USF Libraries
USF Digital Collections

Antecedents and consequences of emotional dissonance

MISSING IMAGE

Material Information

Title:
Antecedents and consequences of emotional dissonance understanding the relationships among personality, emotional dissonance, job satisfaction, intention to quit and job performance
Physical Description:
Book
Language:
English
Creator:
Diamond, Laurie K
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Emotional labor
Turnover
Ipip
Debt collectors
Helpfulness
Dissertations, Academic -- Psychology -- Doctoral -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The primary goal of this research was to explore the antecedents and consequences of emotional dissonance for debt collectors. The antecedents were personality factors (extraversion, anger, conscientiousness and agreeableness) and pro-social factors. The consequences of emotional dissonance were job satisfaction, intention to quit and job performance. A path model was developed to explain the relations among the studys measures in a sample of 188 full-time debt collectors. The path analysis results failed to show strong relations between personality and emotional dissonance. However, strong relations were found between emotional dissonance, job satisfaction, intention to quit, and performance. Job satisfaction acted as a mediator between emotional dissonance and intention to quit as well as emotional dissonance and performance.
Thesis:
Thesis (Ph.D.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Laurie K. Diamond.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 115 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 - 001681137
oclc - 62865813
usfldc doi - E14-SFE0001044
usfldc handle - e14.1044
System ID:
SFS0025365:00001


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 Ka
controlfield tag 001 001681137
003 fts
005 20060215071309.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 060110s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001044
035
(OCoLC)62865813
SFE0001044
040
FHM
c FHM
049
FHMM
090
BF121 (Online)
1 100
Diamond, Laurie K.
0 245
Antecedents and consequences of emotional dissonance
h [electronic resource] :
b understanding the relationships among personality, emotional dissonance, job satisfaction, intention to quit and job performance /
by Laurie K. Diamond.
260
[Tampa, Fla.] :
University of South Florida,
2005.
502
Thesis (Ph.D.)--University of South Florida, 2005.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 115 pages.
Includes vita.
3 520
ABSTRACT: The primary goal of this research was to explore the antecedents and consequences of emotional dissonance for debt collectors. The antecedents were personality factors (extraversion, anger, conscientiousness and agreeableness) and pro-social factors. The consequences of emotional dissonance were job satisfaction, intention to quit and job performance. A path model was developed to explain the relations among the studys measures in a sample of 188 full-time debt collectors. The path analysis results failed to show strong relations between personality and emotional dissonance. However, strong relations were found between emotional dissonance, job satisfaction, intention to quit, and performance. Job satisfaction acted as a mediator between emotional dissonance and intention to quit as well as emotional dissonance and performance.
590
Adviser: Michael Brannick.
653
Emotional labor.
Turnover.
Ipip.
Debt collectors.
Helpfulness.
690
Dissertations, Academic
z USF
x Psychology
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1044



PAGE 1

Antecedents and Consequences of Emotional Dissonance: Understanding the Relationships Among Personality, Emotional Dissonance, Job Satisfaction, Intention to Quit and Job Performance by Laurie K. Diamond A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Michael T. Brannick, Ph.D. Walter C. Borman, Ph.D. Judith Becker Bryant, Ph.D. Doug Rohrer, Ph.D. Paul E. Spector, Ph.D. Date of Approval: March 22, 2005 Keywords: emotional labor, turnover, IPIP, debt collectors, helpfulness Copyright 2005, Laurie K. Diamond

PAGE 2

Dedication To my wonderful husband Mitch, you have stood by me at my very worst and still maintained that I was wonderful. You still remember where the Yellow River is even though this information has been lost to me for years! You held me up when I couldn’t find my feet, stood behind me when I could find my feet and have always been the guiding light that I look for every morning. Even when you were diagnosed with a brain tumor and facing death head-on you told me that no matter what I was going to continue and finish my degree. For your selflessness and faith, you have my everlasting love, respect and admiration. For my children, Andrew, Lauren, Kevin and Tori, who helped me study when I needed to and never complained when my work came before their personal wants, I thank you. You four have been the best gifts I could ever have received. I am blessed to have each and every one of you. Thanks for giving of yourself so that I could do something that was important to me. Remember, no matter what, you still come first. To my family, Mom, Dad, Kim and Cheryl, I would never have made it this far if it wasn’t for all of you lending the emotional support that you did over the years. How lucky I am to have such an incredible family. Finally, to my friends, Erica and Melanie what more can I say than I love you and watch out world, because now there are three incredible female doctors out there! The third one (me) would never be out there without your unwavering friendship.

PAGE 3

Acknowledgements To Dr. Brannick who has never waivered in his belief that I could become the person that I am today. Your ability of knowing when to push and when to sit back has allowed me to succeed when I didn’t think it was possible. Your tirelessness in reading and editing drafts is beyond compare. Allowing me the opportunity of working with you for many long years is a gift that I will never forgot and debt I will never be able to repay. I am honored to have been able to work with you. I truly hope that one day, all the hair that I have caused you to pull out will grow back. Thanks for pacing with me when I needed to explain another mystifying idea and sharing my quirky sense of humor with me. One day I hope to be able to call you Mike. To Dr. Bryant, thank you for being by my side for all those years from undergrad, to getting my master’s to finishing my doctorate, you have no idea what your presence at my side has meant to me all these years. You mean more to me than you can possibly imagine. Finally to Dr. Spector, thank you for pushing me so hard even when I didn’t want to move one more inch. You saw something in me that even I was able to see until after you pushed me over the edge and the view was so much better from the other side. Your insight is much appreciated. Thank you for helping me craft a dissertation I can be proud of.

PAGE 4

i Table of Contents List of Tables iii List of Figures iv Abstract v Introduction 1 Development and Theoretical Perspectives of Emotional Labor/Emotional Dissonance 2 Empirical Studies of Emotional Dissonance 21 Limitations of Past Empirical Research 23 Current Emotional Dissonance Research 24 Personality as Antecedents of Emotional Dissonance 28 Extraversion/Introversion 28 Emotional Stability/Neuroticism 30 Agreeableness or Likeability 32 Conscientiousness 33 Openness to Experience 34 Pro-Social Behavior Helpfulness 34 Consequences of Emotional Dissonance 38 Job Satisfaction 38 Intention to Quit 40 Performance 41 Gender Differences 43 Summary 46 Method 47 Participants 47 Measures of Antecedent Variables 49 Pro-Social (Helpfulness) Behaviors 49 Five Factor Personality 50 Dependent Measures 51 Emotional Dissonance 51 Job Satisfaction 54 Intention to Quit 54 Performance 54 Procedure 55

PAGE 5

ii Results 57 Data Analysis 57 Tests of Hypotheses with Path Analysis 63 Gender Differences 65 Post-Hoc Gender Analyses 71 Discussion 72 References 85 Appendices 94 Appendix A: Helpfulness Measure 95 Appendix B: IPIP 96 Appendix C: DEELS 98 Appendix D: Means, Standard Deviations and Correlations between all Variables and Emotional Dissonance, Faking Emotions and Suppressing Emotions 100 Appendix E: Nature of Work Measure 101 Appendix F: Intention to Quit Measure 102 Appendix G: Informed Consent 103 Appendix H: Demographic Questionnaire 106 About the Author End Page

PAGE 6

iii List of Tables Table 1 Participant Pool Demographic Information 48 Table 2 Means, Standard Deviations, Alphas and Comparisons of Measures 50 Table 3 Frequency and Percentage of Missing Data 58 Table 4 Means, Standard Deviations and Correlations between all Variables of Interest 59 Table 5 Regression Analyses 60 Table 6 Female Means, Standard Deviations and Correlations between all Variables of Interest 68 Table 7 Male Means, Standard Deviations and Correlations between all Variables of Interest 69 Table 8 Post-Hoc Gender Demographic Comparisons 72

PAGE 7

iv List of Figures Figure 1. Morris and Feldman (1997) Antecedents and Consequences of Emotional Labor Model 11 Figure 2. Grandey (2000) Emotional Labor Model 14 Figure 3. Glomb and Tews (2004) Emotional Labor Model 18 Figure 4. Proposed Path Model including Hypotheses and Proposed Relationship Directions 27 Figure 5. Path Model including Path Coefficients 62 Figure 6. Path Model including Path Coefficients for Females 66 Figure 7. Path Model including Path Coefficients for Males 67

PAGE 8

v Antecedents and Consequences of Emotional Dissonance: Understanding the Relationships Among Personality, Emotional Labor, Job Satisfaction, Intention to Quit and Job Performance Laurie K. Diamond ABSTRACT The primary goal of this research was to explore the antecedents and consequences of emotional dissonance for debt collectors. The antecedents were personality factors (extraversion, anger, conscientiousness and agreeableness) and prosocial factors. The consequences of emotional dissonance were job satisfaction, intention to quit and job performance. A path model was developed to explain the relations among the study’s measures in a sample of 188 full-time debt collectors. The path analysis results failed to show strong relations between personality and emotional dissonance. However, strong relations were found between emotional dissonance, job satisfaction, intention to quit, and performance. Job satisfaction acted as a mediator between emotional dissonance and intention to quit as well as emotional dissonance and performance. In addition, gender differences were explored, and it was found that the relations between emotional dissonance and job satisfaction were weaker for females than for males. Implications and recommendations for future research in this area are discussed.

