(1, 212) = .02, p > .05. Research Question 4 What is the difference in perceived fear of negative evaluation in the church setting co mpared to the non church setting for individuals of different races?

PAGE 141

130 Research Hypothesis 4 The difference in FNE between the church and non-church setting is the sa me for different races. Levene’s test for both comparisons i ndicated no evidence of heterogeneity of variances. A repeated measures ANOVA was conducted to te st the effect of race on self-reported fear of negative evaluation as measured by the BFNE-S, General. A statistically significant main effect for setting was found, F (1, 208) = 6.40, p < .05. No statistically significant within-subjects interaction effect between setting and race was observed, F (3, 208) = 0.840, p > .05. No statistically significant between-subjec ts main effect for race was found, F (3, 208) = 2.16, p > .05. Research Question 5 What is the difference in levels of perceived acceptance between people inside and outside the church setting? Research Hypothesis 5. Levels of per ceived acceptance by people in the church setting are higher than the levels of percei ved acceptance by people outside the church setting. For five items on the perceived acc eptance checklist, respondents were asked to indicate their levels of acceptance when thinking about situations outside the church and inside the church as well. Respondents were also asked to indicate perceived level of acceptance wh ile at their place of work or business and when with their family at home. For both versions of the Perceived Acceptance Checklist, scores from the five items were summed to produce a score, which could range from 5 to 25. Listwise deletion was utilized for missing values. Cronbach’s alpha for the Perc eived Acceptance-General checklist

PAGE 142

131 scores was less than optimal at .67 (95% CI = .60, .74), with a value of .73 (95% CI = .67, .78) for the Perceived Acc eptance checklist, Context-specific scores. Table 30 presents descriptive statis tics for the individual items on the perceived acceptance checklists, as well as two items (perceived acceptance at one’s work or place of business and with one’s family) for which only one rating was obtained. The Perceived A cceptance Checklist, General, mean and standard deviation (8.73 and 2.48, respectively) were very similar to the Perceived Acceptance Checklist, Cont ext-specific, mean and standard deviation (8.66 and 2.71 respectively). Both distribut ions of scores were relatively normal, with skewness and kurtosis values less than the absolute value of 1, except that the Perceived Acceptance Checklist, Cont ext-specific had a kurtosis value of 1.94, indicating a slight l eptokurtic distribution. Because of the markedly kurtotic di stributions for two items, perceived acceptance by God in the general setti ng and in the Context-specific setting (5.34 and 6.96, respectively), a Wilcox on signed ranks test was conducted to investigate the difference bet ween reported levels of perceived acceptance. The Wilcoxon signed ranks test failed to reveal a statistically significant difference between mean levels of perceived acc eptance in the general setting (Mdn = 7.20) and in the Context-s pecific setting (Mdn = 7.20), Z = -1.111, p > .05. The median and standard deviation for perceiv ed acceptance at work of place of business and with one’s family were si milar to medians and standard deviations for the other items on the checklist; however, perceived acceptance with one’s family was positively skewed (2.62) and markedly leptokurtic (7.39).

PAGE 143

132 Table 30 Descriptive Statistics for Items on the Perceived Acceptance Checklist Perceived Acceptance, General Perceived Acceptance, Context-specific N MinMax M SD Skewness Kurtosis N Min Max M SD Skewness Kurtosis By people in general. 237141.810. 630.581.23 236141.810.710.861.20 By friends and acquaintances. 236131. 470.540.51-0.93 237151.540.651.272.90 When you meet someone you do not know. 235142.130.730.13-0.40 236142.030.750.24-0.48 By yourself. 237151.800.871. 171.48 236151.780.901.331.89 By God. 236151.500.912. 2305.34 235151.450.862.526.96 Non-comparative Items At your place of work or business. 234151. 750.861.432.63 -------With your family at home. 235151.420. 862.627.39 -------Note: Scale ranged from 1 (Very A ccepted) to 5 (Very Unaccepted)

PAGE 144

133 Research Question 6 What is the difference in self-reported levels of comfort outside the church setting and in side the church setting? A paired samples t test was conducted to answer Research Question 6 because each participant provided two scores (Hatcher & Stepanski, 1994). Research Hypothesis 6 Levels of self-reported comfort for people in the church setting are higher t han the levels of comfort fo r people outside the church setting. Scores from the 10 items for each scale were summed to produce a score, which could range from 10 to 50. Listwise deletion was utilized for missing values. Internal consistency reliabili ty of the Comfort Scale, General, and Context-specific scores was assessed for the sample of 224 and 221 participants, respectively. Cronbach’s alph a coefficient for both versions of the Comfort Scale revealed that scores from both versions of the Comfort Scale had exceptional internal consis tency. Cronbach’s alpha for scores pertaining to the Comfort Scale, General, as well as to t he Context-specific, was .86 (95% CI = .83, .88). The item-total correlation for the Comfort Sc ale, General, ranged from .51 to .68, and for the Comf ort Scale, Context-specific from .49 to .67. The lowest item-total correlation for the Co mfort Scale, General was .5 for Item 3 (comfort with small groups) and for the Co mfort Scale, Context-specific, the lowest item-total correlation was .49 for Item 8 (comfort with being assertive). The highest item-total correlation for t he Comfort Scale, G eneral, was .67 for Item 5 (comfort in social situations) and the highest item-total correlation for the Comfort Scale, Context-specific, was .67 fo r Item 5 (comfort in social situations).

PAGE 145

134 Table 31 Descriptive Statistics for the Comfort Sc ale, General and C ontext-specific Comfort Scale, General Comf ort Scale, Context-specific Item Level of comfort with: N M SD Skewness Kurtosis N M SD SkewnessKurtosis 1. Being the focus of attention 2332.84 1.24 0.18 -1.00 2352.88 1.24 0.13 -1.02 2. Large groups 2362.67 1.26 0.32 -1.01 2362.56 1.25 0.44 -.84 3. Small groups 2352.29 1.08 0.57 -.52 2342.18 1.10 0.86 .077 4. Authority figures by virtue of knowledge (intellectual superiors, experts) or by virtue of role (police, teachers, superiors at work) 2382.19 .94 0.45 -.49 2382.04 0.94 0.76 0.22 5. Social situations in general 2372.30 1.00 0.64 .07 2382.13 0.99 0.82 0.37 6. New interpersonal situations in general 2332.39 1.01 0.47 -.39 2342.26 0.97 0.60 -.22 7. Strangers 2372.64 1.07 0.34 -.60 2372.48 1.02 0.57 -.17 8. Situations where assertiveness is required (e.g., when complaining about faulty service in a restaurant) 2372.70 1.17 0.35 -.81 2372.65 1.16 0.41 -.75 9. Being evaluated or compared with others (e.g., when being interviewed, when being criticized) 2382.94 1.17 0.08 -1.13 2362.90 1.22 0.13 -1.00 10, An opposite sex group or a member of the opposite sex 2352.73 1.11 0.30 -.76 2372.56 1.09 0.51 -.41 221 224 Note: The minimum for all items was 1 (Very Comfor table) and the maximum was 5 (Very Uncomfortable).

PAGE 146

135 Table 31 displays descriptive statisti cs for the Comfort Scale, General, and the Comfort Scale, Context-specific as well as values for skewness and kurtosis. Data screening revealed the scores for both instruments were approximately normally distri buted. Inspection of t he skewness and kurtosis values revealed that for the Comfort Sc ale, General, all skewness values were less than the absolute value of 1 and two va lues for kurtosis were greater than the absolute value of 1 (1.01 and 1.13). Fo r the Comfort Scale, Context-specific, all skewness values were less than the abs olute value of 1, and one kurtosis value was greater than the abs olute value of 1 (-1.02). The distribution of scores for the Co mfort Scale, General, as well as Context-specific, were approximately no rmal, with skewness and kurtosis values all being less than the absolute value of 1. Responses to the 10 level of comfort questions were subjected to a paired samples t test. The paired-samples t test revealed a statistically significant diffe rence between mean levels of comfort in the general setting and in the Context-specific setting, t (219) = 1.37, p > .05. The mean score on the Comfort Sc ale, General, was 23.31 ( SD = 6.76) and the mean for the Comfort Scale, C ontext-specific, was 22.64 ( SD = 6.70). The effect size, utilizing the Dunlap et al.’s (1996) formula for a paired t test, was negligible (0.09). Research Question 7 To what extent do context-specific issues influence self-reported levels of fear of negative evaluation in the church setting?

PAGE 147

136 Research Hypothesis 7 Greater depth and breadt h of involvement in church activities are associated with reduced self-report ed fear of negative evaluation. Section B of the Personal Concer ns and Issues Survey included six questions about various aspects of t he respondents’ relationship with other church members and regular attenders. After careful consideration, it was decided to utilize responses to four of the six questions. The four questions utilized in the current study were: length of membership at C hurch A, regularity of attendance, number of activities par ticipated in per month, and number of close friends at Church A. The decision was made not to use responses to the remaining two questions (length of membership at any church and approximate number of acquaintances) to form co mposite variables. There seemed to be substantive differences in what the variables were believed to be measuring. Length of membership at Church A was chosen ra ther than membership at any church because it was less skewed than was length of membership at any church (72% of the respondents had been members of so me church for 10 or more years, whereas only 28% had been members at Church A for the same length of time) and it seemed to measure the characteristi cs of the person most relevant to the environment under investigation. Approximat e number of close friends at Church A was utilized because it can reasonably be assumed that this would be a more valid indicator of how often the respondent had meaningful interpersonal contact than would the approximate num ber of acquaintances. It was believed that the

PAGE 148

137 regularity of attendance and t he number of activities participated in per month would give a measure of the depth of affiliation. A multiple regression was utilized to answer Research Question 7. Table 32 contains means, standard deviations, and Pearson correlations. The bivariate correlations, utilizing pair-wise deletion, revealed that only one predictor variable had a statistically significant relationship with fear of negative evaluation in the church setting: friends ( r = -.14, p < .05). This small effe ct size indicates that individuals with lower levels of fear of negative evaluation repor ted having slightly more friends at Church A than do individuals with higher levels of fear of negative evaluation. The correlations between the remaining context variables and BFNES, Context-specific scores [membership at Church 1 ( r = -.07, p > .05), attendance ( r = .03, p > .05), and activities ( r = .02, p > .05)] were not statistically significant. Utilizing multiple regression, BFNE -S, Context-specific scores were regressed on the linear combination of membership, attendance, friends, and activities. The equation containing these three variables accounted for 1.6% of the variance in BFNE-S, Context-specific scores, F (4, 203) = .802, p > .05, R2 = .016. Beta weights, or standardized mult iple regression coefficients, were reviewed to ascertain the relative import ance of the four c ontext variables in predicting scores on the BFNE-S, Contextspecific. Table 33 indicates that none of the context variables had statistically si gnificant beta weights. The activities

PAGE 149

138 and friends variables might have multicolinearity because the structure coefficients were large and the beta weights were small. Table 32 Descriptive Statistics for Context Variab les (Membership, Attendance, Activities and BFNE-S, Context-specific Scores) Item M SD N Length of Membership 3.80 2.19 229 Regularity of Attendance 3.65 0.92 229 Number Activities Participated in per Month 2.16 1.02 234 Number of Close Fr iends 2.96 1.43 234 Note. Context variables were on an ordinal scale. Table 33 Standardized Multiple Regression C oefficients for Context Variables (Membership, Attendance, Activities, and Friends) Item Beta Structure Coefficients t Length of Membership -.03 .40 -.35 Regularity of Attendance .07 .61 .87 Number Activities Participated in per Month .02 .15 .23 Number of Close Friends -.11 -.24 -1.33 Research Question 8 How do shy people typically think, feel, and behave in an environment hypothesized to have less fear of negative evaluation and selffocus?

PAGE 150

139 Research Hypothesis 8 At least 75% of indivi duals with high levels of FNE will report thoughts, feelings, and behaviors related to six church situations that are consistent with the Clark and We lls (1995) model and that will be at least 10% more than those with low levels of FNE. It should be noted that BFNE-S, Contextspecific results were divided into three levels of shyness for Research Question 7, based on the distribution of scores in the sample. For Research Questions 8 and 9, the BFNE-S, Contextspecific results were divided into thr ee levels of shyness, based upon norms reported in research involving community samples in a previous study utilizing the general population (Duke et al., 2006), as depicted in Table 34, and consideration of the distribution of scores. Table 34 Shyness Levels for Research Questions 8 and 9 Shyness Level BFNE-S, Context-specific, Score Minimal-to-Low 12 to 28.62 Low 28.631 to 35.96 High 35.97 to 48.00 The responses to the write-in qu estions in the Personal Concerns and Issues Survey were entered into an Ex cel spreadsheet, to facilitate coding. Based on the Clark and Wells (1995) m odel of social phobia, the author constructed a coding table (see Table 35). The author explained and discussed the Clark and Wells model of social phobia in detail with a second coder, a

PAGE 151

140 colleague. Additionally, the second c oder read selected portions of research regarding shyness and the Clark and Wells (1995) model of social phobia. Discussion included examples of behavio rs, thoughts, and feelings that were consistent and inconsistent with the model Thoughts that were consistent with the model were, for example, “I wish I wasn’t so uncomfortable to go up to someone and start a conversation, ”which is avoidance (a safety behavior) as is “When can I leave to go home?” A comment inconsistent with the model was, for example, when one respondent reported th inking “How can I make this a moment filled with purpose?” wh ile meeting with a friend. Table 35 Coding Guidelines Based on the Clark and We lls (1995) Model of Social Phobia Focus of Analysis Clark and Wells Model Component Examples Behaviors Safety Behaviors Avoiding situations. Avoiding initiating interpersonal contact. Minimizing the stress of interpersonal contact by averting eyes, speaking in short sentences, etc. Thoughts Excessively high standards for social performance. “I must not show any sign of weakness” and “I should only speak when other people pause” Conditional beliefs concerning consequences. “If I am quiet, people will think I am boring'” and “If people get to know me, they will not like me.” Unconditional negative beliefs about the self, “I am odd (or different)” and “I am unlikable” Feelings Anxiety and other emotional distress, such as fear or worry. Anxious. Insecure. Intimidated. Comments irrelevant to the model “No specific thoughts”, “That’s a good point. I never looked at it that way” or “Listen closely; I only have hearing in one ear”

PAGE 152

141 Working together, the author and sec ond coder coded a sample of five sets of comments regarding whether the comment was consist ent with the Clark and Wells (1995) model. (Each set of comments consisted of one participant’s responses describing his/her thoughts, f eelings, and behaviors in the six church situations, i.e., each “set” of comment s consisted of 18 responses from one participant.) Working independently, the author and second coder then coded another sample of five sets of comment s. The percent of agreement between the author and the second coder was 89%. After discussion, the percent of agreement was 93%, which was considered a satisfactory rate of agreement. These 10 responses were withdrawn fr om the data set and were not utilized subsequently, except as training materi al for coding conducted for Research Question 9. The remaining responses to the writ e-in items were ordered by BFNE-S, Context-specific scores, and five sets of comments were selected from minimalto-low, medium, and high levels of shy ness. Items having short, medium and long responses, based on visual scan, we re selected. The order of comments was then randomized with respect to BFNE -S, Context–specific score, and the score was removed from the file used for coding so that both coders were unaware of the self-reported shyness leve l of the person whose comments they were coding. The total number of comments analyzed was 270; each of 15 participants had 18 comments (thoughts, fee lings, and behaviors in six situations). The total

PAGE 153

142 word count for all comments was 4,493. Utilizing comment as the unit of analysis, the author and the second coder coded the 270 write-in comments. Comments that were irrelevant to t he Clark and Wells model and comments that were ambiguous were marked as uncodable. The initial rate of agreement was 87%. After discussion, the rate of agr eement was 92%. The comments marked as uncodable were removed from subs equent analysis. Table 36 presents the proportion of comments that were marked as uncodable and as codable with respect to theory. Only comments c odable with respect to the theory were included in the final calculation of responses that were consistent or inconsistent with theory. Table 36 Percent of All (270) Comments Irre levant and Relevant to Theory Shyness Level Number (Percent) of Uncodable Comments Number (Percent) of Codable Comments Number of Comments per Shyness Level Minimal-to-Low 48 (53%) 42 (47%) 90 Medium 44 (49%) 46 (51%) 90 High 35 (39%) 55 (61%) 90 Note: Sample size was five individuals per shyness level, for a total of 15. Table 37 and Figure 5 present results of the coding for Part A of Research Question 8. Results varied by shyness leve l. For individuals with minimal–to-low shyness, 26% of behaviors were consist ent with the Clark and Wells model. For individuals with a medium level of shy ness, 39% of behaviors were consistent

PAGE 154

143 with the model. Finally, 62% of behaviors of individuals with a high level of shyness were consistent with the Clark and Wells model. Table 37 Percent of Comments Cons istent with the Clark and Wells (1995) Model by Shyness Level Shyness Level Number (Percent) Comments Consistent with Theory Number (Percent) Comments Inconsistent with Theory Total Number of Relevant Responses Coded Minimal-toLow 11 (26%) 31 (74%) 42 Medium 18 (39%) 28 (61%) 46 High 34 (62%) 18 (38%) 55 Note: Sample size was five individual s per shyness level, for a total of 15. Figure 5 Percent of Comments C onsistent and Inconsistent with the Clark and Wells (1995) Model of Social Phobia Percent of Comments Consistent with Clark and Wells (1995) Model by Shyness Level 26% 39% 62% 74% 61% 38% 0% 20% 40% 60% 80% 100% LowMediumHigh Shyness Level Consistent Inconsistent

PAGE 155

144 Table 38 presents examples of comments that were coded as consistent with theory and as inconsistent with theory. Table 38 Examples of Comments Consistent and Inconsistent With Theory (Clark and Wells’ [1995] Model of Social Phobia} Scenario Comment Consistent with Theory Comment Inconsistent with Theory When in a group I hope I’m not called on. I may say something wrong or not express what I truly think or feel. Comfortable with this group of people. Like others are looking at me. Happy, contented, enlightened. When with friends Be polite as expected. Happy to be with friends. Will someone please ask me how I am doing? Feeling connected to friends and acquaintances. When at a social function Stick with people I know and try to find someone to talk to. Overall: Happy to be part of a joyous occasion. Standing at the edges of Comfortable with the

PAGE 156

145 the room. group of people at the gathering. When interacting with authority Reserved, stern, uncomfortable, shy. I would be feeling good about the possibility to help. This person is better than me (although I know they are not), and I need their approval. Glad that I attend a church with a loving pastor and family. The comments were coded again utilizing phrase as the unit of analysis. From the total number of phrases (281), 34 (12%) comm ents were irrelevant and were removed from the analysis, leaving a total of 247 phrases to be coded as consistent or inconsistent with t he Clark and Wells (1995) model. When utilizing phrase as t he unit of analysis, a sim ilar pattern of results was found, as displayed in Table 39 and Fi gure 6. Those participants with lower levels of shyness had more phrases for descriptions of their thoughts, feelings, and behaviors that were inconsistent wit h the Clark and Wells (1995) model than those participants with higher levels of sh yness (80% inconsistent compared with 20% consistent). For individuals with high levels of shyness, 68% of the phrases they used when describing their t houghts, feelings, and behaviors were consistent with the Clark and Wells (1995) model, and only 32% of their comments were inconsistent.

PAGE 157

146 Table 39 Percent of Phrases Consistent with the Clark and Wells (1995) Model by Shyness Level Shyness Level Number (Percent) Phrases Consistent with Theory Number (Percent) Phrases Inconsistent with Theory Total Number of Relevant Phrases Coded Minimal-toLow 14 (20%) 57 (80%) 71 Medium 37 (46%) 44 (54%) 81 High 65 (68%) 30 (32%) 95 Totals 116 131 247 Note: Sample size was five individuals per shyness level, for a total of 15. Percent of Phrases Consistent with the Clark and Wells (1995) Model by Shyness Level20% 46% 68% 80% 54% 32% 0% 20% 40% 60% 80% 100% Minimal-toLowMediumHigh Sh y ness Percent of Occurrences Consistent Inconsistent Figure 6 Percent of Phrases Consistent with the Clark and Wells (1995) Model by Shyness Level

PAGE 158

147 The comments were coded again, utilizing individual participant as the unit of analysis. As Table 40 and Figure 7 i ndicate, participants with higher levels of shyness were somewhat more likely to report behaviors, t houghts, and feelings that were consistent with theory more often than did t hose with lower levels of shyness. The results of this analysis indicate what percentage of the 15 individual participants made comments consis tent with theory. The finding that the trend in results by participant as the unit of analysis is similar to that for results by comment as well as by phras e as the unit of analysis, lends another measure of credibility to the findings. It s uggests that more individual participants with higher levels of shyness repor ted behaviors, thoughts, and feelings consistent with theory than did participants with lower levels of shyness, not merely that more comments or phrases were made consistent with theory, as would have been the case had only one or two participants with a high level of shyness made comments cons istent with theory. Table 40 Descriptive Statistics for Percent of Pa rticipant Comments Consistent With Theory by Level Percent of Participant Comments Consistent with Theory Shyness Level Mean SD Range Minimal-to-Low 21% 21% 50% Medium 32% 18% 42% High 70% 14% 34% Note: Sample size was five individuals per shyness level, for a total of 15.

PAGE 159

148 Percent of Participants Making Comments Consistent and Inconsistent with the Clark and Wells (1995) Model21% 32% 70% 79% 68% 30% 0% 20% 40% 60% 80% 100% LowMediumHigh Shyness Level Consistent Inconsistent Figure 7 Percent of Participant Comments Consistent and Inconsistent with the Clark and Wells Model Of the 127 comments that were mark ed as uncodable, approximately 90 concerned thoughts, feelings, and behaviors in two of the six church situations (in worship services and when performing a spec ific job for which the individual had volunteered.) These comments were subj ected to inductive analysis, and results are described below. The remainder of the comments marked as uncodable with respect to theory were highly diverse and contained no information relevant to the analysis and consequently were not coded. The comments regarding thoughts, feelings, and behaviors during wors hip services and when conducting a job, though not directly rela ting to the Clark and Wells (1995) model, were related to the findings of Spurr and Stopa (2002). These comments indicated that when in worship services or conducting a job, nearly all behaviors, thoughts, and

PAGE 160

149 feelings were focused on the task involved or else were focused on others, rather than self, for all levels of shyness. With only 15 comments in each of six sections for the behavior, thoughts, and feelings questions regarding worship serv ices and jobs, at most four coding categories were utilized for each question. In reporting what they did during worship services, 14 respondents talked about activities that would typically be expected, such as singing, standing w hen everyone else does, worshipping, or reading the Bible. One response was about the length of service. Only one individual described behavior consistent with the Clark and Wells model, stating that “I sit or stand quietly, trying not to receive any attention.” Most of the comments ( n = 8) about thoughts during worship services concerned thoughts about God, four comme nts indicated self-reflection about one’s relationship with God, and two addres sed concern about the other person’s welfare. Only one respondent reported thoughts consistent with the Clark and Wells model: “Sometimes I am intimidated to worship at church because I worry that people are watching and judging. So metimes I am able to break through and worship and other times I just struggle.” The responses describing feelings during worship services were mixed. Five were positive, four were both posit ive and negative, three were both positive and negative, one comment was not applic able, and one person did not answer this question. Again, only one comment was typical of the thoughts a shy individual might be expected to have, a ccording to the Clark and Wells (1995) model.

PAGE 161

150 Responses that concerned behaviors thoughts, and feelings when performing a job were also analyzed inducti vely. Fourteen of the 15 individuals whose comments were selected for analysis had a formal job to perform in the church. Each of the 14 responses regarding behavior when performing a job was different, as expected because each per son had a different job. Examples include, “Working with children,” “Singi ng. I am in the choir,” and “Greeting people as they come in.” Regarding w hat the respondents were thinking while performing their jobs, eight of the individuals reported having thoughts about the task they were performing or ways to help others, three made positive statements such as “I love what I do” one individual was wondering why more people did not help, and one per son reported feeling un comfortable teaching a Sunday School class even though prepar ations had been made. Regarding feelings while performing a job in the c hurch setting, nine individuals reported positive feelings (such as “Joyful and focu sed” or “Thankful for the opportunity to help out”), one reported both positive and negative feeli ngs, one reported feeling “anxious and uncomfortable” and two m ade comments about the work itself or why they had not done more. Based on results presented in this se ction, Research Hypothesis 8 was partially supported. For each of the th ree units of analysis (comment, phrase, and individual participant), as depicted in Table 41, the pattern of response was the same. As previously described, indivi duals with high levels of shyness made more comments consistent with the Clark and Wells (1995) model than individuals with medium levels of shyne ss, and individuals with medium levels of

PAGE 162

151 shyness made more comments consist ent with theory than individuals with minimal to low levels of shyness. Table 41 Coding Results by Unit of Analysis Shyness Level Unit of Analysis Percent of Units Consistent with Theory Minimal to Low Comment 26% Medium 39% High 62% Minimal to Low Phrase 20% Medium 46% High 68% Minimal to Low Individual Participant 21% Medium 32% High 70% Note: Sample size was five individuals per shyness level, for a total of 15. Research Question 9 To what extent is self-r eported fear of negative evaluation associated with attentional focus upon se lf and negative quality of thought in the six church situations?

PAGE 163

152 Research Hypothesis 9 Focus upon self and negat ive quality of thought related to the six church situations ar e associated with higher levels of selfreported fear of negative evaluation. The responses to the 18 write-in items described above were also analyzed for attentional focus and qualit y of thought by the author and the second coder. The author explained to the second coder the nature of the task referenced in each of the four scenario s. Working together, the author and second coder coded five sets of comments (previously utilized for training) for focus of attention and quality of though t. Then, working independently, the author and second coder coded a second set of five sets of co mments, attaining a 90% level of agreement, which rose to 92% after discussion. The author and the second coder coded the 270 comments fo r focus of attention and quality of thought, attaining a 91% level of agr eement. Table 42 and Figures 8 and 9 present the results. For Research Q uestion 9, all 270 comments were relevant to the content being analyzed, and all comments were coded. Table 42 Results of Coding for Focus of Attention and Thought Quality Shyness Level Focus of Attention Thought Quality Task or Others Self Positive Negative MinimalTo-Low 85 (94%) 5 (6%) 68 (75%) 22 (25%) Medium 74 (82%) 16 ( 18%) 59 (65%) 31 (35%)

PAGE 164

153 High 58 (64%) 32 (36%) 45 (50%) 45 (50%) Note: Sample size was five individuals per shyness level, for a total of 15. Relationship Between Focus of Attention and Shyness Level94% 82% 64% 6% 18% 36% 0% 20% 40% 60% 80% 100% Minimal to LowMediumHigh Shyness LevelPercent of Occurrences Task/Others Self Figure 8 Relationship Between Focus of Att ention and Level of Shyness (as measured by the BFNE-S, Context-specific) Relationship Between Quality of Thought and Shyness Level75% 65% 50% 25% 35% 50% 0% 20% 40% 60% 80% 100% Minimal to LowMediumHigh Shyness LevelPercent of Occurrences Positive Negative Figure 9 Relationship Between Quality of T hought and Level of Shyness (as measured by the BFNE-S, Context-specific)

PAGE 165

154 As was seen with the analyses for Research Question 8, level of shyness was found to be related to the focus of att ention. For this sample of participants, individuals reporting a minimal level of shyness tended to focus on the task at hand and to have a positive quality to their reported thoughts. As the level of shyness increased, respondents reported focu sing more on the self, rather than the task at hand, and their t houghts tended to have a negativ e quality more often. Table 43 presents examples of coding fo r the Focus of Thought and Quality of Thought analyses. Table 43 Examples of Coding for Focus of Thought and Quality of Thought Positive Quality of Thought Negative Quality of Thought Focus on Task or Others “I would feel comfortable with the group of people at the gathering” and “How can I help others?” (social situation) “The praise and worship and sermon could each be shorter” and “There is a lot of emotionality and we cater to these people.” (worship situation) Focus on Self “Talking with others and genuinely enjoying “I feel awkward whenever I am standing around”

PAGE 166

155 myself” (social situation) and “How grateful I am for all that God has done for me and my family.” (worship service) (social situation) and “I wish these people would know the real me and the conversation would not be such a surface conversation.” (group situation) To explore the statistical significance of the difference between quality and focus of thought for different levels of shyness, coded results were totaled. Comments reflecting a focus of thought on task or others received a score of 1, as did comments indicating a positive qua lity of thought. Comments indicating a focus of thought on self or a negative quality of thought, re ceived a zero. Data screening revealed the Focus on Task or Others and Positive Thought Quality data to have skewness and kurtosis values less than the absolute value of 1. Two one-way ANO VAs were conducted to measure the relationship for the 15 participant s between scores on the BFNE-S ( M = 34.31, SD = 8.66) and Focus of T hought on Task or Others ( M = 14.47, SD = 3.02) and Positive Quality of Thought ( M = 10.73, SD = 3.37). A one-way ANOVA for Focus of Thought was statistically significant, F (2, 12) = 8.19, p < .01. The means were 17.0, 14.8, and 11.6 for minimal, medium and high levels of shyness respectively, i ndicating that lower levels of shyness were associated with a tendency to focu s more on the task at hand or others,

PAGE 167

156 rather than on the self. Po st hoc multiple comparison tests revealed the Scheffe test for focus of thought was statistically significant at the .01 level between minimal to low and high levels of shyness. A one-way ANOVA for Qualit y of Thought was also statistically significant, F (2, 12) = 9.48, p < .01. The means fo r Positive Quality of Thought were 13.6, 11.8, and 9.0 for minimal, medium and hi gh levels of shyness respectively, indicating that lower levels of shyne ss were associated with a tendency to have a more positive quality of thought. Post hoc multiple comparison tests for Quality of thought revealed the Scheffe test was st atistically significant at the .05 level between minimal to low and high levels of shyness. Research Hypothesis 9 is supported by the results of this analysis. The relationship between focus upon task or others and level of shyness and the relationship between positive quality of thought and shyness were found to be statistically significant. This chapter has presented results of the data analyses conducted for this study. Results for both quantitative and qualitative data were reviewed. In Chapter 5, these results will be interpreted with respect to previous research. Chapter 5 also explains the limitations of the current study and discusses implications for future research.