PAGE 9

1 Introduction In the past 20 years, researchers have been increasingly interested in the role of emotions within the workplace (Abraham, 1999; Cote & Morgan, 2002; Wharton, 1993). Of particular interest for researchers has been the role of the management of one’s emotions in the workplace or what is typically called emotional labor (Hochschild, 1983). Recent research has shown that the appropriate management of emotions within the workplace is important for many different types of jobs (Diefendorff & Richard, 2003; Grandey, 2000; Morris & Feldman, 1996) although typically the jobs actually studies are all confined to the service sector. With the service sector growing rapidly, currently accounting for 59.2% of all jobs (Schnaue, 2005) in the United States alone, the topic of emotional labor and emotional dissonance has become increasingly important. More specifically, researchers want to know what constitutes emotional labor and emotional dissonance, how emotional dissonance should be measured, and the antecedents and consequences of emotional dissonance. Because emotional labor and emotional dissonance research is st ill in the beginning stages, there are no definitive answers to these important issues. I begin with a review of the emotional labor theories to date, the development of the emotional dissonance construct, the empirical studies completed to date on emotional dissonance, and the limitations of those studies. I will then discuss the contribution to the literature that this study provides in both confirming and extending past research.

PAGE 10

2 Development and Theoretical Perspectives of Emotional Labor/Emotional Dissonance In 1983 Hochschild published her seminal book entitled “The Managed Heart.” Dr. Hochschild, a sociologist, based her ground-breaking book largely on her study of flight attendants. Through her research, Dr. Hochschild found that the emotions displayed by many flight attendants were different than the emotions the flight attendants were actually experiencing. For example, flight attendants were required to smile at their passengers, regardless of whether they felt like smiling or if the passenger was not being agreeable. It was this discrepancy that led Dr. Hochshild to coin the term “emotional labor”. The definition supplied by Dr. Hochschild was “the management of feeling to create a publicly observable facial and bodily display for a wage” (p.7). While experiencing conflicting emotions was common from a personal level, this was the first time that the issue of conflicting emotions due to organizational demands was presented. Dr. Hochschild termed this change as “the commercialization of human feeling.” Hochschild (1983) elaborated on the range of experienced emotional labor by likening it to the toe and the heel of a human foot. In Hochschild’s words, “the corporate world has a toe and a heel, and each performs a different function: one delivers a service, the other collects payment for it” (p. 137). Hochschild further elaborated that the “toe” (customer service) has to feel trust and enthusiasm, while the “heel” (collection) has to feel distrust and ill will. The two types of workers have to manage extreme but opposite emotions in order to work effectively. According to Dr. Hochschild, the conflict between

PAGE 11

3 felt emotion and the required organizational emotional display resulted in emotive dissonance. Emotive dissonance is similar to cognitive dissonance in that a person will experience discomfort when there is a mismatch between actual felt emotion and displayed emotion and seek ways in which to bring the discrepancy into balance. Hochschild posited two different avenues that people may use to reduce emotive dissonance: surface acting and deep acting. Surface acting focuses on outward behavior and Hochschild likened surface acting to method acting. In surface acting, the “actor” is not actually experiencing the displayed emotion, he or she is simply portraying the emotion for the audience to see and interpret. Surface acting, according to Hochschild, is “the art of an eyebrow raised here, an upper lip tightened there” (p.38). Deep acting, on the other hand, focuses on inner feelings and refers to attempts to internalize the required emotion so that the natural, unaffected display will be congruent with the role requirement. Hochschild discussed deep acting as having two different avenues. The first avenue in deep acting involves a person directly willing the desired emotion. The second avenue in deep acting involves a person using their imagination to create thoughts, images, and personal memories to induce the required emotion. An example of this would be thinking of a happy event, such as getting married, having a child, or even winning the lotto to create a feeling of happiness. Hochschild believed that emotional labor had negative consequences. Theoretically, Hochschild posited that there were three different scenarios, all with different consequences. First, one type of employee identifies too much with the

PAGE 12

4 emotional demands of the job. Over time, Hochschild believed that this type of individual would suffer from burnout, stress and depersonalization. The second type of employee distinguishes themselves from the job itself. While Hochschild believes that this type of behavior will reduce the burnout, it may make the employee feel “phony” (p. 188), because at any time they may feel that they are either overacting or underacting. The third type of employee is one who distinguishes themselves from the emotional display itself. This type of employee realizes that the display is just an act and Hochschild believed that this type of employee may suffer from estrangement and withdrawal behaviors. To summarize, Hochschild’s position was that emotional labor occurred any time that a person had to manage their feelings to create a publicly observable facial and bodily display for a wage. According to Hochschild, emotive dissonance occurred when there was a discrepancy between actual feeling and displayed feeling. Emotive dissonance would then require a person to employ emotion management techniques to lessen the dissonance. The emotion management techniques that Hochschild presented were surface acting and deep acting. In addition, Hochschild believed that the performance of emotional labor would result in negative consequences. Although Hochschild’s work was ground-breaking, it was not without limitations. For example, Hochschild did not develop a system of measuring emotional labor. Instead, Hochschild presented a list of 15 different occupations that involved “substantial amounts of emotional labor” (p.245).

PAGE 13

5 In 1993, Ashforth and Humphrey published their theory regarding emotional labor. Ashforth and Humphrey reviewed and expanded on Hochschild’s (1983) theory by focusing on the behavior itself rather than the underlying emotions. The rationale for taking a behavioral view of emotional labor was based on their belief that it is the actual behavior which is observed and directly affects customers and clients and that people are able to conform with display rules without having to “manage” them. The use of the term “display rules” was argued by Ashforth and Humphrey to be a more relevant term than “feeling rules” which Hochschild used. Ashforth and Humphrey drew upon the Rafaeli and Sutton (1989) research in defining display rules as a function of societal, occupational or organizational norms that dictate which emotions should be displayed during a service encounter. Based on this position, Ashforth and Humphrey’s definition of emotional labor was, “the act of displaying the appropriate emotion, i.e., conforming with a display rule” (p.90). Therefore, according to these researchers, emotional labor was the actual display of appropriate emotions. Using this definition, the researchers argued that emotional labor was performed through surface acting, deep acting or by the expression of genuinely felt emotion. This research was one of the first to incorporate genuine emotion into the emotional labor construct. These researchers argued that the problem with past conceptions of emotional labor was that they did not allow for the genuine expression of emotion. For example, a nurse may genuinely feel sympathy for an injured child and therefore does not need to use surface or deep acting to comply with the display rule.

PAGE 14

6 Ashforth and Humphrey believed that theoretically emotional labor could result in both positive and negative consequences. As for the positive consequences, Ashforth and Humphrey believed that if the expressed emotion was perceived by the audience as sincere, then compliance with the display rules would be positively associated with task performance. In addition, if employees were given more latitude for self-expression in the performance of emotional labor, then increased personal well-being would follow. As for the negative consequences, Ashforth and Humphrey believed that emotional labor was a “double-edged sword” (p. 95) and that employees who comply with desired display rules may prime customer expectations that can not be met. In addition, these researchers believed that performing emotional labor would result in emotive dissonance and selfalienation. However, using the social identity theory, these researchers believed that social identity would moderate the relationships between emotional labor and both positive and negative consequences. In summary, the Ashforth and Humphrey theory was behaviorally oriented and introduced genuine expression of emotion as a way of performing emotional labor in addition to surface and deep acting. This theory posited emotional labor to be considered as a form of impression management. The performance of emotional labor was theorized to result in both positive and negative consequences. Following Ashforth and Humphrey (1993), Morris and Feldman published their theory of emotional labor in 1996. According to Morris and Feldman, emotional labor was “the effort, planning, and control needed to express organizationally desired emotion during interpersonal transactions” (p. 987). Morris and Feldman argued that emotional