PAGE 168

157 CHAPTER 5 DISCUSSION This chapter summarizes the major fi ndings of this study and interprets those findings in light of the research questions, after which the results are compared with previous research. Limitati ons of the study are also reviewed. Finally, implications of the findings are discussed, and recommendations for future research are presented. Overview of Method and Summary of Findings This study addressed nine research questions regarding the psychometric properties of the BFNE-S and the effect of reduced fear of negative evaluation on shyness. The sample was taken from an evangelical church, with a modal age range of 50-60. Data were collected utilizing Surveyonkey, a data collecting software, as well as paper/pencil. The survey consisted of demographic questions, the 12 items of the BFNE-S General, 12 items of the BFNE-S Specific, 10 items about comfort in vari ous social situations, 7 items regarding perceived acceptance in various situati ons, as well as 18 write-in responses about typical thoughts, feelings, and behaviors in church situations that were similar to the 10 situations that researc hers have identified as being difficult for shy people. The BFNE-S performed reas onably well, and 62% of the responses about thoughts, feelings, and behaviors reported by individuals with high levels of shyness were consistent with the Clark and Wells (1995) model of social phobia, an extreme form of shyness, as com pared with 39% and 26% of the responses

PAGE 169

158 of individuals with medium and low levels of shyness, respectively. Table 44 presents a summary of the findi ngs for the nine research questions in this study. Table 44 Summary of Major Findings Purpose Research Question Research Hypothesis Analysis and Results 1. To evaluate the psychometric properties of the BFNE-S, General and Context-specific, in a sample taken from the church setting, a previously unstudied population. What are the psychometric properties of the BFNE-S, General and Context-specific, i.e., the church? The BFNE-S, General and Context-specific, displays psychometric properties in the sample that are similar to those demonstrated for other populations taken from university or clinical settings. (A) Descriptive statistics and Cronbach alpha for scores from the BFNE-S, General and Context-specific, were generally similar to those in previous studies, although the means were lower and the standard deviations larger. (B) Confirmatory factor analysis for scores from the BFNE-S, General and Context-specific supported a one-factor model of shyness in the setting utilized in the current study. 2. To compare levels of perceived fear of negative evaluation (FNE) inside and outside the church setting. What is the difference in perceived fear of negative evaluation in the church setting compared to the non-church setting? Perceived FNE is lower in the church setting compared to the non-church setting. A dependent samples t -test to examine differences in means revealed a statistically significant lower level of perceived fear of negative evaluation in the church setting than outside the church setting. The effect size was negligible (0.01). 3. To compare levels of perceived fear of negative evaluation inside and outside the church setting across gender. What is the difference in perceived fear of negative evaluation in the church setting compared to the non church setting for males and females? The difference in FNE between the church and nonchurch setting is the same for males as for females. A repeated measures ANOVA revealed no statistically significant difference in perceived FNE for males and females in the church setting compared to the non church setting. 4. To compare levels of perceived fear of negative evaluation inside and outside the church setting across race. What is the difference in perceived fear of negative evaluation in the church setting compared to the The difference in FNE between the church and nonchurch setting is the same for different races. A repeated measures ANOVA revealed no difference in perceived FNE for Caucasians, African Americans, Hispanics, and persons of multiracial background in the church

PAGE 170

159 non church setting for individuals of different races? setting compared to the non church setting. 5. To compare the levels of perceived acceptance by people inside and outside the church setting. What is the difference in perceived acceptance between people inside and outside the church setting? Levels of perceived acceptance by people in the church setting are higher than the levels of perceived acceptance by people outside the church setting (A) Descriptive statistics for the items on the perceived acceptance checklist, outside and inside the church setting, were very similar. (B) A dependent samples t test failed to reveal a statistically significant difference in levels of perceived acceptance by people inside and outside the church setting. 6. To compare levels of self-reported comfort for people in the church setting compared to outside the church setting. What is the difference in self-reported levels of comfort outside the church setting and inside the church setting? Levels of comfort perceived by people in the church setting are higher than the levels of comfort outside the church setting. The dependent samples t -test revealed a statistically significant difference between mean levels of comfort in the general setting and in the Context-specific setting. The effect size was negligible (.09). 7. To understand how Context-specific issues (extent of involvement in church activities) relate to selfreported fear of negative evaluation in the church setting To what extent do Contextspecific issues relate to selfreported levels of fear of negative evaluation? Greater depth and breadth of involvement in church activities are associated with reduced selfreported fear of negative evaluation. A multiple regression for fear of negative evaluation using length of membership, regularity of attendance, number of activities participated in per month, and number of close friends as predictor variables failed to reveal a statistically significant relationship. 8. To seek confirmation or disconfirmation of the Clark and Wells (1995) model via examining the extent to which the thoughts, feelings, and behaviors of shy individuals in the church setting are consistent with the model. How do shy people typically think, feel, and behave in an environment hypothesized to have less fear of negative evaluation and self-focus? At least 75% of the responses of individuals with high levels of FNE will report thoughts, feelings, and behaviors related to six church situations that are consistent with the Clark and Wells (1995) model and that will be at least 10% more than those with low levels of FNE Sixty-two percent of the thoughts, feelings, and behaviors for individuals with high levels of shyness in the church setting were consistent with the Clark and Wells (1995) model compared with 39% of individuals with medium levels of shyness and 26% of individuals with low levels of shyness. 9. To seek support of theory via examining the extent to which attentional focus is To what extent is self-reported fear of negative evaluation Focus upon self and negative quality of thought related to the six church (A) A one-way ANOVA revealed a statistically significant difference between attentional focus scores and

PAGE 171

160 related to selfreported levels of fear of negative evaluation. associated with attentional focus upon self and negative quality of thought in the six church situations? situations are associated with higher levels of selfreported fear of negative evaluation. scores from the BFNE-S, Context-specific, for individuals with minimal to low and high levels of shyness. (B) A one-way ANOVA revealed a statistically significant difference between thought quality scores and scores from the BFNE-S, Context-specific, for individuals with minimal to low and high levels of shyness. Research Question 1 Confirmatory factor analyses of the BFNE-S, General and Context-specific, were conduc ted to investigate whether the data suggested utilizing a two-factor model, ra ther than the one-fact or model on which most of previous literature was based. The data revealed nearly identical fit indices for both versions of the BFNE-S and highly similar parameter estimates and modification indices. The confirma tory factor analyses supported the onefactor model of shyness reported in extant literature and, consistent with parsimony, the one-factor model was utilized in the curr ent study. Accordingly, researchers in the area of shyness can continue with a greater degree of confidence that the items on the BFNE-S represent an essentially unidimensional construct. The higher number of stat istically significant modi fication indices for the Context-specific version of the oneand two-factor models may have been due to order effects. It is speculated t hat respondents may have hurried through the Context-specific version of the BFNE -S because it immediately followed the general version. They may have been impat ient responding to the items they

PAGE 172

161 thought they had responded to previously if they did not read the directions carefully. Research Question 2 In considering findings regarding the BFNE-S, it should be kept in mind that t he current study utilized two versions of the BFNE-S. Respondents were asked to comple te the questions when thinking about situations in general outside the chur ch, and these responses constituted the distribution of scores for the BFNE-S, General. Respondents were asked to complete the questions again when thinking about situations in general inside the church; those responses comprised the distribution of score s for the BFNE-S, Context-specific. The most significant finding conc erning the BFNE-S was that the instrument performed reasonably well with a sample of individuals recruited from a church setting. The distributions of scores were similar in that they approximated a normal distribut ion; however, the means for both versions were lower and the standard deviations larger than those reported in previous research for non-socially anxious samples (C ollins et al., 2005; Duke et al., 2006; Rodebaugh et al., 2004; Weeks et al., 2005). The means for the BFNE-S, General and Context-specific, were compared with those in previous studies. Cohen’s d effect sizes were calculated to assess the practical significance of the diffe rence in scores between the current study and previous ones. Effect sizes are ranked in order of size, as displayed in Table 42. The largest effect size (2.59) was found for the BFNE-S, General, with the Collins et al. (2005) study, which consisted of i ndividuals with social phobia. The

PAGE 173

162 smallest effect size (0.05) was for the BFNE-S, General, and the non-socially anxious sample in the Weeks et al. ( 2005) study, as depicted in Table 45. Table 45 Effect Sizes Reported for BFNE Scale (ordered by effect size) Effect Size Study Author(s) N Sample Description M SD Alpha BFNE-S, General BFNE-S, Contextspecific Watson, 2009 232 Members, regular attenders, and visitors in a large evangelical church – BFNE-S, Context-specific 25.22 11.09 .94 --Watson, 2009 226 Members, regular attenders, and visitors in a large evangelical church – BFNE-S, General 26.50 10.39 .93 --Weeks et al., 2005a 1385 Non socially anxious 26.81 4.78 .90 0.05 0.26 Collins et al. 2005b 30 Community sample (non-anxious) 29.20 8.20 .97 0.27 0.37 Rodebaugh, Woods, Thissen, Heimberg, Chambless, & Rapee, 2004 1,049 Archival data (anxious and non-anxious) 29.41 7.72 -0.35 0.50 Duke et al., 2006 355 Individuals in a shopping mall 32.30 7.34 .94 0.99 1.16 Collins et al., 2005b 99 Individuals with panic disorder 39.80 12.50 .97 1.61 1.69 Weeks et al., 2005a 138165 Individuals with social anxiety 46.91 9.27 .92 1.87 1.91 Collins et al., 2005b 82 Individuals with social phobia 51.50 7.30 .97 2.59 2.57 a The Weeks et al. (2005) study consisted of two samples. bThe Collins et al. (2005) study consisted of three samples. Inter-item reliability was assessed with a subsample ( n = 107). It is possible that the lower means were found because participants in the sample were older. The modal age r ange for the current study was 50 to 60 years (51% were 50 or more years of age, with 77% being 40 years of age or

PAGE 174

163 more) and the modal age range was 50 to 60 years, whereas previous studies focused on younger participants. The co rrelation between the BFNE-S, General, and age in the current study was r = -.19, p < .01. The correlation between the BFNE-S, Context-specific, and age was r = -.22, p < .01. The findings also suggest that the sh yness regular church attenders report experiencing in the church environment is similar to what they experience outside the church environment, though not as intense. The reader might recall that 91% of the survey respondents reported attending church 3 or more times per month. The larger standard deviations for sco res from the BFNE-S, General and Context-specific versions, compared to th ose from previous studies, might be explained, again, by differences in the par ticipants’ ages. It is possible that greater diversity in age led to greater diversity in scores. A repeated measures ANO VA was conducted to test the effect of age on self-reported fear of negativ e evaluation as measured by the BFNE-S, General. A repeated measures ANOVA was conducted to test the relation between age and self-reported fear of negative evaluation as measured by the BFNE-S, General and BFNE-S, Context-specific Levene’s test for both comparisons indicated no evidence of heterogeneity of va riances. A statistically significant main effect for setting was found, F (1,215) = 13.79, p < .01. No statistically significant interaction effect between setting and age was observed, F (2,215) = 2.16, p > .05. A statistica lly significant between-subjects main effect for age was found, F (2,215) = 6.94, p < .01. Pair-wise com parisons among the three age groups revealed that individuals in t he 31 to 60 age group had less fear of

PAGE 175

164 negative evaluation than did the 21 to 30 age group, and individuals in the 60 plus age group had less fear of negative ev aluation than the 21 to 30 age group. It would have been interesting to compar e the stability of the participants’ responses across settings (in church and outside church) with participants’ scores on the state/trait anxiety inventor y (Spielberger et al.,1970). Such a comparison might have helped account fo r the relatively small difference between self-reported st ate anxiety inside and outside church. If the sample had many individuals with high trait anxiety, t hose individuals would likely tend to report high levels of anxiety in any sit uation. The decision was made not to include this instrument, however, because the survey was quite lengthy. A chi square was utilized to determi ne whether there was a statistically significant relationship between the ques tion on the Zimbardo’s Shyness Survey (1974), “Are you shy?” and the scores on the BFNE-S, Context-specific, were compared utilizing a Pearson chi square. This relationship was statistically significant, 2 (2) = 17.82, p < .001), which supported t he validity of the BFNE-S as a measure of shyness. Research Questions 3 and 4. Another important finding was that the results of this study indicated no statis tically significant differences between the genders or among the races in levels of perceived fear of negative evaluation inside and outside the church setting. This seemed unsurprising because research regarding gender differences is mixed (Bruch et al., 1989; Pilkonis, 1977), as discussed previously in Chapter 2. Research has been conducted on the effect of cultural influences on shy ness; however, information on race, but not

PAGE 176

165 on cultural background, was co llected. This represents a limitation of the current study. Research Question 5 It was hypothesized that individuals would report feeling more accepted inside the church t han outside. However, no statistically significant difference was found in self-r eported levels of acceptance. This finding is consistent with the results fo r the BFNE, which showed little difference in perceptions of the church setting and outside the church. Research Question 6 The hypothesized difference between comfort outside and inside the church when in sit uations known to be difficult for shy people was found to be statistically signi ficant; however, the effect size was negligible. This very small difference is consistent with the small differences found for other variables that were us ed in comparing participant responses outside the church with responses for inside the church. The effect sizes were negligible. Research Question 7 Four aspects of the respondents’ relationship with other church members and regul ar members were utilized to explore the effect of Context-specific issues (depth and breadth of involvement in church activities) to self-reported levels of fear of negative evaluation in the church setting, as measured by the BFNE-S, Context-specific The four aspects were length of attendance, regularity of a ttendance, number of activities participated in per month, and number of close friends. Of the four, only one (number of close friends) was found to have a statistically significant relationship with fear of negative evaluation. A multiple regressi on analysis also failed to reveal any

PAGE 177

166 statistically significant relationships bet ween the context variables and fear of negative evaluation. Again, this lack of measurable difference is consistent with the lack of or small diff erences between participant responses pertaining to outside the church and participant respons es pertaining to inside the church. Research Question 8 Regarding the Clark and Wells (1995) model of social phobia, it was predicted that indi viduals with high levels of shyness would tend to report a greater percentage of st atements reflecting safety behaviors, excessively high standards for social perfo rmance, conditional beliefs concerning consequences, unconditional negative bel iefs about the self, and anxiety and other emotional distress, such as fear or worry compared to individuals with lower levels of shyness. This pred iction was supported in that 62%, 39%, and 26% of the self-reported behaviors, thought s, and feelings of individuals with high, medium and low-to-minimal levels of shyness in the six church situations, respectively, were consistent with the Clark and Wells (1995) model of social phobia. This finding suggests that the majori ty of individuals with high levels of shyness have the same kinds of thoughts, feelings, and behaviors in the church setting as in other settings. Much of the research has been conducted in the highly competitive environment of the university or in th e clinical setting. It was hypothesized that the church setting would hold less potential for fear of negative evaluation. Although the cu rrent study revealed that the church setting showed only slightly less potential for fear of negative evaluation, it should be noted that participants reported feeling fairly comfortable and accepted, as measured by the

PAGE 178

167 items addressing those topics. The means were 8.73 and 8.66, respectively, for five items regarding acceptanc e outside and inside the church setting. The scale for responses ranged from 1 to 5, with 1 indicating the participant felt “Very Accepted” and 5 indicated “Very Un accepted. Additionally, means for the 10 items regarding the comfort scale were 23.31 and 22.64 respectively for 10 items regarding comfort outside and inside the c hurch setting. The scale ranged from 1 to 5, with 1 indicating a feeling “V ery Comfortable” and 5 indicating feeling “Very Uncomfortable.” It was thus not possible to understand whether operating in an environment with less potential threat enables individuals with high levels of shyness, or social phobia, to over come the hidden assumptions and habits delineated in the Clark and Wells (1995) model. As mentioned earlier, these assumptions and habits involve safety behavio rs (such as avoiding situations), excessively high standards (“I must not show any sign of weakness”), and negative emotions, such as anxiety. The way in which the data were collected also could have affected the responses. Conducting interviews, with the opportunity to ask clarifying follow-up questi ons, might also have yielded different results. It is also possible that some people who attend church, even those who attend regularly, perceive the environment as holding even greater potential for being evaluated negatively in that they do not feel they are accepted. In citing Schaller (1978), McIntosh and Ma rtin (1992) concluded that:

PAGE 179

168 there is considerable evidence which s uggests that at least one-third, and perhaps as many as one-half, of all Protestant church members do not feel a sense of belonging to the congr egation of which they are members. They have been received into membership, but have never felt they have been accepted into the fellowshi p circle. (p. 77) Research Question 9 The statistically significant and practically significant relationship between attenti onal focus scores and shyness levels as measured by the BFNE-S, Context-specific was related to the results described above. Individuals with high levels of sh yness were found to focus more often on the self than when in social situations than individuals with minimal to low levels of shyness, who tended to focus more oft en on the task at han d or other people. This finding also lends support to prev ious research findings (Spurr & Stopa, 2002). Additionally, statistically significant and practically significant relationships were found between thought quality scores and shyness levels as measured by the BFNE-S, Context-specific. High levels of shyness were found to be associated with negative thought qual ity. That is, individual s with high levels of shyness tended to have more negative thoughts when experiencing the scenarios utilized in this study than did individuals with low to minimal shyness levels. This is consistent with Cl ark and Wells’ (1995) ex plication of the processes activated when an individual wit h social phobia perceives a social threat, whether that threat is real or not, in that the individual’s negative assumptions are activated. These negative assumptions include, as described

PAGE 180

169 earlier, self critical thoughts, such as “I ’m unacceptable” or “I am weird.” These findings are consistent with previous re search and further support the utility of including efforts to change the focus of attention when attempting to modify shy behaviors and their consequences. A second major finding regarding Res earch Question 9 concerned the focus of attention. Of s pecial interest was the fact that the Clark and Wells (1995) model of social phobia accounted fo r less than 1% of thoughts, feelings, and behaviors in two settings where the par ticipant would typically be expected to focus attention on the task or others, rather than on the self. The two settings were when in worship services and when fu lfilling the duties of a designated job, such as singing in the choir or helping with children. A key f eature of the Clark and Wells (1995) model is the shift in focu s of attention from the environment to self. The shift in focus of attention is so foundational to the model that the first part of the cognitive treat ment for social phobia bas ed on the model begins with encouraging the patient to “. . drop their safety behaviors and focus their attention on the other person(s) in the interaction and on what is being said” (Clark, 2001, p. 421). In this study, i ndividuals with high levels of shyness reported acting, thinking, and feeling mu ch like individuals with low-to-minimal levels of shyness when their attenti on was focused on activities that were presumably important to t hem (i.e., participating in worship and performing a job for which they had volunteered). Why thes e individuals were able to shift focus away from self is not clear. Future re search will need to inve stigate this issue.

PAGE 181

170 Participants reported that during wors hip services they were thinking “ How much I love God and how grateful I am for all he has done for me and my family”, and that they were “Thanking God for getti ng me where I am and how I got here”, and “Considering God, thinking what my part in His plan I have . ” When conducting a job, some individuals said they were thinking things like, “Hope I can make all welcome, comfortable and make them smile” and “I can make a difference in these young lives. I am so thankful to have this opportunity.” Comparison of Findings with Theoretical Framework Two salient characteristics of the sample should be kept in mind when considering results for the current st udy. As mentioned previously, a large percentage of survey responde nts were older (51% were 50 or more years of age, with 77% being 40 years of age or more) and the modal age range was 50 to 60 years. The ages of participants in previous studies tended to be much lower, as displayed in Table 46. Many survey respondents were also long-time members of Church A (46% had been members for five or more years). It seems likely that these two findings account fo r at least some, perhaps a great deal, of the disparity between act ual and predicted results Table 46 Participant Ages in Previous Studies of the BFNE Scale Study Author(s) N Sample Description M Range Collins et al., 2005 82 Individuals with social phobia 38 17-68 99 Individuals who experience

PAGE 182

171 panic disorder 30 Community sample (nonanxious) 33 20-49 Weeks et al., 2005 138-165 (missing data varied) Individuals with social anxiety 32.39 -138-165 Non socially anxious 33.12 -Duke et al., 2006 355 Individuals from a shopping mall 43 18 to 86 Rodebaugh, Woods, Thissen, Heimberg, Chambless, and Rapee, 2004 1,049 Archival data (anxious and nonanxious) 22.5, 23.2, 20.5, and 20.7 -Results of this study generally lend support for the Clark and Wells (1995) cognitive model of social phobia, which emphasizes the role of the shift in attention that occurs when an individua l perceives, whether accurately or inaccurately, that he or she is about to be evaluated negativel y. Although 75% had been hypothesized, more than one half (62%) of the behaviors, thoughts, and feelings of individuals with self-r eported high levels of shyness were consistent with the Clark and Wells (1995) model when the individuals were in an environment determined, albeit by self-r eport measures, to hold slightly less potential for fear of negativ e evaluation. The reader will recall that according to

PAGE 183

172 Weeks et al. (2005), fear of negative evaluation is the core feature of social anxiety disorder, or shyness. Also lending support to the Clark and Wells (1995) model is the finding that the model explained less than 1% of behaviors, thoughts, and feelings of individuals when they were engaged in activi ties that directed their attention to a task and/or to others, rather than the self, such as attending a worship service or performing a job. Recall that in the Clark and Wells (1995) model, one of the two key components that maintain social phobia is focusing of attention upon self, with the other being negativ e thoughts about the self. When one is in a worship service, one’s attention is generally on the speaker and the singers and God. When performing a job, an individual typicall y is concentrating on what has to be accomplished. The findings support the as sertion by Wells (2001) that many shy individuals possess adequate social skill s and are able to function quite adequately in social situations once their attention is focused outward, rather than inward. The 6 situations utilized in the current study were extremely similar to the 10 situations known to be difficult for shy people (Crozier, 2001). The discomfort reported by shy individuals in their write-in comments r egarding thoughts, feelings, and behaviors in four of t he six situations also supported this component of the current study ’s theoretical framework. The four situations where individuals with higher levels of shyness reported being uncomfortable more often than did individuals with lower levels of shyness were when interacting in groups, in social situations, with friends and acquaintances, and

PAGE 184

173 with persons in authority. P earson correlations revealed that higher levels of fear of negative evaluation were asso ciated with higher levels of discomfort (The response options for the comfort items r anged from 1 [“Very Comfortable”] to 5 [“Very Uncomfortable”]. Spec ifically, the correlation between the BFNE-S Context-specific, and comfort in general outside the church was r = .21, p < .01, and the correlation with comfort inside the church was r = .17, p < .05. The correlation between the BFNE-S, General and comfort in general outside the church was r = .22, p < .01 and with comfort inside the church was r = .18, p < .01 The findings of the current study al so provided further evidence that scores obtained with the BFNE-S have very good psychometric properties, and these properties are consist ent with those reported by Or sillo (2001). Cronbach alphas for both versions of the BFNE-S were excellent and similar to those reported in previous research (Collins et al., 2005; Duke et al., 2006; Rodebaugh et al., 2004; Weeks et al., 2005). The c onfirmatory factor analyses provided additional evidence supporting the validity of the one-fa ctor model of shyness reported in the liter ature (Crozier, 2001). Limitations The findings of this study should be interpreted with caution due to limitations and possible threats to internal validity as well as to external validity of the findings. Regarding limitations, order effects were not assessed for the components of the Personal Concerns and Issue Survey. All respondents completed the BFNE-S, General followed by the BFNE-S, Context-specific.

PAGE 185

174 Second, another limitation of the study was that co mpleting the BNFE-S scale might have influenced how participants de scribed their thoughts, feelings, and behaviors. It is also possible that extr emely shy individuals might have declined to participate due to discomfort in discl osing information about what is often experienced as a significant and embarra ssing personal problem. Alternatively, participants who had low levels of shy ness might have declined participation if they believed the topic of t he study to be unimportant. If that were the case, the findings might not have been valid for the sample utilized in the study. The researcher attempted to minimize t he possibility of biased selection of participants when presentations were made to solicit participants by emphasizing the need for non-shy as well as shy individu als to participate. The researcher also provided reassurances that all data would be treat ed confidentially, stating that any reports that were written woul d not include information that would enable individuals to be identified. The reader will recall that the data were collected electronically and participants did not need to interact socially or in any way with the researcher. The distance created by colle cting data electronically might have reduced fear of negative evaluation, which wo uld have, presumably, allowed participants to give a more accurate response than if they had been interacting fa ce to face. The church selected for the study wa s an already-formed group that likely differed in important ways from other churches or other settings in which perceived potential for negative evaluati on might exist. This limitation,

PAGE 186

175 unavoidable due to study design, was handled by exercising an abundance of caution in drawing conclusions and in making generalizations. Threats to Internal Validity The most important threat to internal validity was that the data were correlational, and the researc her had no ability to control the setting. Researcher bias was another threat to internal validity. It was pos sible that the halo effect occurred during analysis of qualitative data because the researcher had prior knowledge about the partici pants in general and personal assumptions about how participant beliefs might account fo r some of the results (Onwuegbuzie, 2003b); however, every attempt was made to avoid letting that knowledge influence the interpretation of the findings. Results and interpretations of all data were discussed with committee members un til consensus was obtained. During data analysis, it was understood that kno wledge of the participants’ shyness levels as measured by the BFNE-S, Contex t-specific, might cause the researcher to perceive most or all participant respons es as findings consistent with theory. To mitigate that possibility, the researc her, as well as the second coder, were not aware of the participant’s shyness le vel, and findings were discussed with colleagues who served as disinterest ed peers who had no stake in the findings and interpretations (Lincoln & Guba, 1985, p. 308). Finally, a reflective researcher journal was maintained, as a tool to make obvious, and thus more avoidable, any hidden assumptions that might have influenced interpretation. Confirmation bias, the tendency for a re searcher to find what he or she wants to find and to ignore, or misinter pret, anything else (Nickerson, 1998),

PAGE 187

176 represented a considerable threat to internal validity. The res earcher could have perceived level of shyness as influencing participant responses. As Greenwald, Pratkanis, Leippe, and Baum gardner (1986) note, this ty pe of potential bias is most often present when the ai m of a study is to test, ra ther than create, theory. Careful attention was paid to this thr eat through several means. Throughout the study, and particularly during the analysis stage, the researcher discussed her ideas with disinterested colleagues w ho had no personal stake in the research and who were unfamiliar with the church environment. Findings were also discussed with peers who were familia r with the church environment. Additionally, a second coder was utilized in analyzing the responses to the openended questions about thoughts, feelings, and behaviors in the six church situations. The coder was trained, sa mple comments were coded together, and discussion of the findings ensue d until a satisfactory rate of agreement (90% or greater) was obtained. Additionally, the re searcher kept a re flexive journal, as described by Lincoln and Guba (1985). Throughout the study, the researcher made journal entries to record relevant information about herself as well thoughts and decisions about methods and analyses. The reflexive journal was a valuable tool in maintaining objectivity and avoiding biases. The researcher had to work constantly to be aware of her personal bias—that is, the strong desire to find a way to help individuals with high levels of shyness feel more comfortable in the church setting and begin finding out what might keep them from beco ming more connected with the church. Entries made before data were collected indi cate a high level of emotion, both

PAGE 188

177 positive and negative, which made it challenging to be objective. Statements included: “It is frustrating to have to do so much paperwork, in such detail [the IRB process], just to collect data. I know it is necessary but it takes so long and I want to get on with it”, and “It is exciti ng to finally be submitting the IRB. It doesn’t seem possible after all these year s of wanting to help shy people that I am finally doing the research!” Entries while data were being colle cted and a report was being written for the church indicate, again, frustration. “I know I need to be patient but I wish I did not have to collect 250 surveys!”, “It is taking so long to analyze the comments for the church questions. I know it will take as long to write this as it does to write a report at work, and I am learning a lot about part of the data. I know I cannot wait until after the dissertation is finis hed to do this, so I will just keep working and know that it will all get done eventually.” After the church report was co mpleted and the researcher began analyzing the data for the disse rtation, it was difficult to be patient with the need to proceed one step at a time. One comment is a good example of the impatience to find out the results all at onc e: “Why does it take so long for every aspect of every question? Why does it have to be so tedious? Well, if I try to go fast, I will make mistakes.” Perhaps the most important items re corded in the journal concerned the interpretation of data. Personal re flection reveals that had there not been numerous discussions with co-chairs the results would have been interpreted and presented, albeit unintentionally, in a biased and inaccurate manner. For

PAGE 189

178 example, the researcher had initially coded the write-in co mments and then had the second coder indicate the percent age of agreement. When a co-chair required that the comments be coded inde pendently by the second coder, the researcher was frustrated wit h the delay that caused. One comment was “I do not want to wait and have the coder do the coding independently. I have used the percent of agreement bef ore. I can see the poin t, though. . “ After the coding had been re-done, one entry notes, “I can see it changed the results, so I am glad for having done it. It made the anal ysis more accurate.” The researcher was again frustrated initially with the nex t draft with the necessity to go back and re-categorize all the comments, adding a category regarding whether the comments were relevant to the topic or not Again, it was worth it, even with all the extra work, because it dramatically improved the accuracy. One comment was, “I hope this is the last time I hav e to re-do this piece but if there is something else wrong, I want to find it out and fix it.” Somewhere along the way, the researcher developed a degree of pat ience, an essential quality for one who wishes to conduct quality research. The process of analyzing data and interpreting the results was lengthy and involved numerous emails and tel ephone conference calls between the researcher and the co-chairs. It felt at times to the researcher that the discussions were obstructive but pers onal reflection always revealed the accuracy of the co-chair’s comments and perceptions. In sum, the deep interest in the subject was a great benefit to the researcher in conducting the current study in that it provided motivation and the

PAGE 190

179 patience to undergo a lengthy, detailed proc ess. However, it also created a threat to the validity of interpretations that were made regarding both the qualitative component of the study as well as statements relating findings of the current study to previous research. It was difficult not to see the results as indicating the causal linkages that were predicted. The res earcher experienced in a very meaningful way the benefits of ex tensive collegial discussion, especially when topics of personal interest are in volved. Recording personal reflections after engaging in discussions helped work through and resolve many questions and helped clarify thinking that became, at times, muddied by intense personal interest. Threats to External Validity Threats to external validity in the cu rrent study included population validity. The utilization of one church congregati on from which to draw the sample constituted a threat to populat ion validity, and it is rec ognized that findings of the current study may not be generalizable to c hurches of other denominations or to churches differing significantly from the congregation utilized in the study. The potential inability to generalize findings bey ond the church setting to the everyday world is an even larger threat to ecolog ical validity, but unav oidable due to study design. The threat of self-selection or “vol unteer bias” (Bordens & Abbott, 2004, p. 122) is of particular relevance to t he current study. Rosenthal and Rosnow (1975) have reported that persons who volu nteer for research often tend to be more social than do non-volunteers. It is possible that individuals who were less

PAGE 191

180 social could have declined to participate. An attempt was made to minimize the potential effects of volunteerism by emphas izing the critical need for participants with social anxiety. Regarding the potential threat to the trustworthiness of findings (Lincoln & Guba, 1985), or legitimation threats pertain ing to the qualitative component of the study (the extended response questions), prolonged engagement was inherent in study design, in that the researcher had approximately 29 years of experience in the church setting, as well personal ex perience with being shy. Typically, the purpose of prolonged engagement is to ensure, to the extent pos sible, that the researcher has sufficient experienc e with the phenomenon or culture under investigation (Linco ln & Guba, 1985). Another aspect of study design is that the researc her had given the subjects comprising the study much thought whereas participants were asked to comment about situations to which they may not have given much thought. Had the participants been allowed to reflect on the questions for a period of time before responding, more depth and breadth of responses would likely have been obtained. Throughout the study, the researcher attempted to maintain awareness of the possibility of personal bias potentiall y distorting the findings. As Lincoln and Guba (1985) state, “. . awar eness [of that possibility]. . is a great step toward prevention” (p. 304). Although the re searcher was removed from the data collection process and there was no personal interaction with the participants, the researcher recognized the potential fo r personal beliefs and assumptions to

PAGE 192

181 influence interpretations. The reflective journal and discussions with colleagues and peers helped reduce t hat possibility. Several methods were utilized to enhance researcher legitimation. According to Patton (1990), triangulati on is “. . the combination of methodologies in the study of the same phenomena or programs” (p. 187). In an effort to obtain maximal l egitimation, negative case analysis was utilized. As Patton (1990) explains, “Where pattern s and trends have been identified, our understanding . is increased by consi dering the instances and cases that do not fit within the pattern” (p. 463). Particul ar attention was given to responses to thoughts, feelings, and behavior comments t hat seemed to indicate shyness operates differently in the c hurch setting than in the every day world. In fact, one third of the scenarios utilized in the study (behaviors, thoughts, and feelings during worship services and when performing a job) were subjected to negative case analysis. These scenarios included individuals with high BFNE scores but comments inconsistent with the Clark and Wells (1995) model. Comments from individ uals with high levels of shyness that were consistent with the Clark and Wells ( 1995) model were (when interacting with authority figures),“This person is better t han me (although I know they are not), and I need their approval.” and (regarding social occasions), “I do not go to many just social things because it is hard to make small talk and I feel awkward standing around.” Inconsistent comments included statements such as thinking when conducting a job, “I hope I can make all welcome, comfortable and make

PAGE 193

182 them smile.” and (when in a group) “What encouraging words can I speak? How can I make this a moment filled with purpose?” Individuals with low-to-minimal levels of shyness reported that when with friends they would “ . visi t, say hi, what are you up to, ask how are things“ and they would be “having fun and joking around” at a social occasion. In some cases, individuals with low-to-minima l levels of shyness made comments consistent with the Clark and Wells (1995) Model. One participant said that when teaching a class “Most of the time I felt uncomfortable in front of the class leading the discussion even t hough I felt I was prepared.” Benefits of mixed methods research. As Onwuegbuzi e and Leech (2004, p. 770) state, the “ability to extract signi ficance from . data is compromised by the limitations inherent in the method of extraction.” Onwuegbuzie and Leech contend that the Interpretation of signi ficant findings in both quantitative and qualitative research can be enhanced by mixed methods data analyses. The interpretation of significant findings in the quantitative portion of this study was undertaken sequentially (Onwuegbuzie & Le ech, 2004). The qualitative data were utilized to ascertain the level of consistency among the thoughts, feelings, and behaviors of individuals with high levels of shyness, as measured quantitatively by the BFNE-S, and a wellestablished theory, the Clark and Wells (1995) model of social phobia. From the inception of t he current study, the resear cher planned to utilize mixed methods techniques. Collins, Onwuegbuzie, and Sutton (2006) conceptualize mixed methods research as involving 13 steps, beginning with “. .