PAGE 15

7 labor was multi-faceted and was based on an interactionist model of emotion. According to these researchers, the majority of the previous research done on emotional labor used frequency as the only measure of emotional labor. Frequency in this context merely means a count of how frequently the appropriate emotional displays are exhibited during interactions between the service providers and the clients. While Morris and Feldman acknowledged that frequency is an important indicator and dimension of emotional labor, they did not feel that frequency alone captured the planning, control or skill need to regulate and display emotions. Therefore, Morris and Feldman posited three additional dimensions of emotional labor: attentiveness to required display rules, variety, and emotional dissonance. According to Morris and Feldman, display rules will have different levels of required attentiveness. The more attentiveness to a display rule that is required, the more energy and effort (both psychological and physical) will be expended by the employee. The more energy and effort expended by the employee will result in greater levels of emotional labor. To properly measure attentiveness to display rules, Morris and Feldman proposed that both the duration and intensity of the interactions between the service providers and the clients be measured. The duration of emotional display was based on Sutton and Rafaeli’s (1988) work which involved convenience store clerks. In their study, Sutton and Rafaeli found that when the clerks’ interactions with clients were short, the clerks typically used a highly scripted interaction format such as a simple thank you or a smile. As a result of these findings, Morris and Feldman (1996) proposed that short transaction require little effort

PAGE 16

8 while longer interactions would require more effort and therefore result in greater levels of emotional labor. According to Morris and Feldman, intensity refers to “how strongly or with what magnitude an emotion is experienced or expressed” (p. 990). Morris and Feldman included intensity as a facet of attentiveness to display rules based on the work of Frijda, Ortony, Sonnemans, and Clore (1992), who argued that intensity of emotion by the service provider is the number one predictor of whether clients and customers change their behaviors during service interactions. Morris and Feldman then tied intensity in to Hochschild’s (1983) concept of deep acting discussed above. It was Morris and Feldman’s belief that the intense emotions require greater effort and in order to express these intense emotions, employees would need to use thoughts, images and memories to induce the desired intense emotion. Therefore, intensity of emotion was linked to deep acting with greater intensity resulting in greater use of deep acting which then resulted in greater emotional labor. The third component, variety, speaks to the issue of the variety of emotions to be displayed. According to Morris and Feldman, emotions can vary within jobs as well as across jobs. Variety within jobs is common within the service industry as different emotions may be required as the transaction occurs or organizations may have different display rules based on the situation. For example, a sales clerk may start out displaying friendly emotions when they greet the customer and then may have to change to concern or sympathy when they ascertain that the customer is there to return a defective product. Another example would be that according to the organization, the clerks are to spend

PAGE 17

9 more time with customers when business is slow compared to less time with customers when business is brisk. Both of these examples illustrate how emotions can vary within the same job. An example of variety of emotions across jobs would be a customer service clerk who deals with customers face-to-face in a busy store compared to a support technician that interacts with his or her customers via the phone. In this example, it is likely that the customer service clerk experiences much more emotional variety than the support technician. Morris and Feldman posited that the greater the displayed emotional variety, the greater the emotional labor. The final component, according to Morris and Feldman, is emotional dissonance. Emotional dissonance reflects the conflict between actual felt emotion and displayed emotion. Adding emotional dissonance as a component of emotional labor was a shift in the literature according to Morris and Feldman because up until that time emotional dissonance had been viewed as a consequence of emotional labor. Morris and Feldman argued that emotional dissonance should be considered a dimension of emotional labor because it is the mismatch between felt and displayed emotion that requires greater control, effort and skill for people to display the required emotion thus resulting in greater emotional labor. A year later, in 1997, Morris and Feldman honed their theory of emotional labor down to three dimensions, frequency, duration and emotional dissonance. However, in both publications, the researchers theorized that there were antecedents and consequences of emotional labor. Figure 1 graphically depicts the Morris and Feldman (1997) model. As can be seen, the theorized antecedents were: 1) explicitness of display rules, 2)

PAGE 18

10 routineness of task, 3) job autonomy and 4) power of role receiver. These antecedents were then theorized to be linked to the three dimensions of emotional labor: frequency, duration and emotional dissonance. The proposed consequences of emotional labor in this model were emotional exhaustion, job satisfaction and role internalization.

PAGE 19

11 Figure 1 Morris and Feldman (1997) Antecedents and Consequences of Emotional Labor Model Antecedents 1. Explicitness of Display Rules 2. Routineness of Task 3. Job Autonomy 4. Power of Role Receiver Emotional Labor Frequency of Interaction Duration of Interaction Emotional Dissonance Consequences 1. Emotional Exhaustion 2. Job Satisfaction 3. Role Internalization

PAGE 20

12 The Morris and Feldman work is viewed as a theory that is focused on the expressive behavior rather than the management of emotions. As such, this theory does not specifically address surface or deep acting, although they did acknowledge the utility of both surface and deep acting as way ways in which people could manage their emotions. Instead, these researchers believed that the main focus should be appropriate expressive behavior because that is what organizations desire. How the employees manager their emotions so that they can produce the desired emotional display was of little consequence in this theory. The benefit that Morris and Feldman model brought was to provide a model of emotional labor that depicts emotional labor as a multi-dimensional construct and to provide additional ways to measure emotional labor other than simply measuring frequency of displays. In addition, Morris and Feldman detailed anticipated antecedents and consequences of emotional labor. In 2000, Grandey published her work which argued for a new way to conceptualize emotional labor. Grandey began her theory by outlining the different theoretical perspectives that had been posited in the past (Hochschild, 1983; Ashforth and Humphrey, 1993; Morris and Feldman, 1996). Grandey argued that an integrated definition and theoretical model was needed. In an attempt to provide an integrated definition and model, Grandey looked for similarities across all three theoretical positions. In Grandey’s opinion, the similarity across theories was that individuals could regulate their emotional expressions at work. Based on this understanding, Grandey

PAGE 21

13 defined emotional labor as, “the process of regulating both feelings and expressions for the organizational goals” (p. 97). The model that Grandey proposed (see Figure 2) incorporated different aspects of all three different theoretical positions and was based on emotion regulation theory. From Hochschild (1983), Grandey incorporated deep acting and surface acting, from Ashforth and Humphrey (1993) Grandey incorporate incorporated display rules as situational cue cues preceding emotional labor, and from Morris and Feldman (1996), Grandey incorporated frequency, duration and variety also as situational cues which preceded emotional labor. As can be seen in Figure 2, Grandey did not propose any antecedents to emotional labor, just situational cues that act as triggers for emotional labor to be performed. The situational cues that Grandey proposed were based largely on the Morris and Feldman (1996) work as three of the situational cues (frequency, duration and variety) were proposed by Morris and Feldman as components of emotional labor itself. In addition, Grandey posited that display rules were also situational cues. Grandey added emotional events as a pre-curser to emotional labor which had not previously been proposed.

PAGE 22

14 Figure 2 Grandey (2000) Emotional Labor Model Situational Cues Emotional Regulation Process Long-Term Consequences Grandey’s model also introduced moderators. These moderators are separated as two different groups, one for individual factors and one for organizational factors. Grandey’s conceptualization of consequences is also separated into individual well-being and organizational well-being consequences. Grandey’s model indicates that burn-out, job satisfaction, performance and withdrawal behavior are all potential consequences of emotional labor. Grandey’s contribution to the emotional labor literature is a model based on emotion regulation theory that attempts to incorporate certain aspects from all previously espoused theories for a comprehensive, integrated model. In this model, surface acting Emotional Labor Deep acting: Modify Feelings Attentional Deployment Cognitive change Surface acting: Modify Expressions Response modulation Interaction Expectations Frequency Duration Variety Display Rules Emotional Events Positive events Negative events Individual Well-Being Burnout Job Satisfaction Organizational Well-Being Performance Withdrawal Behavior Individual Factors Gender Emotional expressivity Emotional intelligence Affectivity (NA/PA) Organizational Factors Job autonomy Supervisor support Coworker support

PAGE 23

15 and deep acting, previously seen as mechanisms used to manage emotional labor, act as the two components of emotional labor itself. This model does not include emotional dissonance as a component part. Moderators and consequences in this model are separated as to whether they are addressing the individual or the organization. At approximately the same time as the Grandey (2000) model was being presented, Kruml and Geddes (2000) were publishing their research findings that emotional labor is comprised of two factors: emotive effort and emotive dissonance. Kruml and Geddes found that the emotive dissonance factor represents the degree to which employees’ expressed emotions align with their true feelings. Kruml and Geddes argued that a high score on this factor would represent more surface acting while a low score on this factor would represent passive deep acting, genuine expression of emotion. Conversely, the emotive effort factor can be viewed as a measure of active deep acting with a high score on this facet representing high levels of active deep acting. Kruml and Geddes’ contribution to the literature is that emotive effort is a construct that explicates the labor involved in emotional labor. The next major theory of emotional labor was presented by Brotheridge and Lee in 2003. Somewhat in contrast to the Morris and Feldman (1996 and 1997) theory and the Kruml and Geddes (2000) model, Brotheridge and Lee (2003), felt that emotional labor does not always lead to emotional dissonance and therefore emotional dissonance should not be a component of emotional labor. As reasoning for this assertion, the researchers posited that workers who genuinely feel the required display emotion do not experience emotional dissonance. However, although these researchers asserted that