PAGE 194

183 determination of the goal of the study . [and ending with] “. . writing the final report, and . reformulating the research questions (p. 69-70).” The goals of the study were to extend the usefulness of a well-researched theory as well as to explore the psychometric properties of an extensively used measure of shyness in a previously unexplored setting. A quantitative measure was necessary to ascertain which individuals had high, medium, and low-to-minimal levels of shyness. Quantitative measures were also utilized to gauge the nature of the environment and the effect of characteristi cs of that environment on participants. Comparisons of findings with existi ng theory were made possible through qualitative analysis. Implications for Future Research To explore further the valid ity of scores obtained with the BFNE-S, it would be useful to conduct a study in diffe rent environments and seek to understand individuals in those environments with high and low fear of negat ive evaluation. The data could be used to determine whet her the environment was categorized as having high or low potential for negative evaluation. As the current study has indicated, assumptions can be inaccurate. It should also be helpful to include the State-Trait Anxiety Inventor y (Spielberger, et al, 1970) as one of the measures in a future study comparing the effect of different environments on shyness. Trait anxiety could be controlled for statistically. Future research regarding the chur ch setting should utilize a sample consisting of more individuals who are not long-term me mbers and regular attenders. As stated prev iously, had the sample in the current study included

PAGE 195

184 fewer individuals with a long-term relati onship with the church, the results might have been substantially different. Although the relationship between fear of negative evaluation and length of membership was not strong, participants who were relatively new to the church mi ght have reported feeling less accepted and the open-ended comments for indivi duals with medium levels of fear of negative evaluation might have been more consis tent with the Clark and Wells (1995) model. Had such been the case, they mi ght have thought, felt, and acted more like a person with a high level of shyness. Exploration of the 10 situations known to be difficult for shy people would al so be helpful. Interviews could be conducted or surveys could be administered to explore why individuals with high levels of shyness feel uncomfortable in such situations and what has either helped alleviate their levels of discomfor t or what they believe would provide a greater degree of comfort. Fi nally, a follow-up study involving in-depth interviews over a fairly long period of time wit h shy individuals could provide deeper understanding of the functioni ng of the Clark and Wells (1995) model of social phobia. Additionally, future studies coul d probe how people use their religious beliefs to help them cope with difficult si tuations, in particular shyness. It is possible that coping mechanisms are differ ent for individuals with high levels of shyness compared to those with medium or low levels. It would also be interesting to explore if strong religious beliefs change an individual’s perspective about shyness and about being evaluated.

PAGE 196

185 Based on results of this study (R esearch Question 9), a study to gauge the effect on shyness of a shift in attenti on from self to others or the situation at hand might provide guidance for treat ment. A long-term study could be undertaken with participants keeping diaries of how they felt when they were able to concentrate on other people rather t han themselves when they were engaged in social situations. Analyzing the comments in the diaries could provide information on what kinds of things help in dividuals focus on others, rather than self. In concert with keeping diaries, or perhaps in a separate study, individuals could wear an unobtrusive device to record physiologic responses, such as heart rate, when in various situations. Conclusions The field of measurement has benefitted from the current study in that it provides additional evidence regarding the generalizability of scores yielded by the BFNE-S across settings. The psychom etric properties of the BFNE-S were found to be robust in a setting not utilized in previous studies. The BFNE-S can be used with greater confidence as a result of the current st udy with older adults operating in a church setting. The BFNE -S can facilitate research and hence extend theory. Confirma tory factor analysis revealed a good, although not optimal, fit for the one-factor model of shyness. The current study has provided evi dence supporting one of the most wellresearched theories of shyness. Resear ch indicates shyness can be affected by many environmental conditions (Henderson & Zimbardo, 2001). Two of the most

PAGE 197

186 critical environmental factor s are fear of negative evaluat ion (Weeks et al., 2005) and self-focus (Spurr & Stopa, 2002). Shyness has been studied most often either in the highly com petitive environment of coll ege, where evaluation is central, or in clinical populations w here self-focus is obviously paramount. Descriptions from shy indivi duals of their thoughts, f eelings, and behaviors in the church setting were compared with descriptions in previous research. Comparison of participant responses with existing theory, which is based upon data collected in traditional settings, such as the university, helped gauge the extent of conver gence with theory. Results have suggested that, for indivi duals with high levels of shyness, shyness manifests itself in the same way in an environment different than the ones utilized in most previous research We can continue research with a greater degree of confidence that existi ng theory is robust and that it is generalizable to many settings. This information should help inform future research in this area and assist with efforts to alleviate what is a significant social problem for approximately 40% to 50% of the population at some time in the life span (Carducci, 2000; Zimbardo et al., 1974). Furthermore, results of this study lend support to the utility of urging shy i ndividuals, in therapy or when utilizing self-help methods to overcome shyness, to learn to focus their attention on the task at hand rather than on t he self when entering a social situation. For all levels of shyness, nearly all of the reported thoughts when involved in two specific tasks (attending worship servic es and when performing a volunteer job) were focused on the task or on other individ uals, rather than self. It is the focus

PAGE 198

187 on self when entering a social situation that sets the cycle of negative thoughts and feelings described in the Clark and Wells (1995) model. The field of education has benefitted fr om the current study because shy behaviors can significantly interfere with performance in school (Collins, 1996). This study suggests that for all levels of shyness, shifting the focus of attention from self to others or the task at hand did help individuals exhibit less shy behaviors and have more positive thoughts, both of which should benefit shy students. As Collins noted, being able to participate in discussion and feeling free to ask questions are important in educ ation. Shy students of all ages could be particularly encouraged to change their focus of attention from self to the task at hand or others. It is hypothesized that doing so might help reduce anxiety and enhance learning. As mentioned previously and displayed in Table 41, two salient characteristics of the sample should be k ept in mind when considering results. A large percentage of survey respondents were older (51% were 50 or more years of age, with 77% being 40 years of age or more) and were long-time members of Church A (46% had been members for five or more years). It seems likely that these two findings account for at leas t some of the disparity between mean scores on the BFNE-S in this study compared to previous research and the expected difference between f ear of negative evaluation inside and outside the church setting. Research indicates that most individuals experience a period in their lives when they are shy (Carducci, 1999). Further, “many of those who are not shy now report being shy at some time in the past” (Zimbardo, 1977, p. 5). It

PAGE 199

188 is hypothesized that had the sample in cluded more people under 40 years of age, there would have been a larger percentage of individuals reporting moderate or high levels of shyness, although the difference between shyness inside and outside the church might have remained similar. In the sample utilized in the current study age was negatively correlated with levels of shyness. That is, older individuals tended to reported lower levels of shyness than did younger individuals in this sample. Specifically, the correlation between the BFNE-S, General, and age was r = -.19, p < .01. The correlation between the BFNE-S, Context-specific, and age was r = -.22, p < .01. Regarding the effect of membersh ip on results, when one has been a member of any organization for five or mo re years, one has presumably made at least a few relatively close friends and many acquaintances and has attained a certain level of comfort when participating in activities of that organization. The relationship between length of member ship and comfort outside the church setting was not statistically significant ( r = .06, p > 05), nor was the relationship between length of membership and comf ort inside the church setting ( r = .09, p > .05) In contrast, the relationships betw een length of membership and number of acquaintances and number of close friends we re, however, statistically significant ( r = .30, p < .01 and r = .37, p < .01, respectively ). Again, had the sample included a larger segment of individuals who were not members and/or who had been attending regularly for only a short per iod of time, the findings might have been more similar to those that were expec ted. For example, if more of the participants had been non-members who had been attending only a few weeks,

PAGE 200

189 there might have been more of a differenc e between perceived levels of comfort and acceptance inside and outsi de of the church. The foregoing limitations of sample characteristics notwithstanding, the current study has provided additional evidence regarding the psychometric properties of one of the most commonly employed measures of shyness, the BFNE-S, in a setting not utilized in previ ous studies. The confirmatory factor analysis exhibited an acceptable level of fi t for the one-factor model of shyness. Finally, analysis of participant responses regarding their thoughts, feelings, and behaviors in situations known to be difficu lt for shy people has provided evidence supporting one of the most well-researched theories of shyness, the Clark and Wells (1995) model of social phobia. The study has also provided implications and suggestions for future research into a significant social pr oblem that affects between 40% and 50% of the popul ation at some time in the life span (Carducci, 2000; Zimbardo et al., 1974).

PAGE 201

190 References American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders – DSM-IV (4th ed.). Washington, DC.: Author. Antony, M. M., Orsillo, S. M., & Roemer, L. (2001). Practitioner’s guide to empirically based measures of anxiety. New York: Kluwer Academic/Plenum Publishers. Ary, D., Razavieh, A., Sorensen, C., Ja cobs, L. C., & Sorensen, C. K. (2005). Introduction to research in education (7th ed.). Belmont, CA: Wadsworth. Babbie, E. (2004). The practice of social research. Belmont, CA: Wadsworth. Beck, A. T., Emery, G., & Greenberg, R. L. (1985). Anxiety disorders and phobias: A cognitive perspective New York: Basic Books. Beck, A. T., Emery, G., & Greenberg, R. L. (1996). Cognitive therapy for the evaluation anxieties. In C. G. Lindemann (Ed.), Handbook for the treatment of the anxiety disorders (pp. 235-260). Northvale: Aronson. Berube, M. S. (Ed.). (1982). American Heritage Dictionary (2nd college ed.). Boston: Houghton Mifflin Company. Boomsa, D. I., & Plomin, R. (1986). Heart rate and behavior of twins. MerrillPalmer Quarterly, 32 141-151. Bordens, K. S., & Abbott, B. B. (2004). Research design and methods: A process approach. (6th ed.). Boston: McGraw-Hill. Boyatzis, R. E. (1998). Transforming qualitative information. Thousand Oaks, California: Sage. Briggs, S. R., Cheek, J. M. & Jones, W. H. (1986). Introduction. In W. H. Jones,

PAGE 202

191 J. M. Cheek, & S. R. Briggs (Eds.) Shyness: Perspectives on research and treatment (pp. 1-14). New York and London: Plenum Press. Briggs, S. R., & Smith, T. G. (1986). The measurement of shyness. In W. H. Jones, J. M. Cheek, & S. R. Briggs (Eds.) Shyness: Perspectives on research and treatment (pp. 47-62) New York and London: Plenum Press. Brown, T. A. (2006). Confirmatory factor analysis for applied research New York: Guilford Press. Bruch, M. A., & Cheek, J. M. (1995) Developmental factors in childhood and adolescent shyness. In R. G. Heimberg, D. A. Hope, M. R. Liebowits, & F. R. Schneier (Eds.), Social phobia: Diagnosis, assessment, and treatment ( pp. 163-182). New York: Guilford Press. Bruch, M. A., Gorsky, J. J. Collins, T. M., & Berger, P. A. (1989). Shyness and sociability reexamined: A multicomponent analysis. Journal of Personality and Social Psychology, 57 904-915. Buss, A. H., & Plomin, R. (1984). Temperament: Early developing personality traits Hillsdale, NJ: Erlbaum. Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin 56 81-105. Carducci, B. J. (1999). Shyness: A bold new approach New York: HarperCollins. Carducci, B. J. (2000). W hat shy individuals do to cope with their shyness: A content analysis. In W. R. Crozier (Ed.). Shyness: Development, consolidation, and change (pp. 171-185) London: Routledge.

PAGE 203

192 Carducci, B. J., & Clark, D. L. (1999). The personal and sit uational pervasiveness of shyness: A replication and extensi on of the Stanford Survey on shyness 20 years later. Unpublished paper, Indiana University Southeast Shyness Institute. Carducci, B. J., & Zimbardo, P. (1997). Are you shy? In M. H. Davis (Ed.), Annual editions, social psychology 1997/98 (pp. 35-41). Guillford, CT: Dushkin/Brown & Benchmark. Carleton, R. N., Collimore, K. C., & Asmundson, G. J. G. (2007). Social anxiety and fear of negative eval uation: Construct valid ity of the BFNE-II. Journal of Anxiety Disorders 21 131-141. Carver, C. S., & Sc heir, M. F. (1996). Perspectives on personality (3rd ed.). Boston, MA: Allyn & Bacon. Cheek, J. M., & Buss, A. H. (1981). Shyness and sociability. Journal of Personality and Social Psychology 41 330-339. Cheek, J. M., & Briggs, S. R. (1990). Sh yness as a personality trait. In W. R. Crozier (Ed.). Shyness and embarrassment: Perspectives from social psychology (pp. 315-337). Cambridge, England : Cambridge University Press. Chen, X. (2000). Social and emotional dev elopment in Chinese children and adolescents: A contextual cross-cult ural perspective. In F. Columbus (Ed.), Advances in psychology research ( Vol. 1, pp. 229-251). Hauppauge, NY: Nova Science. Clark, D. M. (2001). A cognitive perspective on social phobia. In W. R. Crozier &

PAGE 204

193 L. E. Alden (Eds.). International handbook of so cial anxiety: Concepts, research, and interventions rela ting to the self and shyness (pp. 404-430). New York: Wiley. Clark, D. M., & Wells, A. (1995). A cogniti ve model of social phobia. In R. Heimberg, M. Liebowitz, D. A. Hope, & F. R. Schneier (Eds.), Social phobia: Diagnosis, assessment and treatment (pp. 69-93). New York: Guilford Press. Collins, J. (1996). The quiet child London: Wellington House. Collins, K. A., Westra, H. A., Dozois, D. J. A., & Stewart, S. H. (2005). The validity of the brief version of the Fear of Negative Evaluation scale. Journal of Anxiety Disorders 19 345-359. Collins, K. M. T., Onwuegbuzie, A. J., & Jiao, Q. G. (2007). A mixed methods investigation of mixed methods samp ling designs in social and health science research. Journal of Mixed Methods Research, 1 267-294. Cottle, M. (1999, August 2). Selling shyness. New Republic, 24-29. Cox, B. J., Walker, J. R., Enns, M. W., & Karpinski, D. C. (2002) Self-criticism in generalized social phobia and response to cognitive-behavioral treatment. Behavior Therapy 33, 479-491 Crozier, W. R. (1979). Shyness as a dimension of personality. British Journal of Social and Clinical Psychology, 18 121-128. Crozier, W. R. (Ed.). (1990). Shyness and embarrassment: Perspectives from social psychology Cambridge, England: Cambridge University Press. Crozier, W. R. (Ed.). (2000a). Shyness: Development, consolidation, and

PAGE 205

194 change. London: Routledge. Crozier, W. R. (2000b). Shyness and social relationships: Continuity and change. In W. R. Crozier (Ed.). Shyness: Development, consolidation, and change (pp. 1-21) London: Routledge. Crozier, W. R. (2001). Understanding shyness: Psychological perspectives China: Palgrave. Crozier, W. R., Jr., & Alden, L. E. (Eds.). (2001a). International handbook of Social anxiety: Concepts, research, and interventions relating to the self and shyness. New York: Wiley. Crozier, W. R., & Alden, L. E. (20 01b). Origins and development. In W. R. Crozier & L. E. Alden (Eds.), International handbook of social anxiety: Concepts, research, and interventions relating to the self and shyness (pp. 23-28). New York: Wiley. Deardorff, J., Hayward,C., Wilson, K. A., Bryson, S., Hamme, L. D., & Agras, S. (2007). Puberty and gender interact to pr edict social anxiety symptoms in early adolescence. Journal of Adolescent Health 41 102-104. Denzin, N. K. (1978). The research act: A theoretical introduction to sociological methods. New York: McGraw Hill. Dobbs, J. L., Sloan, D. M., & Karpinski, A. (2007). A psychometric investigation of two self-report measures of emotional expressivity Personality and Individual Differences, 43 693-702.