PAGE 24

16 emotional dissonance is not a component of emotional labor, they did propose that emotional dissonance may be associated with surface and deep acting. This assertion arises from Brotheridge and Lee’s (1998) earlier work where they indicated that surface acting may be a manifestation of dissonance, and perhaps a “proxy for emotional dissonance” (p.9). Their 1998 validation attempts included a factor analysis where they found that the surface acting factor contained both the items from the emotional dissonance scale and the items from the surface acting scale. These two sub-scales were positively related (r=0.79, p<.001) and shared comparable relationships with the remaining variables. Based on these findings, the researchers asserted that, “respondents may not be able to distinguish between the two constructs” (p.12) and therefore combined the emotional dissonance items together with the surface acting items to measure one component of emotional labor called surface acting. According to Brotheridge and Lee, emotional labor is a multi-dimensional construct which is comprised of six components. These six components are frequency, intensity of emotional display, variety of emotional display, duration, surface acting and deep acting. The influence of Morris and Feldman (1996) on Brotheridge and Lee is obvious, from the inclusion of the first four components and the influence of Grandey (2000) is obvious from the inclusion of surface and deep acting as actual components of emotional labor. In the scale that Brotheridge and Lee developed to measure emotional labor, the surface acting sub-scale consists of three items that address both suppression and faking of emotions. As both suppression and faking of emotions are key components of

PAGE 25

17 emotional dissonance as they represent a mismatch between actual felt emotion and displayed emotion, it is clear that the Brotheridge and Lee surface acting sub-scale is capturing dissonance as well as surface acting. Brotheridge and Lee’s contribution to the literature was a melding of the Morris and Feldman (1997) theory and the Grandy (2000) theory. In addition, Brotheridge and Lee provided a new, validated measurement tool for emotional labor. The newest conceptualization of emotional labor was provided by Glomb and Tews in 2004. Glomb and Tews acknowledged the confusion which abounds in the emotional labor research domain and argued that consensus among researchers has never been reached from either a theoretical or methodological standpoint. Theoretically, Glomb and Tews argued that there are three main themes that guide the emotional labor research. These themes would include internal states, internal processes and external behavioral displays. Morris and Feldman (1996) would represent a theory with an internal state theme, Grandey (2000) would represent a theory with an internal process theme and Ashforth and Humphrey (1993) would represent a behavioral display theme. Glomb and Tews argued that the emotional labor domain could appear to be in a “theoretical quandary, flooded with a multitude of conceptualizations” (p. 4). However, these researchers also posited that researchers need to recognize the complexity of emotions in the workplace and view emotional labor as a network of related constructs. To illustrate how the different themes may, in fact, be complimentary, Glomb and Tews used the example of display rules motivating an employee to experience emotional

PAGE 26

18 dissonance (internal state), requiring the employee to use surface or deep acting (internal process), resulting in an organizational emotional display (behavior display). The Glomb and Tews conceptualization of emotional labor focuses largely on the behavioral expression and non-expression of felt or unfelt emotions as they respond to display rules. The argument made by Glomb and Tews is that emotional labor is comprised of both the expression of emotions and the non-expression of emotions in accordance to display rules. This conceptualization utilizes two dimensions: one dimension classifies an emotional display as either an appropriate expression or an appropriate non-expression; and the other dimension is a felt continuum which indicated whether the expression or non-expression is consistent with an internal feeling. Figure 3 illustrates the underlying premise of this theory. Figure 3 Glomb and Tews (2004) Model of Emotional Labor FELT 1 2 No Yes No Expressed Display Nothing Felt Appropriate or Displayed Suppressed Display TYPE OF APPROPRIATE DISPLAY 3 4 Appropriate Appropriate Expressed Display Faked Genuine Display Display

PAGE 27

19 Glomb and Tews argue that there are five reasons that their conceptualization of emotional labor extends and is different from previous theories. First, this framework accounts for underlying felt emotion that happens when employees conform to display rules. Second, this framework acknowledges that conformance to display rules can be done in two ways, either by expressing an appropriate emotion or not expressing an inappropriate emotion. Third, this framework recognizes the importance of genuinely felt emotions and also recognizes the suppression and faking of emotions (which may be more taxing than expressing genuinely felt emotions). Fourth, this framework allows for distinction between positive and negative emotions. Finally, this framework encompasses both emotions and moods. Utilizing the framework provided by Glomb and Tews is helpful in understanding their concept of emotional labor and emotional dissonance’s contribution to emotional labor. In Cell 1 of the framework, emotion is neither felt nor expressed and is not measured because it represents an absence of both felt and expressed emotion. Cell 2 of the framework represents negative emotions that are felt but not expressed (an appropriately suppressed display). Cell 3 of the framework represents positive emotions that are expressed but not felt (an appropriate faked display). Cell 4 of the framework represents the match between felt and appropriate display emotions (genuine emotions). Glomb and Tews indicate that the majority of emotional labor research has focused on cells 2 and 3. Cells 2 and 3 would also represent emotional dissonance as there is a mismatch between felt and displayed emotion. Cells 1 and 4 represent a match between compliant display and felt emotion.

PAGE 28

20 Glomb and Tews’ contribution to the literature is that their conceptualization of emotional labor is one in which it is part of a network of related constructs. These researchers offer a way of understanding how the different theories and their underlying themes may be complimentary in nature. In addition, these researchers developed and validated a new, versatile emotional labor measurement tool. Depending on the area of interest, researchers can measure emotional dissonance only (cells 2 and 3) by obtaining scores on suppression and faking of emotions, or researchers can measure emotional labor (cells 2, 3 and 4) which then incorporates genuine expression of emotions. Possibly the greatest contribution that these researchers made to further the understanding of emotional labor was when they stated, “the theoretical orientation a researcher adopts will depend on the research question” (p. 4). This study was designed to examine whether personality characteristics would serve as antecedents to lessen the incongruence between actual felt emotions versus displayed emotion. From a theoretical standpoint, this incongruence would be defined as emotional dissonance. Further this study was designed to then determine if lower emotional dissonance led to increased job satisfaction when then led to decreased intentions to quit and increased job performance. The research question concerns the antecedents and consequences of emotional dissonance. Given this, the theoretical orientation that best fits this research question is that provided by Glomb and Tews as it allows for the measurement of emotional dissonance by capturing the frequency of both suppressed and faked emotions.

PAGE 29

21 Empirical Studies of Emotional Dissonance Several researchers have explored the consequences of emotional dissonance. Morris and Feldman (1997) found a significant positive relationship between emotional dissonance and emotional exhaustion. This finding is extremely interesting in that Morris and Feldman also attempted to make a connection between the frequency and duration components of emotional labor to emotional exhaustion and were unable to do so thus implying two things: one, that it is the emotional dissonance component that affects emotional exhaustion and that emotional dissonance is a distinct component of emotional labor with its own unique consequences. Abraham (1998) also explored consequences of emotional dissonance and found that both job satisfaction and emotional exhaustion were related to emotional dissonance. Job satisfaction was negatively related to emotional dissonance and emotional exhaustion was positively related to emotional dissonance. In addition, job satisfaction’s relationship with emotional dissonance was moderated by social support such that those with high levels of social support were less likely than those with low levels of social support to experience the negative effects of emotional dissonance on satisfaction. Abraham extended these findings in 1999 when she confirmed that job satisfaction mediates the relationship between emotional dissonance and intention to quit as well as emotional dissonance and organizational commitment. According to Abraham (1999), increased emotional dissonance results in decreased job satisfaction, which then leads to both increased turnover intention and decreased organizational commitment.

PAGE 30

22 While validating the emotional labor scale, Brotheridge and Lee (2003) found that emotional exhaustion and depersonalization were significantly correlated with their surface acting subscale. These researchers felt that these findings suggested that the effort to hide one’s true feelings or to pretend to feel those that were expressed is the major contributor to emotional strain. More importantly, these researchers confirmed that surface acting is comprised of both suppressing and faking emotions. If this is the case, then the Brotheridge and Lee surface acting is indeed acting as a “proxy” for emotional dissonance as defined by Glomb and Tews (2004). Using this rationale, then the Johnson (2004) results are relevant in that she found that surface acting was positively and significantly related to emotional exhaustion, and negatively and significantly related to affective well-being and may be considered possible consequences of emotional dissonance. Overall, it would appear that past research has clearly shown the negative affects of emotional dissonance on job satisfaction and turnover intentions. The relationship between emotional dissonance and job satisfaction is of particular interest for this study because job satisfaction, or lack thereof, has been found to be linked to job performance. Past meta-analyses have estimated that the correlation between job satisfaction and job performance is in the .20s (Iaffaldano & Muchinsky, 1985; Petty, McGee & Cavender, 1984). A meta-analysis was also conducted by Mathieu and Zajac (1990) that confirmed the small but positive relationship (r = .13) between organizational commitment and job performance. Individual components of organizational commitment have even stronger positive relationships with job performance (Hackett, Bycio & Hausdorf, 1994). In all of

PAGE 31

23 these cases, job performance was related to variables that have been shown to be strongly affected by emotional dissonance. Limitations of Past Empirical Research. As discussed earlier, Hochschild (1983) likened the range of emotional labor to the toe and the heel of a human foot, with customer service representing the toe and debt collectors representing the heel. Although employees in both of these professions have the propensity to experience great amounts of emotional labor and emotional dissonance, the experience itself is resulting from two opposite emotions. The most likely cause for emotional dissonance in customer service would be the suppression of negative emotions and the faking of positive emotions. Conversely, the most likely cause for emotional dissonance in debt collectors would be the suppression of positive emotions and the faking of negative emotions. Although almost all researchers acknowledge the difference between these two professions, not one empirical study has solely used a profession with the emotional demands equal to that of a debt collector. Overall, the majority of the research has included flight attendants (Hochschild, 1983), telecommunications representatives (Abraham, 1998), sales clerks, teacher’s aids and waiter/waitresses (Cte and Organ, 2002), college students working full or part-time in retail environments (Brotheridge and Lee (2003), and retail store employees and customer service agents (Johnson, 2004). Two studies, Morris and Feldman (1997) and Kruml and Geddes (2000), did include debt collection agencies. However, these samples were combined with customer service samples and no analyses were done to determine if the findings differed as a

PAGE 32

24 result of the type of emotional dissonance experienced. However, it is interesting to note that both studies that used collection agencies as a part of their sample confirmed the presence of emotional dissonance. To date, there has only been one published study (Sutton, 1991) that directly addressed the difficulties of being a debt collector. Unfortunately, while this article provided insight on the debt collecting profession, it was a qualitative study which explored the use of norms (urgency and voice tone) associated with collecting debt and emotional dissonance was not explored. One contribution of the current study was to explore the antecedents and consequences of emotional dissonance in a population that has, to date, been relatively unexplored. Additionally, the current study extends current theory in emotional labor by exploring dissonance in a population that must suppress pleasant emotion emotions and fake unpleasant ones. Current Emotional Dissonance Research Currently, the thrust in the emotional dissonance research area is to determine whether individual differences exist in people’s ability to handle emotional labor and emotional dissonance (Brotheridge, 2003; Kring, Smith & Neale, 1994; Tews & Glomb, 2003; Vey & Bono, 2003). Unfortunately, again, such studies have been limited to customer service jobs. Although the consequences of emotional dissonance may be similar for both the customer service and debt collection workers, the antecedents, and more specifically the underlying personality make-up of these individuals, may be different. In other words, the personality characteristics needed for someone to be upbeat and enthusiastic (suppressing negative and faking positive) all day could be quite

PAGE 33

25 different from the underlying personality characteristics needed for someone to be combative, aggressive and forceful (suppressing positive and faking negative). Of course, whether different personalities result in different emotional dissonance levels across jobs is an empirical question. This research addressed the question of such individual differences in debt collectors. The underlying rationale was that, because emotional dissonance involves incongruence between felt emotion and displayed emotion, if personality attributes lessened the incongruence, emotional dissonance would decrease. More simply put, if the personality attributes predispose people to display required emotions, then incongruence between felt emotions and displayed emotions would be lessened, thus lowering the emotional dissonance. The research community has been very vocal about the need to find individuals that are well suited to the emotional requirements of the job. Sutton (1991) stated: “organizations should focus substantial efforts on attracting and selecting candidates who are disposed to feel and display required emotions” (p. 266). Morris and Feldman (1996) stated: “from a selection perspective…selecting employees on the basis of their general tendency to experience certain emotions may lead to a better fit between an employee’s expressive behaviors and work role requirements” (p. 1006). In the present research, I attempted to answer the call from the many researchers who have explored this path previously. This study was designed to ascertain whether there are certain personality factors that could accurately predict those individuals whose personalities made the congruence between experienced emotion and displayed emotion greater, therefore making these

PAGE 34

26 individuals less susceptible to the perils of emotional dissonance. It was proposed that such personality factors would have a direct effect on the extent of emotional dissonance experienced by different individuals. Those who experienced less emotional dissonance would tend to be more satisfied with their jobs. Those who were more satisfied would be less likely to quit and more likely to perform well on the job. Schematically, the model is shown in Figure 4.

PAGE 35

27 Figure 4 – Path Model including Hypotheses and Proposed Relationship Directions Extraversion N euroticis m (Anger) Agreeableness Conscientiousness Helpfulness Emotional Dissonance Job Satisfaction (Nature of Work) Intention to Quit Performance H6 (-) H1 (-) H2 (-) H3 (+) H4 (-) H5 (-) H8 (+) H7 (-)

PAGE 36

28 Personality as an Antecedent of Emotional Dissonance In the past several decades, much research has been done regarding personality and work. Overall, some researchers now agree that it is useful to consider personality as consisting of five robust factors (Barrick & Mount, 1991). These five robust factors have become known as the Five Factor Model of Personality (FFM) or simply the “Big Five.” The FFM represents a taxonomy, or classification system, of the five major personality traits found, through factor analysis, to be consistently present. These five factors have been found to be consistent across different frameworks (Goldberg, 1981), using different instruments to measure the factors (Costa & McCrae, 1988; McCrae & Costa, 1985, 1987, 1989). All of the five factors have been found to be relatively independent of cognitive ability (McCrae & Costa, 1987), which is typically used for employment screening purposes. The five factors have been commonly labeled as: Extroversion/Introversion, Emotional Stability or Neuroticism, Agreeableness or Likability, Conscientiousness, and Openness to Experience or Culture. Extroversion/Introversion Extroversion/Introversion is widely agreed to be the first dimension in the Five Factory Personality Model (Barrick & Mount, 1991). A person who scores high on this trait would be someone is sociable, gregarious, assertive, talkative and active. This factor was examined by Tews and Glomb (2003) in their exploration into whether certain personality traits affect the ability to express and/or suppress emotions differently. Unfortunately, these researchers did not explore the faked negative and suppressed

PAGE 37

29 positive dimensions in this study, both of which may have greater impact when looking at the role of a debt collector. These researchers found that extraversion was positively related to expressing genuinely felt positive emotion (r = .46, p <.01) as well as positively related to faking positive emotions (r =.27, p < .10). The measures used in this study were self-report measures wherein the participants indicated how frequently they expressed certain emotions when they actually felt that way, how frequently they expressed emotions when they didn’t feel that way and how often they kept certain emotions to themselves. Because of its content, this measure can be viewed as a measure of emotional dissonance rather than emotional labor. Given that the sample that these researchers utilized was customer service agents who are required to be pleasant at all times to their customers, these findings, while important, are not surprising. The results of this study reflect the importance of alignment between experienced emotion and required display emotion. By virtue of the extraversion trait, these customer service agents are pre-disposed to be sociable which may be viewed as a match between pre-disposition and required display if the required display is for the customer service agents to be social to their customers. Therefore, the difference between experienced emotions and required display emotions is relatively low, thus allowing these customer service agents to experience lower levels of emotional dissonance. These researchers found significant support for extraverts being able to express genuine positive emotions and limited support for ability to fake positive emotions. The

PAGE 38

30 facet of extroversion that w ill be required by the debt collectors will be the ability to talk to their customers and to be assertive while they are attempting to collect the debt from the debtor. Another finding in the Tews and Glomb (2003) study that may be of greater import to the current study is that extraversion was not found to be significantly related to self-reported measures of how frequently customer service workers had to suppress negative emotions (this was inconsistent with the researchers’ expectations). In other words, extraverts did not feel that they had to suppress negative emotions that they were experiencing. Taken together with the above findings, this would mean that those scoring high on the extraversion trait will not feel the need to restrain negative emotions while also be able to fake positive emotions. Although expression of negative emotion is frowned upon in the customer service arena, it is viewed as an asset in the debt collection arena. Based on the previous findings, and taking into consideration the job differences between customer service agents and bill collectors, the following hypothesis was posited: Hypothesis 1 – The relations between Extraversion and emotional dissonance will be negative. Emotional Stability/Neuroticism Emotional Stability or Neuroticism is the next factor in the FFM. Emotional Stability represents one end of the factor’s spectrum, while Neuroticism represents the other end of the spectrum.