PAGE 206

195 Duke, D., Krishnan, M., Faith, M., & St orch, E. A. (2006). The psychometric properties of the Brief Fear of Negative Evaluation Scale. Journal of Anxiety Disorders, 20 807-817. Dunlap, W. P., Cortina, J. M., Vaslow, J. B., & Burke, M. J. (1996). Meta-analysis of experiments with matched groups or repeated measures designs. Psychological Methods 1 170-177 Erdfelder. E.. Faul. F. & Buchner. A. (1996). GPOW ER: A general power analysis program. Behavior Research Methods. Instruments, & Computers. 28, 1-11. Eysenck, H. J. (1956). The questionnai re measurement of neuroticism and extraversion. Revista de Psicologia 50 113-40. Glaser, B. G., & Stra uss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research Chicago: Aldine. Gorsuch, R. L. (1983). Factor analysis (2nd ed.). Hillsdale, NJ: Erlbaum. Greenwald, A. G., Pratkanis, A. R., Lei ppe, M. R., & Baumgar dner, M. H. (1986). Under what conditions does theory obstruct research progress? Psychological Review, 93 216-229. Hatcher, L., & Stepanski, E.J. (1994). A step-by-step approach to using the SAS system for factor analysis and st ructural equation modeling Cary, NC: SAS Institute Inc. Hatcher, L. (1994). A step-by-step approach to using the SAS system for univariate and multivariate statistics. Cary, NC: SAS Institute Inc.

PAGE 207

196 Hartman, L. M. (1983). A me tacognitive model of social anxiety: Implications for treatment. Clinical Psychology Review 3 435-456. Heimberg, R. G., & Barlow, D. H. ( 1988). Psychosocial treatments for social phobia. Psychosomatics, 99 27-37. Heimberg, R. G., Hope, D. A., Liebowit s, M. R., & Schneier, F. R. (1995). Social phobia: Diagnosis, assessment, and treatment. New York: Guilford Press. Heinrichs, R. M., Rapee, R. M., Alden, L. E., Bogels, S..G., Hofmann, K., & Sakano, O. (2006). Cultural differences in perceived social norms and social anxiety. Behavior Research and Therapy, 44 1187-1197. Henderson, L., & Zimbardo, P. G. (2001). Shyness as a clinical condition: The Stanford model: In W. R. Cr ozier & L. E. Alden (Eds.) International handbook of social anxiety: Concepts, research, and interventions relating to the self and shyness (pp. 430-448). New York: Wiley. Hofmann, S. G. (2000). Self-focused a ttention before and after treatment of social phobia. Behavior Research and Therapy 38, 717-725. Hsu, L., & Alden, L. (2007). Social anx iety in Chineseand European-Heritage students: The effect of assessment format and judgment s of impairment. Behavior Therapy, 38 120-131. Ingram, R. E. (1990). Self-f ocused attention in clinical disorders: Review and a conceptual model. Psychological Bulletin. 107 156-176 Johnson, R. B., & Turner, L. A. (2003). Data collecti on strategies in mixed

PAGE 208

197 methods research. In A. Tas hakkori & C. Teddlie (Eds.), Handbook of mixed methods in social and behavioral research (pp. 297-319) Thousand Oaks, CA: Sage. Jones, W. H., Cheek, J. M. & Briggs, S.R. (1986). Shyness: Perspectives on research and treatment. New York: Plenum Press. Kagan, J., & Reznick, J.S. (1986). Shyne ss and temperament. In W. Jones, J. M. Cheek, & S. R. Briggs. Shyness: Perspectives on research and treatment (pp. 81-90) New York: Plenum Press. Keller, M. B., Wood, J. J., McLeod, B. D., Sigman, M., We i-Chin H., & Chu, B. C. (2003). The lifelong course of soci al anxiety disorder: A clinical perspective. Acta Psychiatr Scand 108 (Supplement 417), 85-94. Kimble, C. E., & Zehr, D. H. (1982). Se lf-consciousness, information load, selfpresentation, and memory in social situation. The Journal of Social Psychology 118 39-46. Leary, M. R. (1983). A brief version of the Fear of Negative Evaluation Scale. Personality and Social P sychology Bulletin, 9 371-376 Leary, M. R., & Kowa lski, R. M. (1995). Social anxiety. New York: Guilford Press. Leech, N. L., & Onwuegbuzie, A. J. (2007). An array of qualitative data analysis tools: A call for data analysis triangulation. School Psychology Quarterly, 22, 557-584. Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry Beverly Hills, CA: Sage. Maxwell, J. A. (2005). Qualitative research design: An interactive approach (2nd ed.). Newbury Park, CA: Sage.

PAGE 209

198 McDaniel, P. A. (2003). Shrinking violets and Cas par Milquetoasts: Shyness, power, and intimacy in the United States, 1950-1995 New York: New York University Press. McIntosh, G., & Martin, G. (1992). Finding them, keeping them: Effective strategies for evangelism and assimilation in the local church. Nashville, Tennessee: Broadman Press. McNeil, D. W, Ries, B. J., & Turk, C. L. (1995). Behavioral assessment: Selfreport, physiology, and overt behavior. In R. G. Heimberg, D. A. Hope, M. R. Liebowits, & F. R. Schneier (Eds.). Social phobia: Diagnosis, assessment, and treatment (pp. 202-231) New York: Guilford Press. Morgan, D. L. (2007). Par adigms lost and pragmatism regained: Methodological implications of combining qua litative and quantitative methods. Journal of Mixed Methods Research, 1 48-76. Morris, C. G. (1982). Assessment of shyness Unpublished manuscript, University of Michigan. Musa, C. Z., & Lpine, J. P. (2000). Cogniti ve aspects of social phobia: A review of theories and experimental research. European Psychiatry, 15 (1) 59-66. New York International Bible Society. (1978). The Holy Bible, New International Version. Grand Rapids, MI: Zondervan. Nickerson, R. S. (1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2 175-220. Ollendick, T. H., & Hirshfeld-Becke r, D. R. (2002). The developmental psychopathology of social anxiety disorder. Biological Society 51 44-58.

PAGE 210

199 Onwuegbuzie, A. J. (2003a.). Effect sizes in qualitative research: A prolegomenon. Quality and Quantity: Inte rnational Journal of Methodology, 37 393-409. Onwuegbuzie, A. J. (2003b.). Expanding the framework of internal and external validity in quantitative research. Research in the Schools 10 (1), 71-89. Onwuegbuzie, A. J., & Johnson, R.B. (2006). The validity issue in mixed research. Research in the Schools 13 (1), 48-63. Onwuegbuzie, A. J., & Leech, N. L. (2004). Enhancing the interpretation of “significant” findings: The role of mixed methods research. The Qualitative Report, 9 770-792. Retrieved May 1, 2009, from http://www.nova.edu/ssss/QR/QR94/ onwuegbuzie.pdf Onwuegbuzie, A. J., & Leech, N. L. (2005). On becoming a pragmatist researcher: The importance of combin ing quantitative and qualitative research methodologies. International Journal of Social Research Methodology: Theory & Practice 8 375-387. Onwuegbuzie, A. J., Slate, J. R., Leech, N. L., & Collins, K. M. T. (2007). Conducting mixed analyse s: A general typology. International Journal of Multiple Research Approaches 1 (1), 4-17. Onwuegbuzie, A. J., & Teddlie, C. (2003) A framework for analyzing data in mixed methods research. In A. Ta shakkori & C. Teddlie (Eds.), Handbook of mixed methods in soci al and behavioral research (pp. 351-383) Thousand Oaks, CA: Sage.

PAGE 211

200 Orsillo, S. M. (2001). Measures of social phobia. In M. M. Antony, S. M. Orsillo, & E. Roemer (Eds.), Practitioner’s guide to empiri cally based measures of anxiety (pp. 165-187). New York: Kluwer. Patton, M. Q. (1990). Qualitative evaluation and research methods (2nd ed.) Newbury Park: Sage. Pilkonis, P. A. (1977). The behavio ral consequences of shyness. Journal of Personality 45, 596-611. Pilkonis, P. A., & Zimbar do, P. G. (1979). The personal and social dynamics of shyness. In C. E. Izard (Ed.), Emotions in personality and psychopathology (pp. 131-160). New York: Plenum Press. Pines, A., & Zimbardo (1978). The personal and cultural dy namics of shyness: A comparison between Israelis, American Jews, and Americans. Journal of Psychology and Judaiism, 3 81-101. Pollard, C. A., & Henderson, J. G. (1988). Types of so cial phobia in a community sample. Journal of Nervous and Mental Disease 16 440-445. Rapee, R. M. (1995). Descrip tive psychopathology of so cial phobia. In. R. G. Heimberg, M. R. Liebowits, D. A. Hope, & F.R. Schneier (Eds.), Social phobia: Diagnosis, assessment, and treatment (pp. 55-56). New York: Guilford Press. Rapee, R. M., & Heimberg, R. G. (1997). A cognitive b ehavioral model of anxiety in social phobia. Behavioral Research Therapy 35 741-756. Rodebaugh, T. L., Woods, C. M., Thissen, D. M., Heim berg, R. G., Chambless, D. L., & Rapee, R. M. (2004). More in formation from fe wer questions: The

PAGE 212

201 factor structure and item properties of the orig inal and brief Fear of Negative Evaluation Scale. Psychological Assessment, 16 169-181. Rosenthal, R., & Ro snow, R. L. (1975). The volunteer subject. New York: Wiley. Rubin, H. J., & Rubin, I (2005). Qualitative interviewing. The art of hearing data (2nd ed.). Thousand Oaks, CA: Sage. SPSS Inc. (1998). SPSS Base 8.0 for Windows User's Guide SPSS Inc., Chicago, IL. Spurr, J. M., & Stopa, L. (2002). Self -focused attention in social phobia and social anxiety. Clinical Psychology Review. 7 947-975. Stein, M. B. (Ed.). (1995). Social phobia: Clinical and research perspectives. Washington, D.C.: American Psychiatric Press. Tashakkori, A., & Teddlie, C. (1998). Mixed methodology: Co mbining qualitative and quantitative approaches Thousand Oaks, CA: Sage. Tashakkori, A., & Teddlie, C. (2003).The past and future of mixed methods research: From data triangulation to mixed model designs. In A. Tashakkori & C. Teddlie (Eds.), Handbook of mixed methods in social and behavioral research (pp. 671-701) Thousand Oaks, CA: Sage. Tashakkori, A., & Teddlie, C. (Eds.). (2003). Handbook of mixed methods in social and behavioral research. Thousand Oaks: Sage. Trochim, William M. The Research Me thods Knowledge Base, 2nd ed.. Internet WWW page, at URL: (version current as of 10/20/2006).

PAGE 213

202 Trower, P., & Gilbert, P. ( 1989) New theoretical concepti ons of social anxiety and social phobia. Clinical Psychology Review, 9 19-35. Watson, D., & Friend, R. (1969). Measurement of so cial-evaluative anxiety. Journal of Consulting and Clinical Psychology, 33, 448-457. Watson, F. (2009). Shyness in the context of reduced fear of negative evaluation: A mixed methods case study Unpublished doctoral dissertation, University of South Florida. Weeks, J. W, Heimberg, R. G., Hart, T. A., Fresco, D. M., Turk, C. L., Schneier, F. R., & Liebowitz, M. R. (2005). Empirical valid ation and psychometric valuation of the brief fear of negative evaluati on scale in patients with social anxiety disorder. Psychological Assessment, 17 179-190. Wells, A. (2000). Modifying social anxie ty: A cognitive approach. In W. R. Crozier. Shyness: Development, consolidation, and change (pp. 207-226). London: Routledge. Willems, E. P., & Raush, H. L. (1969). Naturalistic viewpoints in psychological research New York: Holt, Rinehart, & Winston. Yin, R. K. (1984). Case study research: Design and methods .. Thousand Oaks, CA: Sage. Yin, R. K. (2003). Case study research: Design and methods (3rd ed.). Thousand Oaks, CA: Sage. Zimbardo, P. G. (1977). Shyness: What it is, what to do about it Reading, MA: Addison-Wesley.

PAGE 214

203 Zimbardo, P. G., Pilkonis, P. A., & Norwood, R. M. (1974). The silent prison of shyness. Palo Alto,CA: Office of Naval Research Technical Report Z-17, Stanford University.

PAGE 215

204 APPENDICES

PAGE 216

205 Appendix A: Personal Concerns and Issues Survey Your help is needed for a research study that will: assist in evaluating th e needs of the congregation, o be part of a doctoral dissertation for a member, and lay the foundation for future research We are asking visitors, attendees and mem bers to take this survey, which has three sections. Section A contains questions about our church that are not part of the dissertation research. Sections B and C, the dissertation re search, will explore how feeling accepted affects relationshi ps at church, especially for people with social anxiety. Your help is vital, even if you hav e no social anxiety at all. Future research will build upon the findings from these th ree sections of the survey. The University of South Florida Instituti onal Review Board requ ires that I keep your study records confident ial. All records will be kept secure in my home office. No names or any other identifyi ng information will be used in any report. The four dissertation committ ee members (professors at the University of South Florida) and the senior Pa stor at ______ will see the resu lts of the study but they will not know which individuals gave the responses. If members of the Instituti onal Review Board or with the Department of Health and Human Services need to see the study re cords, by law, they must keep the records completely confidential. Taking the survey may help you understand more about yourself. There are no known risks in taking the survey but if you experience emoti onal discomfort, you will be given the name of a qualified counselor you may contact. We may publish what we learn from this study. If we do, we will not let anyone know your name or anything else that would let people know who you are. If you have any questions problems regarding this study, please email me at ______________or call me at ____________ (bef ore 8:00 p.m.). You should not feel that t here is any pressure to take part in the study. If you want to take part, please turn the page. Thank you very much for taking time to share your thoughts.

PAGE 217

206 Appendix A: Continued SECTION A Section A is not part of the dissertation research Answers to these questions will be in a separate r eport for the church. 1. Are you a member of t he Leadership Team? (Please circle) Yes. No. 2. How did you hear about this survey? Please check only one option. Staff meeting A staff member A Leadership Team member Sunday School Wednesday night service Sunday morning service Other (Please explain) 3. What could our church do to be a more loving church for you? ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ 4. What is your greatest need wit h which our church could help you? ________________________ __________________ ______________________ _________________________ ____________________ ___________________ __________________ ________________ _________________ _____________ ________________________ __________________ ________________ ______ ________________________ __________________ ______________________ If you wish, you may submit your survey now. We hope you will continue with Section B, which should take about 20 minutes.