PAGE 39

31 For purposes of the current research, the term Neuroticism will be used. Therefore, a person who scores high on this trait would be someone who is anxious, depressed, angry, is easily embarrassed, very emotional, worried and insecure. Again, this personality trait was explored by Tews and Glomb (2003) in their research of emotional dissonance and the five-factor model. These researchers posited several hypotheses relating to neuroticism. However, the most important finding, as it relates to the current study, is that neuroticism was positively related to expressing genuinely felt negative emotions. The importance of this hypothesis is that the current study dealt with debt collectors. Debt collectors are expected to routinely express negative emotions. Tews and Glomb (2003) found that the standardized coefficient for neuroticism explaining negative emotion expression was .51 (p < .01). Because these researchers were measuring expression and suppression of different types of emotions, this finding shows that those participants who scored high on the neuroticism trait were more easily able to express negative emotions. Neuroticism, as measured by the IPIP, is composed of six different facets: anxiety, anger, depression, self-consciousness, immoderation and vulnerability. Of the six facets, the most relevant for the current study is anger. In the context of the current study, debt collectors are all expected to display anger at customers (the debtors); angry collectors would have a more closely aligned balance between experienced emotion and displayed emotion. Based on the job description of debt collector, it makes sense then that anger would more closely align with expressed emotion than the combined measure

PAGE 40

32 of Neuroticism. The close alignment would result in less emotional dissonance. Therefore, based on the job requirements of a debt collector having to display anger and on the rationale provided above, the following hypothesis was posited: Hypothesis 2 – The relations between Anger and emotional dissonance will be negative. Agreeableness or Likeability The third factor in the five-factor model is generally interpreted as Agreeableness or Likeability. A person who scores high on this trait would be one who is courteous, flexible, trusting, good natured, cooperative, forgiving and soft-hearted. Returning to the Tews and Glomb (2003) research, agreeableness was hypothesized to be positively related to faking positive emotions as well as to be positively related to suppressing negative emotions. Agreeableness was not found to be positively related to faking positive emotions but was found to be positively related to suppressing negative emotions (r = .19, p <.05). Again, it needs to be noted that the Tews and Glomb (2003) study used customer service agents as the participants. In the current study, debt collectors were used. Debt collectors’ job duties differ radically from those of customer service agents. Within the debt collection field, the finding that people who score high on the agreeableness trait also tend to suppress negative emotions is an important finding. Suppressing negative emotions is not encouraged within the debt collection industry. Therefore, this personality factor may be of extreme importance for the current study inasmuch as there should be great disparity between the experienced emotion and the required display

PAGE 41

33 emotion. Indeed, supervisors within the collection industry urge their collectors to be firm; too much empathy is not helpful as can be seen from the following quote from the Sutton (1991) article: “The sob stories make it hard sometimes, especially for new collectors, and especially when they seem to be telling the truth. But you’ve got to get tough with them when they sound down. If you don’t get the money, another creditor will” (p. 261). Based on the debt collection industry requiring their employees to not be agreeable, together with the Tews and Glomb (2003) findings regarding agreeableness, the following hypothesis was posited: Hypothesis 3 – The relations between Agreeableness and emotional dissonance will be positive. Conscientiousness The fourth factor is best known as conscientiousness. Although there has been some debate about what the actual factor is measuring, it is most commonly agreed that a person scoring high on this factor would be one who is dependable, responsible, thorough, organized, hardworking, achievement-oriented and persevering. Barrick and Mount (1991) found conscientiousness to be a consistent predictor of job performance for all occupational groups and all criterion types. Based on this finding, the researchers stated, “this aspect of personality appears to tap traits which are important to the accomplishment of work tasks in all jobs” (p. 17). Tews and Glomb (2003) also looked at this factor in their research. In that study, it was found that conscientiousness was negatively related to suppressing negative

PAGE 42

34 emotions ( = -21, p < .10), which would suggest that conscientious people are less likely to keep negative emotions to themselves. Much like the agreeableness factor, this finding is significant when one considers the debt collection industry. Again, simply by the nature of the job itself, debt collectors are not expected to keep negative emotions to themselves. Instead, debt collectors are given free reign to express negative emotions to those from whom they are attempting to collect a debt. Based on the Barrick and Mount (1991) and the Tews and Glomb (2003) findings, that conscientious people are driven to complete their work tasks regardless of the situation and that they have a difficult time suppressing negative emotions, the following hypothesis was posited: Hypothesis 4 – The relations between Conscientiousness and emotional dissonance will be negative. Openness to Experience The fifth factor is called Openness to Experience. A person scoring high on the Openness trait is imaginative, cultured, curious, original, broad-minded, intelligent and sensitive. The Barrick and Mount (1991) meta-analysis did not find any strong relationships between Openness to Experience and any organizational variable. Given that the traits associated with Openness were not related to the job functions of a debt collector and on the lack of prior theoretical findings, Openness was not a personality factor explored in this study. Pro-Social Behavior – Helpfulness The trait helpfulness may also play a role in the congruence between felt and displayed emotion for bill collectors. Although helpfulness may not seem an obvious

PAGE 43

35 choice, its importance was noted in a qualitative summary that explored the role organizational norms play in the way that debt collectors interacted with debtors (Sutton, 1991). A supervisor from the sample was quoted as saying: “Remember, even if they have to get tough to get the payment, the collector is helping the debtor. The collector has to tell himself: ‘I’m helping this person to save their credit rating. If they don’t pay me, they may never be able to buy a car or a house’” (p. 261). One type of measure that predicts helpfulness is the Prosocial Personality Battery (PSB) (Penner et al., 1995). The PSB was designed to predict a wide range of prosocial reactions and behaviors. The PSB measure attempts to tap two dimensions, OtherOriented Empathy and Helpfulness. The first dimension, Other-Oriented Empathy, measures five related traits: empathic concern, ascription of responsibility, other-oriented moral reasoning, perspective taking and mutual-concerns moral reasoning. This factor attempts to assess cognitions and affect (Penner et al, 1995). People who score high on this factor tend to feel responsibility and concern for other people. The second factor of the PSB, Helpfulness, has two underlying constructs: personal distress and self-reported ‘altruism’ (Penner et al, 1995). This factor appears to assess both behavioral tendencies and affect. The self-reported altruism construct addresses and measures past helping behaviors as an indicator of future helping behaviors while the personal distress construct addresses and measures the affective reaction to others’ distress. People who score high on this factor tend to be altruistic and help others but they do not tend to experience self-oriented discomfort when another person is in

PAGE 44

36 extreme distress. Within the collection industry, collectors have frequent contact with others who are in extreme distress, making this factor of great interest for the proposed study. The PSB’s relationship with other personality factors has been explored. Specifically, Penner and Fritzsche (1993b) compared the correlations between the PSB and the Revised NEO Personality Inventory (NEO PI-R; Costa and McCrae, 1992). In this study Penner and Fritzsche found a significant correlation between the OtherOriented Empathy factor and Agreeableness (r = .52, p <.001), but a nonsignificant correlation between Helpfulness and Agreeableness (r = .11, ns). This non-significant correlation between Helpfulness and Agreeableness was unexpected as altruism is one of the facets of Agreeableness. This finding can be interpreted to mean that the Helpfulness factor may account for additional variance in predicting emotional dissonance above and beyond the four personality factors already designated in Figure 4. Three additional studies lend support to the theory that the PSB has predictive capabilities above and beyond those found by using typical personality based models. Freifeld (1993) explored the relationship between the PSB and two interpersonal styles, nurturance and dominance. The results showed that although both factors of the PSB correlated significantly with nurturance, the correlation between nurturance and the Helpfulness factor was significantly lower (r = .23) than the correlation between nurturance and Other-Oriented Empathy (r = .52). More interesting, however, was the relationship found between Helpfulness and Dominance (r = .34, p < .001). Poindexter (1994) replicated the Freifeld (1993) study and did not find the same

PAGE 45

37 results. In this study, Poindexter confirmed the positive relationship between nurturance and Other-Oriented Empathy (r =.61, p < .01) but was unable to replicate the previous findings of a positive relationship between nurturance and Helpfulness (r=.01, ns). The relationship between Helpfulness and Dominance was replicated (r=.34, p <.01). Although these findings would cast doubt as to whether there truly is a relationship between Helpfulness and Nurturance, it confirms the strong, significant relationship between Helpfulness and Dominance. The final study (Penner and Fritzche, 1993a) used both the NEO PI-R and the Interpersonal Adjective Scale – Revised (IAS-R) to investigate the relationships between the PSB, assertiveness and self-confidence. In this study, a positive, significant relationship was found between Helpfulness and assertiveness as well as between Helpfulness and self-confidence. Taken together, these findings can be interpreted to mean that people who score high on the Helpfulness factor are the type of people who are likely to engage in prosocial actions because they believe that their help will be effective. Low agreeableness, dominance, assertiveness and high self-confidence are all traits that are favored when selecting collectors. Due to the job duties associated with being a bill collector, these traits are seen as desirable for job performance. Therefore, based on the findings of the studies discussed above, it follows that the factor of Helpfulness found in the PSB will be a good predictor of emotional dissonance in the collection industry. The more helpful a debt collector is, the more the collector is able to see his or her job as helping the debtor out of financial ruin thus lessening the distance

PAGE 46

38 between the felt emotion and the displayed emotion. Based on the findings above, the following hypothesis was posited: Hypothesis 5 – The relations between Helpfulness and emotional dissonance will be negative. Consequences of Emotional Dissonance Job Satisfaction Dissatisfaction on the job is one of the consequences that can arise from emotional labor. Job satisfaction is an attitudinal variable defined by how much people like or dislike their job (Spector, 1997). Job satisfaction can be measured at the global level or at the facet level depending on the researcher’s particular area of interest. Depending on the level of interest, global or facet, there are many different assessment tools that can be used to measure job satisfaction. A popular global measure of job satisfaction is the Job in General Scale (also known as the JIG, Ironson, Smith, Brannick, Gibson and Paul, 1989). Common measures for facet scales include the Job Satisfaction Survey (JSS; Spector, 1985), the Job Descriptive Index (JDI; Smith, Kendall and Hulin, 1969), and the Minnesota Satisfaction Questionnaire (MSQ; Weiss, Dawis, England and Lofquist, 1967). The JSS (Spector, 1985) has nine facets including pay, promotion, supervision, fringe benefits, contingent rewards, operating conditions, coworkers, nature of work, and communication. Of most interest with regard to this study is the nature of work facet. This facet directly assesses the employee’s satisfaction with the work itself. Because the thrust of this study is that debt collectors can

PAGE 47

39 experience high levels of emotional labor due to the nature of the work, the use of this particular facet makes the most sense. Previous research has explored the link between global job satisfaction and emotional dissonance. Morris and Feldman (1997) reported that they found a significant relationship (r = -.37, p<.001) between emotional dissonance and global job satisfaction. Abraham (1999) explored the linkage between emotional dissonance and job satisfaction and found a significant negative (r = -.22, p< .05) relationship. Further analysis indicated that job satisfaction acted as a mediator between emotional dissonance and intention to quit. This study provided data necessary to replicate Morris and Feldman’s as well as Abraham’s findings. One study (Wharton, 1993) found job satisfaction to be positively related to emotional labor. In this study, however, job titles were merely coded dichotomously as either a known or unknown occupation to entail emotional labor. The designation of whether an occupation was one that entailed emotional labor was drawn from Hochschild’s (1983) initial listing of occupations. The unusual findings (Wharton, 1993) may be the result of differing job content rather than differing amounts of emotional labor within jobs. Based on the findings above as well as the theory, the following hypothesis was posited: Hypothesis 6 – The relations between Job Satisfaction, more specifically satisfaction with the nature of the work, and emotional dissonance will be negative.

PAGE 48

40 Intention to Quit The correlates of job satisfaction, both negative and positive, have been well documented through previous research. The negative consequence of interest for this particular study, however, is the relationship between job satisfaction and intention to quit. This particular consequence holds great interest for the debt collection industry. Reducing turnover is an extremely salient issue for this particular industry, which records turnover rates averaging between 60% and 110% annually. Such high turnover rates come at an extremely high price for the industry in terms of re-staffing and training incoming employees. Reducing the amount of turnover experienced by this industry could save the collection industry millions of dollars per annum. Although there have been many published articles addressing the satisfaction/intention to quit relationship (Arnold and Feldman, 1982; Bluedorn, 1982; Hollenbeck and Williams, 1986), the main thrust of support comes from analyses using meta-analytic data by Tett and Meyer (1993). In this particular study, strong support was found for the relationship between satisfaction and turnover intention ( r = -.53, p < .05). Contrary to their prediction, these researchers also found that global measures of job satisfaction did not correlate more strongly than facet measures of job satisfaction with intention to quit. The path analyses computed in Tett and Meyer (1993) also provided a clear picture of the relations among satisfaction, intention to quit, and turnover. Tett and Meyer (1993) confirmed that dissatisfaction leads directly to turnover intentions rather than through organizational commitment as originally thought. As found in previous

PAGE 49

41 meta-analyses (e.g., Steel and Ovalle, 1984), turnover intentions were the strongest predictor of actual turnover. Based on the findings above, the following hypothesis was posited: Hypothesis 7 – The relations between job satisfaction and intentions to quit will be negative. Performance To help clarify the relationship between job satisfaction and performance, a metaanalysis was performed by Iaffaldano and Muchinsky (1985). In this study, the researchers looked at multiple aspects of the subject relationship to ascertain not only the true relationship between job satisfaction and performance but also the best ways to measure the variables of interest. Overall, these researchers found that the relationship between job satisfaction and performance was small (.146 uncorrected; .17 corrected) but positive. Next, the researchers turned to the issue of the usage of different measurement tools and whether the tools themselves affected the satisfaction/performance relationship. For example, these researchers explored whether the satisfaction/performance relationship differed depending on whether a global or facet measure of job satisfaction was used. The researchers also explored whether using an objective or subjective measure of performance would alter the relationship between job satisfaction and performance. The results of this study confirmed that how the variables are measured does not result in a large difference in the strength of the relationship.

PAGE 50

42 For purposes of the present study, the nature of the work is believed to be the underlying cause of emotional labor. Because previous findings confirm that there is little difference in the relationship between job satisfaction and job performance when job satisfaction is measured globally or at a facet level, the facet measure of satisfaction with nature of the work was employed. The facet measure appears to be the most relevant for the purposes of the current study. In addition, the above researchers also explored whether objective or subjective measures of performance would affect the relationship between satisfaction and performance. The subjective measures of performance included both self-reports and other of sources, such as supervisory ratings. Although both of these subjective measures were known to have problems such as bias and distortion, the researchers cited Mabe and West (1982) which indicated that self-report measures may be a more valid indicator of performance than typically believed. The objective measure of performance utilized was the basic units of production. Iaffaldano and Muchinsky (1985) found that there was little difference in the correlations between satisfaction and performance whether performance was measured with objective or subjective means. Objective measures have some advantages such as not being inflated as may be the case when self-reports are used to measure how participants rate their own performance, or being a stricter measure of performance than supervisor ratings are used which may indicate personal employee preferences rather than actual performance. Therefore, based on the above findings, it was hypothesized that: Hypothesis 8 – The relations between satisfaction with the nature of work and an objective measure of performance will be positive.

PAGE 51

43 Gender Differences Previous research has indicated that there may be gender differences in two different variables examined in this study. First, gender differences have been identified with emotional labor. Some research has shown that the primary difference is that women are expected to show more emotions than men. For example, women are expected to smile more than men (Birnbaum, Nosanchcuk and Croll, 1980). If the display rules are more extreme for women than for men and both men and women feel equally at odds with the given display, this would result in greater emotional dissonance for women. Hochschild (1989) herself stated that, “the world turns to women for mothering, and this fact silently attaches itself to many job requirements” (p. 182), which would result in increased emotional management for women. Researchers have argued that woman may show more emotions than men because they have been socialized to act warmer and friendlier (Deusch, 1990), have a greater ability to encode and present their emotions (LaFrance and Banaji, 1992), and have a greater need for social approval (Hoffman, 1972). There are two points that are interesting about the above studies. First, most of the literature relates to expressing positive emotions rather than negative emotions. Indeed, one of the major reasons that Hochschild (1983) indicated that women may experience emotional labor is because service jobs were typically performed by females. Second, most of the research in this area is rather dated and it is questionable whether such findings would generalize to the modern female employee. Based on both of these

PAGE 52

44 issues, it is impossible to posit a specific hypothesis regarding gender differences, but they were examined on an exploratory basis. The second variable that has been shown to have gender differences associated with it is the personality trait of anger. Researchers have been interested in gender differences in anger in individuals of all ages. A study by Buntaine and Costenbader (1997) explored gender differences in elementary school aged children from both rural and urban environments. This study found that there weren’t differences in self-reported anger but there were differences in the expression of anger in that boys reported significantly higher levels of aggressive responses. In addition, urban children reported higher levels of anger than rural children. Moving on to middle-school children, Underwood, Hurley, Johanson and Mosley (1999) found that, when provoked, girls made fewer negative comments and made fewer negative gestures. What is interesting in this study is that it involved 8, ten, and twelve old children, and as the children became older, they became more neutral (facial expressions and gestures) when provoked. When the older children were talking, they also made fewer negative comments. Based on this research, it would seem that as children age, they become more adept at controlling their anger. Turning to adults, from a theoretical viewpoint, Sharkin (1993) gave an excellent explanation of the various anger and gender theories. Sharkin reports that expressing anger is often viewed as an empowerment tool for women and yet for men, expressing anger is viewed as something that needs to be controlled or dealt with. Overall, the theoretical position is that anger expression is perceived to be more acceptable for men

PAGE 53

45 than for women. However, empirical testing does not necessarily support the theoretical positions previously extolled (e.g. Averill, 1983; Biaggio, 1989; Spielberger et al. 1985; Sharkin & Gelso, 1991). The lack of empirical findings led Sharkin to assert that it is not really clear how women and men may differ, if they do at all, in the experience and expression of anger (p. 387). Two additional studies shed light on the gender/anger question. A study by Milovchevich, Howells, Drew and Day (2001) explored both gender and gender roles and their relationship to anger. These researchers found that males and females do not differ significantly in their expression of anger, it is the gender role that made a difference, with participants classified as masculine gender role reporting higher levels of trait anger and anger expression and lower levels of anger control. A more recent study by Brebner (2003) found that females have higher frequency rates of anger but males have higher intensity ratings of anger. Simply put, these finding indicate that females get angry more often than males. However, when males do become angry, the intensity of that anger is significantly greater than for the females. Due to the contradictory findings regarding trait anger and its relationship to gender, as well as the paucity of research that addresses whether there are gender differences in trait anger as it relates to emotional dissonance, it is impossible to posit a specific hypothesis on gender and anger. However, this is not to say that these potential gender differences are not important. Therefore, gender differences in anger, and how they relate to emotional dissonance, were examined from an exploratory view in this study.