PAGE 218

207 Appendix A: Continued SECTION B Instructions : Please answer the questions below thinking about situations in general outside the church There are no right or wrong answers. The goal is to find out how you as an individual feel. Not at all like me Slightly like me Moderately like me Very like me Extremely like me 1 2 3 4 5 Statement Rating 1. I worry about what other people will think of me even when I know it does not make any difference. 2. I am concerned if I know people ar e forming an unfavorable impression of me. 3. I am frequently afra id of other people noticing my shortcomings. 4. I often worry about what kind of impression I am making on someone. 5. I am afraid others will not approve of me. 6. I am afraid that peopl e will find fault with me. 7. Other people’s opinions of me bother me. 8. When I am talking to someone, I worr y about what they may be thinking about me. 9. I am usually worried about w hat kind of impression I make. 10. If I know someone is judging me, it has a big effect on me. 11. Sometimes I think I am too concer ned with what other people think of me. 12. I often worry that I will say or do the wrong things.

PAGE 219

208 Appendix A: Continued SECTION B Continued Instructions : Please answer the questions below thinking about situations in general inside the church There are no right or wrong answers. The goal is to find out how you as an individual feel. Not at all like me Slightly like me Moderately like me Very like me Extremely like me 1 2 3 4 5 Statement Rating 1. I worry about what other people will think of me even when I know it does not make any difference. 2. I am concerned if I know people ar e forming an unfavorable impression of me. 3. I am frequently afraid of othe r people noticing my shortcomings. 4. I often worry about what kind of impression I am making on someone. 5. I am afraid others will not approve of me. 6. I am afraid that peopl e will find fault with me. 7. Other people’s opinions of me bother me. 8. When I am talking to someone, I worry about what they may be thinking about me. 9. I am usually worried about w hat kind of impression I make. 10. If I know someone is judging me, it has a big effect on me. 11. Sometimes I think I am too concer ned with what other people think of me. 12. I often worry that I will say or do the wrong things.

PAGE 220

209 Appendix A: Continued SECTION B Continued Instructions : Please answer the questions below thinking about situations in general outside the church Very Comfortable Comfortable Neither Un comfortable Very Un comfortable 1 2 3 4 5 Situation Rating 1. Being the focus of attention 2. Large groups 3. Small groups 4. Authority figures by virtue of knowledge (int ellectual superiors, experts) or by virtue of role (police, teachers, superio rs at work) 5. Social situations in gener al 6. New interpersonal situations in general 7. Strangers 9. Situations where assertiv eness is required (e.g., when complaining about faulty service in a restaurant) 9. Being evaluated or compared with others (e.g., when being interviewed, when being criticized) 10. An opposite sex group or a mem ber of the opposite sex

PAGE 221

210 Appendix A: Continued SECTION B Continued Instructions : Please answer the questions below thinking about situations in general inside the church Very Comfortable Comfortable Neither Un comfortable Very Un comfortable 1 2 3 4 5 Situation Rating 1. Being the focus of attention 2. Large groups 3. Small groups 3. Authority figures by virtue of knowledg e (intellectual superiors, experts) or by virtue of role (police, teachers, superiors at work) 4. Social situations in gener al 5. New interpersonal situations in general 6. Strangers 7. Situations where assertiveness is requi red (e.g., when complaining about faulty service in a restaurant) 8. Being evaluated or compared with ot hers (e.g., when being interviewed, when being criticized) 9. An opposite sex group or a mem ber of the opposite sex

PAGE 222

211 Appendix A: Continued SECTION B – Continued Instructions : Please answer the questions below thinking about situations in general outside the church Very Accepted Accepted Neither Un accepted Very Unaccepted 1 2 3 4 5 How accepted do you feel . Rating 1. By people in general 2. By friends and acquaintances 3. When you meet someone you do not know 4. By yourself 5. By God Using the same scale, how accepted do you feel in general . 1-5 6. At your place of work/business (I f you are not employed outside the home, please respond based on how you feel when you go into a relatively formal setting, like renewing your driver’s license) 7. With your family at home

PAGE 223

212 Appendix A: Continued SECTION B – Continued Instructions : Please answer the questions thinking about situations in general inside the church Very Accepted Accepted Neither Un accepted Very Unaccepted 1 2 3 4 5 How accepted do you feel . Rating 1. By people in general 2. By friends and acquaintances 3. When you meet someone you do not know 4. By yourself 5. By God

PAGE 224

213 Appendix A: Continued SECTION B – Continued Instructions : Please answer the following questions by circling the option that best describes you. 1. How long have you been a member of this church? Not a member Less than one year 2 to 3 years 4 to 5 years 5 to 10 years 10 to 20 years Over 20 years 2. For how many years of your life have you been a member of any church? Not a member Less than one year 2 to 3 years 4 to 5 years 5 to 10 years 10 to 20 years Over 20 years 3. Approximately how many times a month do you attend worship services? 0 1 to 2 3 to 4 5 to 8 8 to 12 4. Approximately how many ti mes a month do you participate in church activities outside worship services ? 0 1 to 2 3 to 4 5 to 10 5. Approximately how many close fr iends do you have at this church? 0 1 to 2 3 to 4 5 to 10 More than 10 6. Approximately how many acquaintances do you have at this church? 0 1 to 2 3 to 4 5 to 10 More than 10 7. Please circle one option for each of the three categories below. GENDER AGE IN YEARS RACE/ETHNICITY Male Less than 21 Caucasian Female 21 to 30 African American 31 to 40 Hispanic or Latino 40 to 50 Asian 50 to 60 American Indian or Alaskan Native 60 to 70 Multiracial 70 plus

PAGE 225

214 Appendix A: Continued Section B – Continued For the following questions, please circle the correct option 1. Do you consider yourself to be a shy person? Yes No (If no, skip to #4) 2. If yes, have you always been shy? Yes No 3. If you are currently shy, is that in most or only in some situations? Most situations Some situations 4. Was there ever a prior time in y our life when you were shy? Yes No 5. How desirable is it for you to be shy? (Please circle one.) Very undesirable Undesirable Neither Desi rable Very desirable Thank you very much for taking time to answer Section B. We hope you will continue with the next section, which should t ake about 20 more minutes.

PAGE 226

215 Appendix A: Continued SECTION C In the spaces provided, please describe what you most often do, think and feel in the church situations below. Every individual is different and there are no right or wrong answers. Please provide as much detail as you can. Feel free to write on the back or to use an extra sheet of paper if you need more space WORSHIP SERVICES If I were with you in a typical church service, what would I probably see you do? ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ What are some of the t houghts you might be having during a worship service? ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ Please describe how you would be feeling ________________________ __________________ ______________________ ________________________ __________________ ______________________ __________________ __________________ __________________ __________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ __________________ __________________ __________________ __________

PAGE 227

216 Appendix A: Continued SMALL GROUP GATHERINGS (for example, Sunday School class) If I were with you in a small group gather ing (like Sunday Sc hool, what would I probably see you do? ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ What are some of the t houghts you might be having? ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ Please describe how you would be feeling ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________

PAGE 228

217 Appendix A: Continued SECTION C – Continued SOCIAL EVENTS (LIKE WEDDINGS, BABY SHOWERS, HOLIDAY CELEBRATIONS, ETC.) If I were with you at a social even t, what would I probably see you do? ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ _____________________ ___________________ What are some of the t houghts you might be having? ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ Please describe how you would be feeling ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________

PAGE 229

218 Appendix A: Continued CONTACTS WITH FRIENDS AND ACQUAINTANCES AT CHURCH If I saw you with some of your friends and acquaintance, what would I probably see you do? ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ What are some of the t houghts you might be having? ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ Please describe how you would be feeling ________________________ _________________________ _______________ __________________ __________________ ____________________________ __________________ __________________ ____________________________ __________________ __________________ ____________________________ ________________________ _________________________ _______________ __________________ __________________ ____________________________

PAGE 230

219 Appendix A: Continued SECTION C Continued INTERACTION WITH AUTHORI TY FIGURES AT CHURCH (E.G., PASTOR, STAFF, GROUP LEADERS) If I saw you with people in authority at c hurch, what would I probably see you do? ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ What are some of the t houghts you might be having? ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ Please describe how you would be feeling ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________

PAGE 231

220 Appendix A: Continued PLACES OF SERVICE (like being a Sunday School teacher or a greeter) If I saw you doing your designated job at church, what would I see you do? If you do not currently have a designated job, please indicate that. ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ What are some of the t houghts you might be having while doing your job? ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ Please describe how you would be feeling ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________

PAGE 232

221 Appendix A: Continued Is there anything you would like to say about this survey or anything else you would like to say? ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ ________________________ __________________ ______________________ ________________________ _____________________ ___________________ Thank you so much for taking time to sh are your thoughts. It took a great deal of thought to answer the questi ons, and your willingness to help is sincerely appreciated.

PAGE 233

222 Appendix B Formula for Effect Size (Dunlap, Cortina, Vaslow & Burke, 1996)* Effect size = t c is the t -statistic from the dependent or correlated t -test and r is the correlation between the pretest and posttest measures. N is the number of pairs of scores for the group. Dunlap, W.P., Cortina, J.M., Vaslow, J.B., & Burke, M.J. (1996). Meta-analysis of experiments with matched groups or repeated measures designs. Psychological Methods 1 170-177.

PAGE 234

223 Appendix C: Full Text of Writein Responses WORSHIP DO WORSHIP THINK WORSHIP FEEL attentive to what is going on and worshipping. I wish I were closer to God. I wish my family were here with me. Sometimes really into the service and involved in worship. At other times, sad, depressed, and disappointed in my relationship with God. Worried about how to overcome these feelings. During the songs, eyes closed, hands raised. Sometimes jumping or dancing, sometimes quietly praying. Sometimes crying, sometimes shouting. During the part where the word comes forth, I would be listening quietly, with my bible in hand and usually taking notes on the sermon. "God I love you so much and I need you." Pour out your presence on your people." "If we would only catch a glimpse of who he is it would change the way we really worship." It depends on the day and what I am going through. I am usually feeling whatever I have been going through that week. (Excited, discouraged, etc). Many times I am feeling thankful to be in his house and fortunate to be praising him. A lot of times I feel this even when I am happy or when I have tears rolling down my face. I just am thankful I can be there. Following the leader singing, worshiping, reading the Bible Considering God, thinking what my part in His plan I have, maybe drifting to other thoughts sometimes Usually good spirits, but sometimes I can be negative or sullen if I have been experiencing down times or life problems Greet everyone as I come in. Sit down closer to the back of the sanctuary. Read the bulletin just before service begins and greet individuals around me. During the music portion of the service I praise the Lord by clapping my hands, lifting my hands, and focusing on the Lord. I am not a great singer so I would sing softly. I would then engage my mind, body, soul and spirit in the message being delivered. Once the service ends I would greet more individuals and return home. Attempting to hear what the Lord wants to speak to me through the message. What can I do to help others in the congregation. Studying the congregation to learn the age and ethnicity of those that are a part of the church. Ways to improve the assimilation of newcomers. Joyful and contemplative.

PAGE 235

224 Appendix C: Continued Greet others, see that my family members are in their respective places of attendance/service, and give my undivided attention to the person who is teaching or preaching. I pay attention to the words of the songs. In particular, there is much language used that the un-churched do not understand, and, there are terms used that even the churched may not fully understand. So, there is the assumption that the songs are understood just by participation, but I think this is a mistake. I feel very good and church and look forward to attending, and cherish the opportunity for my family to also receive from the services. Not a whole lot. I was raised in a very very strict church which frowned on any type of emotion involved in the worship service. I am not at all accustomed to the way in which a Pentecostal worship service is conducted. That is not to say I do not value the service or am not actively participating in my own way. I am slowly but surely becoming adjusted to the different worship styles. I very much prefer what I have experienced at _____ to what I have experienced at _______ or ________. I often times think of how the words and emotions of the message apply to my life. This sometimes is troubling because as a person that is just starting to walk with God again there are many things I need to change in my life (I am getting back to where God needs me to be though). I also focus on the spirit of the message, not just the literal application of it. I try to let the message take on a life of it's own and allow it to illuminate things in my life that need attention. I would probably feel reserved and a little shy. I would also feel anything from sorrow to joy depending on the message and what it is saying to me. Sing, occasionally raise my hands Sometimes I am intimidated to worship at church because I worry that people are watching and judging. Sometimes I am able to break through and worship and other times I just struggle. If I feel that there is tension at worship time or I'm confused about a comment that gets made, I am more likely to be intimidated to worship freely. Sing, pray, lift my hands and praise God, listen to sermon, talk with a couple of people How I need to come every week to be reminded of God's faithfulness, that I would like to get involved in some kind of group when I have time, so I could feel like a part of the church, I feel pretty comfortable at this church, but I am a little anxious about making a good impression.

PAGE 236

225 Appendix C: Continued sing, pray, praise, clap, listen to the pastor preach. TRY to use sign language. Smile. Praise and worship God. Ask God for forgiveness. “Concentrate, focus and listen -don’t get distracted with handsome brothers and pastor’s wife’s beautiful shoes and outfits.” I should close my eyes, lift my head to the heavens, praise God, and forget everything around me. I’d like to be able to focus solely on God & worship. Sing, pray, raise my hands in worship, sit quietly and listen, read my Bible, read the words on the screens, speak to others sitting around me, write a check and put it in the collection plate, read the bulletin Enjoying singing as a form of worship when the songs are familiar and melodically easy to sing. Enjoying praising and worshipping the Lord. Enjoying the freedom to worship in my own way. Enjoying the sermon and gleaning from it what the Lord has to sa y to me. Enjoying seeing the corporate worship and thinking how pleasing it must be to God. Enjoying the joyful spirit in this church and the desire to make the Holy Spirit feel welcome and able to do His work amongst the congregation. May notice a regular who sits around me missing and hoping and praying he or she is OK. May notice an uneasy spirit or something or someone feeling amiss and will pray. Pray that if there are any unsaved individuals in church the Holy Spirit will convict them of their need for Jesus in their lives and that day will be the day of their salvation. Concern for others in the congregation who are going through trials. Happy to be in the presence of the Lord. Happy to be with Christian brothers and sisters. Intent and interested in hearing the Word. Possibly sad if a fellow Christian is sad. Weepy if the Holy Spirit touches me. Feelings can be mixed throughout any given service, but mostly deeply touched by the work of the Holy Spirit. Singing, talking quietly to God, eyes open at times and closed at others. Thanking God for getting me where I am and how I got here. I come in at one level, get brought down mentally, and then lifted mentally and spiritually. I always feel better when I leave than when I arrive.

PAGE 237

226 Appendix C: Continued Sit quietly, make an occasional comment to the person next to me and be respectful while participating in the worship experience. That the praise and worship and the sermon could each be shorter in general. I think that most people have about a 20 minute attention span and that a shorter service is not hindering to the Spirit, and that the flow of the Spirit takes the service longer (time-wise) less often than we think. I feel there is a lot of emotionalism exhibited by many and that sometimes we cater to those people. Stand when asked too,sing,clap,bow my head, greet people with a smile. Wondering why I am not as close to the Lord as I should be. In a forgiving spirit. and focusing on the Teaching of the pastor. uplifted by the end of the service. Stand, clap, sing softly because I am not a good singer, cry and smile both. How much I love God and how grateful I am for all he has done for me and my family. How I wish I could A mixture, happy, sad, tender, excited, peaceful, mostly just happy. worshipping, praying, thinking wanting forgiveness for my weaknesses, wanting the Lord to be close to me. hungry, sometimes so happy, other times kind of sad that I am not a better worshipper. GROUPS GROUP DO GROUPTHINK GROUP FEEL being attentive to the teacher doing the lesson. These people are so friendly and accepting. Comfortable with this group of people.

PAGE 238

227 Appendix C: Continued Talk to those around me and try to make everyone feel comfortable and like they are part of the group. "I wish these people would know the real me and that the conversation would not be such a surface conversation." "How can I help them where they are at?" "I wish I could share my situation, struggles, etc with them and they could know truly know where I was at in my walk with Christ." "I hope something about this conversation will challenge me to go deeper with God." If I am able to be helping someone with their issues I feel good. However, usually the conversations are just surface and I feel like it is a waste of time or like we are being fake. I feel disconnected and lonely. I tend to sometimes avoid social contact in small groups, I find I do not interact with certain others easily I don't want to really be here, trying to find common ground of discussion, something to talk about Possibly distant if I am avoiding contact, but if I am feeling interested in interacting with others, usually good Engaging those around me in conversation so that I can learn about them. Once the study begins I would be listening intently. How can I help others. Does God want me to share anything that would be of value to the small group setting. Joyful and contemplative Eager to participate, and much wanting to engage others in whatever activity we are doing. An expectation of learning something new about God's Word, and either the strengthen of relationships or the making of new ones. Good.

PAGE 239

228 Appendix C: Continued Attempt to interact a little but standing off to the side or in the back of the group. I try to take the situation in. I would attempt to identify where if at all do I fit in with this group. I would also try to identify anything intelligent I may have to add to the situation. I tend to be more reserved with any more then one or two people present. Content Usually talk to one person at a time or sit quietly while others are talking. I might have something to add to the subject, but I would not readily volunteer to share it unless I really felt that the Holy Spirit was prompting me to. Very nervous about sharing my thoughts. Making conversation with one person at a time, asking people questions I want this person to like me. A little anxious if I don't know the people, less anxious if I do know the people, but I am always a little anxious. Listening more than speaking -unless very comfortable with group and topic. How long will this take? I want to be home --unless very comfortable with group. Anxious.

PAGE 240

229 Appendix C: Continued It would be completely dependent on how well I knew the other members of the small group. If I don't know them, I would be quiet, reserved and generally uncomfortable. If I know them, I would be fairly comfortable, talkative, laughing, possibly touch another on the arm or shoulder, be interested in the others and what they have to say generally outgoing. If I don't know the people in the group, I would be listening to them talk, observing their demeanor and behavior, looking for clues as to the sincerity of the individuals, how they treat one another, etc. If I know the group members I would be relaxed, thinking about the strengths of the different individuals, enjoying the interactions, happy to be included. In a group of people I don't know I'd definitely feel very uncomfortable, unsure, probably unhappy really as I don't particularly like making small talk with people I don't know. In a group of people I know, it would be the complete opposite. I would be happy, relaxed, enjoying the whole thing. Standing or sitting quietly and surveying the room. Being polite and greeting others. Observing others, how they are responding to the person talking. Listening to the speaker. Content. Listen, and contribute. I'm not shy about speaking up to clarify or offer input. It would depend on the small group and what the topic of discussion would be. This would also depend on the people and the topic. Wont' raise my hand to give an answer. Quiet afraid to open up, I would be shy.

PAGE 241

230 Appendix C: Continued Usually sit at the back or edge and not say much. Why don't I ever speak up? I had something good to say but now the time is past to say it and I will look weird Usually pretty uncomfortable if we have to talk to each other much. Fine if we are just sitting and listening. Depends.. If I am with other singles that I know I am talking some and helping if needed. If I am with people I don't know well, you won't see me do much of anything. I kind of blend in the background. If comfortable-I never want it to end. If notcan't want until it's over. If comfortable.. glad, happy If not...uncomfortable, sad, fearful sometimes SOCIAL DO SOCIAL THINK SOCIAL FEEL Talking with others and generally enjoying myself. Happy to be part of the event. Comfortable with the group of people at the gathering. Helping, serving, talking to others. "Why do I feel like I have no true friends here even though I know everyone in this room." "I am thankful for the people in my life but I wish we got together more than just at weddings, holidays, baby showers, etc." I would probably be feeling very thankful for the people in my life yet lonely because we never really hang out outside of these events and I wish they walked through every part of life with me.

PAGE 242

231 Appendix C: Continued I would probably be walking around and trying to visit with others, eating, playing with my computer etc Thinking about what I might talk about with someone to see what they are like Good, maybe looking for something to do Assisting with the set up of the event in any way that I can chatting to everyone as I go. Engaging in conversation with those seated/standing around me. How can I help those around me? What words of encouragement can I speak that would benefit the hearers? Make them feel accepted. Comfortable I would participate in whatever level is required, not stepping out to break the order that has been established by the program. Hoping that the program is followed (e.g., getting done on time). As a man, these are formal necessitates that I don't derive much pleasure from. They have their place, it's just that they are not too exciting -but the memory of it will remain special to me. I'm there. Neutral. Mingle with one or two people. Other then that I would probably be on the side or in the back. Just observing and taking it all in. It really depends on the event. I could be indifferent, engaged and happy, or engaged and anxious ( I tend to get anxious in large groups).

PAGE 243

232 Appendix C: Continued I would be sitting quietly unless there was someone there that I felt comfortable to talk with, mostly if they initiated the conversation. I wish I wasn't so uncomfortable to go up to someone and start a conversation. Intimidated Stick with people that I know and try to find someone to talk to. I am uncomfortable if I don't have anyone to talk to or don't know very many people. I wish I didn't have to be here. Anxious, bored Sit at a table. Try hard to be social. When can I leave to go home. Anxious. I would hope there would be someone there I know and would sit with that person. I would speak to others when spoken to, otherwise sit quietly. I do not like to be singled out, made to go up front, or otherwise have attention focused on me. I would be uncomfortable if there was no one at the function I knew and be thinking I wish it would end so I could leave. Alone and uncomfortable if there was no one I knew at the function. Greet other people. Happy. Polite. Happy.

PAGE 244

233 Appendix C: Continued Socialize. Generally with people I know, but I would try to speak with people that I do not know as well. That I like hanging out with my church friends. Welcomed. Either help with the cooking or clean up after a function. Wondering why I am not in a group fellowshipping. so I stay busy. Kinda depressed, yet I stay busy. I do not go to many just social things because it is hard to make small talk and I feel awkward standing around. I usually stand at the edge of the crowd and look for someone I know. I want to leave as soon as I can, usually I wish I had not come. Lonely and weird. Very nervous and tense. Frustrated with myself. not much unless helping with the event depends on who I'm with and the event. the same .. depends FRIENDS DO FRIENDS THINK FRIENDS FEEL Greeting and talking with that person. How friendly everyone is at this church. Accepted. Helping out in areas of ministry at the church. Busy going or doing something for others. "Did I help everyone that needed me." Did I take care of all my responsibilities?" "Will someone please ask me how I am doing." Needed, wanted, used Visit, say hi, what are you up to, ask how are things Nothing in particular Good

PAGE 245

234 Appendix C: Continued Laughing, talking. What encouraging words can I speak? How can I make this a moment filled with purpose? Joyful, extremely comfortable I would just be there. I probably would not be a wall flower but also would not be the center of attention. My thoughts would probably be on the subject at hand. That depends on what the subject at hand was. I could have a full gamut of emotions at this point. I would be talking to them. fairly comfortable Talking and laughing I love these people, and they love me. Relaxed and happy With friend: talk, laugh, listen, share, enjoy company. With Acquaintances: be polite, listen, agree, smile. Friends: Want to make plans and spend more time with them. Acquaintances: Be polite, and proceed to my destination. Welcome. Speak to them, stand around and talk with them, laugh with them. I would be happy to see my friends and happy to interact with them. Happy Greet everyone, smile. Happy to be with friends. Content. Have fun and joke around. That I enjoy these people's company. Loved. I would be keeping my self busy,I have not fit into a group of clickish people. Do I want too! No Why can't I fit in. Depressed, that's why I keep myself busy. I don't want to show how much it hurts. Talk to them, stand around, usually asking about them rather than talking about me. I am okay with people I know but still a little nervous and always wondering if I said the right thing. With new people, I do better sometimes because the talk is just superficial.

PAGE 246

235 Appendix C: Continued helping, talking, playing around some I want to make closer friends, but I am not sure how to do that. I am shy and struggle with accepting myself as I am. So I have a hard time thinking people really like me or want to be around me. So sometimes I wonder if people will ever really love me or just deal with me. I always happy to be with my friends at church. AUTHORITY DO AUTHORITY THINK AUTHORITY FEEL Talking freely and comfortably with that person. Thinking how much I like and respect that person. Comfortable and accepted. Speaking with confidence. At this church, although I would be confident, I might also be defensive even though I wasn't trying to be. "What am I going to get in trouble for this time? Will what I say be used against me? Do they really trust me? Are they judging me? Why does it always seem like a fight?" Frustrated, fake, angry, hurt Listening, find out what is happening, trying to see if I can be part of getting a problem solved Thinking of ways to get a job done, troubleshooting, problem solving Good Say yes sir, no ma'am. Be talkative, friendly, and respectful. Attempt to learn about them and the church. Am I making an idiot of myself? What are they thinking about me? Cautious and somewhat comfortable Respect and appreciation for their position that they fulfill. Thankful. Acknowledgement. A privilege. I know that our leaders our busy, so just to have a few minutes of their time is precious, so I try to make my moments with them encouraging as I know they have a lot of them in fulfilling their position. Good. You would see me exhibit reverence towards a person of authority at church. I would allow them to sit before I did, walk ahead of me, and essentially just follow their lead. I would probably be focused on the details of the persons words and actions. I often find it interesting to observe what people in authoritative people do in situations. In the business world I believe in the principle of casting the shadow of a leader. In a church role I observe what that shadow looks like. Reverence, happiness, intimidation, and maybe some awe. It depends on the situation. I would be comfortable talking with them if they weren't trying to be controlling and it was a friendly encounter. I would probably be worrying about what kind of impression I was making. Comfortable as long as there was no strife.

PAGE 247

236 Appendix C: Continued Be very nice and a little shy This person is better than me (although I know they are not), and I need their approval. Anxious Be respectful. Ask questions. Comfortable --not intimidated. How can I show my appreciation for their hard, intense work. Comfortable. Speak, chat briefly Happy to see the individual. The more contact I had had with the individual, the more comfortable I would be. Greet them with respect. Glad to be in their presents. Welcome, accepted. Speak with them respectfully and honestly. It would depend on the situation. I am not confrontational, so it would take something unusual to get me to "rock the boat". It depends on the person. The less direct involvement I have with the person, the less likely I am to think that my input will make a difference in the decision making process. Listen, may say something casual and respond with a few words. None Okay, at this point as long as they are not degrading me I would ask whatever I needed to and then say thank you and probably walk away. I do not want to bother them. They probably do not even know my name. Nervous and not very comfortable. not much.. I don't talk much to people I don't know well. Not sure Not sure JOB DO JOB THINK JOB FEEL Teaching Sunday School. Most of the time I felt uncomfortable in front of the class leading the discussion even though I felt I was prepared. Anxious and uncomfortable. Working with people, interacting with them, praying with them, giving them answers, helping lead them to the right resources, etc. I love what I do. Happy, thankful for the opportunity to help out, valued Coordinating events, administration, helping with set up, greeting individuals, teaching, praying, connecting newcomers to the church. Am I doing everything as God has intended me to do it? How can I do it better more effectively? Will it make a difference in the lives of others? Joyful, focused, stressed (sometimes), fulfilled. Full bore. Serious when the time calls for it, and cutting up when the time is appropriate. Engaged! Reaching the goal by the prescribed time; satisfying the action items. Fulfilled, and happy to being doing my part. Satisfaction comes from completion of the tasks/project.

PAGE 248

237 Appendix C: Continued Be attentive to the needs of the people I was serving. I would be focused on the task at hand. Dutiful Working in the kitchen I would enjoy serving. Happy to be able to help. I taught a class of teenage girls, and I loved leading a discussion with them. I can make a difference in these young lives. I am so thankful to have this opportunity. A little nervous, but fulfilled. Greeting. Hope I can make all welcome, comfortable and make them smile. Happy. Interacting with individuals, depending on the situation there could be laughter or tears, touching the person on the shoulder or arm. Concern for the individual and what he/she might be experiencing. What can I do to help? Careful not to do or say the "wrong" thing. Depends on the situation possibly happy or possibly sad, possibly comfortable or possibly uncomfortable. Something with my hands and using my life experiences. Glad I could help someone. Happy. Interact with people. Follow instructions. Offer my input when it is appropriate. That we need more people that see service as something that needs to be put into action. That there is a lot more potential for this ministry to grow than what we are experiencing. Be a blessing to the children, none just a joy to be helping with the children Greeting people at the door as they come in. I hope they like our church and we make the feel comfortable. Fairly comfortable because I have done this for a while. Singing.. I'm in the choir, and helping where needed.. I am part of the singles leadership team. When I sing I try to picture the Lord standing in front of me, it helps me focus on Him and to feel close to Him. when singing.. I love it, joy ,peace ,closeness to the Lord. I never want the music to stop.

PAGE 249

238 Appendix D. Num ber and Percent of Missing Data Research Question Analysis Missing Data for BFNEGeneral Missing Data for BFNEContext-specific (1) What are the psychometric properties of the BFNE-S, General and Context-specific, i.e., the church? (A) Descriptive statistics and Cronbach alpha for scores from the BFNE-S, General and Contextspecific. (B) Confirmatory factor analysis for scores from the BFNE-S, General and Context-specific. (A) Descriptives – 13 (5%) Cronbach – 14 (6%) (B) CFA, 1-factor and CFA, 2-factor – 40 (17%) (A) Descriptives – 7 (3%) Cronbach – 8 (4%) (B) CFA, 1-factor and CFA, 2-factor – 40 (17%) (2) What is the difference in perceived fear of negative evaluation in the church setting compared to the non-church setting? A dependent samples t -test to examine differences in means between BFNE-S, general and contextspecific versions 18 (8%) 18 (8%) (3) What is the difference in perceived fear of negative evaluation in the church setting compared to the non church setting for males and females? A repeated measures ANOVA to compare perceived FNE for males and females in the church setting compared to the non church setting. 40 (17%) 40 (17%) (4) What is the difference in perceived fear of negative evaluation in the church setting compared to the non church setting for individuals of different races? A repeated measures ANOVA to compare perceived FNE for Caucasians, African Americans, Hispanics, and persons of multiracial background in the church setting compared to the non church setting. 37 (15%) 37 (15%)

PAGE 250

239 Appendix D (Continued) (5) What is the difference in perceived acceptance between people inside and outside the church setting? (A) Descriptive statisti cs for the items on the perceived acceptance checklist, outside and inside the church setting. (B) A dependent samples t -test for the difference in levels of perceived acceptance by people inside and outside the church setting. (A) 9 (4%) B) 7 (3%) (A) 6 (3%) (B) 7 (3%) (6) What is the difference in self-reported levels of comfort outside the church setting and inside the church setting? A dependent samples t -test of the difference between mean levels of comfort in the general setting and in the Context-specific setting (10 situation checklist) 19 (8%) 16 (7%) (7) To what extent do Context-specific issues relate to self-reported levels of fear of negative evaluation? A multiple regression for fear of negative evaluation using length of membership, regularity of attendance, number of activities participated in per month, and number of close friends as predictor variables 21 (9%) 21 (9%) (8) How do shy people typically think, feel, and behave in an environment hypothesized to have less fear of negative evaluation and self-focus? Percentage of thoughts, feelings, and behaviors for individuals with high levels of shyness in the church setting that were consistent with the Clark and Wells (1995) model. 0 0 (9) To what extent is selfreported fear of negative evaluation associated with attentional focus upon self and negative quality of thought in the six church situations? (A) One-way ANOVA for focus of thought data and scores from BFNE-S, Context-specific. (B) One-way ANOVA for quality of thought data and scores from BFNE-S, Context-specific. 0 0

PAGE 251

ABOUT THE AUTHOR Freda Watson received a Bachelor’s Degree in Psychology from the University of South Florida in 1999 and an M.A. in Curriculum and Instruction, with an Emphasis in Measurement and Resear ch, from the University of South Florida in 2002. She entered the Ph.D. program at the Univ ersity of South Florida in 2002. While in the Ph.D. program, Ms. Wats on was active in research. She has coauthored several publications and m ade several paper presentations at national and regional educational research associati ons. Her primary research interests are shyness, Christian growth families in poverty, and mixed methods research.