PAGE 54

46 Summary The underlying premise of the research is that there are certain personality factors that affect people’s experience of emotional dissonance. The personality factors act, to a certain degree, to cause “fit” between the actual felt emotions and the emotions required to be displayed in the workplace. An example of this fit would be a person who scores high on extraversion who is required to be talkative at work. Such fit would minimize the amount of emotional dissonance required for successful job performance. The personality factors are antecedents of emotional dissonance because they can either increase or decrease the amount of emotional dissonance experienced. In addition, gender differences in emotional labor and anger were explored. Moving to the consequences of dissonance, the amount of dissonance was expected to influence satisfaction with the nature of the work. The expectation was that the greater the emotional dissonance, the lower the satisfaction with the nature of the work. Satisfaction with the nature of the work would then lead to two additional consequences: job performance (positively) and intention to quit (negatively). To help illustrate the information provided above, Figure 4 depicts the original proposed model, the proposed hypotheses and the proposed direction of the relationships among the variables. Although each of the hypotheses could be tested separately, it is more parsimonious to consider them collectively in a path model. Path analysis allows the researcher to estimate all the parameters of interest for hypothesis tests simultaneously. In addition, path analysis allows the researcher to examine the fit of the model as a whole

PAGE 55

47 to obtain additional information about the plausibility of the entire model as a theoretically plausible explanation of the observed relations among variables. Method Participants There were 287 employees of a large mid-western debt collection company that had two separate geographic locations. Of these 287 debt collectors, 223 collectors agreed to participate in this research project. Of the 223 participants, seven were dropped from the study due to belonging to a group whose performance goal was not consistent within the group or across the organization. Four participants were dropped because their performance data were unobtainable, and an additional 29 participants (14 from one location and 15 from the other location) were dropped because they had not been with the organization long enough for their performance data to stabilize, leaving a total participant pool of 183. To better understand the job of debt collector, it is important to know that beginning collectors are placed in small cubicles with a phone and a computer screen. For these collectors, debt collection agencies typically employ what is called an “automatic dialer” that constantly dials debtors’ phone numbers. The collector, therefore, is on the phone for eight hours a day with one call coming in as soon as he or she completes the preceding collection call. Collectors with more experience are typically tapped to handle the more complicated collections such as student loans. In more complicated collections, the collectors actually have a list of “clients” or debtors that they

PAGE 56

48 are responsible for contacting. Relationships between collectors and debtors become possible as collectors move up through the organization. The largest group of collectors, approximately 90% of the total participant pool, were required to use the automatic dialers and did not have any relationships with the debtors. As collectors are promoted within the organization, their numbers become much smaller, resulting in a group considered to be “elite”. Although the methods of contacting clients may differ across collectors, all the collectors are given a monthly monetary goal designating the amount of money they are expected to collect from the debtors for the particular month. The participant pool was comprised of 48% of participants from the first location and 52% from the second location. Of the 183 participants, 44% were male and 56% were female. The ages of the participants ranged from 18 to 61 with a mean age of 29 years. The racial mix of the participants was 81% Caucasian, 11% black, 6% Hispanic, 1% Asian, and 1% listed “other”; 2 participants did not disclose their race. The sample’s demographics appeared to be representative of the host population’s demographics. Additional demographic information is provided in Table 1. Table 1 Participant Pool Demographic Information Variable N Mean SD Range Age 183 29.11 9.46 18 – 61 Debt Collection Experience (in months) 183 25.80 33.05 2 – 240 Tenure with Host Organization (in months) 183 15.06 15.73 2 86

PAGE 57

49 Measures of Antecedent Variables Pro-Social (Helpfulness) Behaviors The Pro-Social (Helpfulness) Behavior measure was obtained by using the Penner et al. (1995) Prosocial Battery’s items that are related to the Helpfulness factor. This measure was comprised of eight items (see Appendix A). Three of the items assessed personal distress level and required the participants to indicate on a one to five Likerttype scale, their agreement with the statement presented. The additional five items assessed the participants’ self-reported altruism and required the participants to indicate on a one to five Likert-type scale, the frequency with which they participated in the actions presented in the statement. Cronbach’s alpha for this variable, as well as for all of the measured variables in this study, was computed and those results can be found in Table 2. The results found in this study were similar to those found in previous studies for most of the variables. Comparisons of all the alphas are also found in Table 2.

PAGE 58

50 Table 2 Means, Standard Deviations, Alphas and Comparisons of Measures Measure N Mean SD Alpha (Std) Comp. Pro-Social Behavior Scale – Helpfulness 182 27.35 4.56 .69 .80A Personal Distress 182 12.551.99.80 .77A Self-Reported Altruism 182 14.793.96.75 .73A Extraversion 172 35.516.91.84 .87B Agreeableness 168 39.346.21.83 .82B Conscientiousness 170 38.536.37.83 .79B Anger 170 24.677.47.88 .88B Discrete Emotions Emotional Scale – Faking and Suppressing 154 59.43 18.60 .91 Positive Emotion Faking 179 13.515.56.86 .87C Negative Emotion Faking 171 15.677.02.90 .88C Positive Emotion Suppression 172 9.354.44.76 .82C Negative Emotion Suppression 171 20.548.90.87 .94C Job Satisfaction 182 17.844.87.85 .76D Intention to Quit 183 7.643.18.78 .91E Performance 183 129.8657.80.77 N/A A Penner et al. (1995) B Goldberg (1999) C Glomb and Tews (2004) D Spector (1985) E Wiesberg and Sagie (1999) Five Factor Personality Four of the five factors of personality: extroversion, conscientiousness, agreeableness and anger as a facet of neuroticism, were measured using the 40 item IPIP (Goldberg, 1999) scale (see Appendix B). Each facet of personality was measured with 10 separate items. The items were presented in statement form and the participants

PAGE 59

51 indicated on a one to five Likert-type scale, their agreement with the statement presented. See Table 2 for computed alphas and alpha comparisons. Dependent Measures Emotional Dissonance The Discrete Emotions Emotional Labor Scale (DEELS), which was developed and validated by Glomb and Tews in 2004, was used to measure the level of emotional dissonance in the participants (see Appendix C). The DEELS scale consists of three subscales: genuine expression, faking, and suppression. However, since this study focuses on emotional dissonance that results from a difference between felt emotion and displayed emotion, only two of the three subscales (faking and suppression) were utilized in the study. Both subscales were comprised of 14 items, of which each item addressed a certain emotion ranging from irritation to enthusiasm. In the genuine expression subscale, the participants were asked to respond to the question, “How often do you genuinely express ________ when you feel that way?” The above blank indicated one of the 14 emotions listed below the question posited to the participants. The participants then indicated their response on a one to five Likert-type scale. The anchors on the scale were “I genuinely express this many times a day” (5), “I genuinely express this a few times a day” (4), “I genuinely express this a few times a week” (3), “I genuinely express this a few times a month” (2), and “I never genuinely express this” (1). In the faking subscale, the participants were asked to respond to the question, “How often do you express feelings of _____ on the job when you really do not feel that

PAGE 60

52 way?” The blank indicated one of the 14 emotions listed below the question posited to the participants. The participants were required to indicate their response on a one to five Likert-type scale. The anchors on the scale were “I express this many times a day when I do not feel it” (5), “I express this a few times a day when I do not feel it” (4), “I express this a few times a week when I do not feel it” (3), “I express this a few times a month when I do not feel it” (2), and “I never express this when I do not feel it” (1). In the suppression subscale, the same emotions were listed for the participants, but in this subscale, the respondents were responding to the question, “How often do you keep feelings of _____ to yourself when you really feel that way?” For this scale, the participants were required to indicate their response on a one to six Likert-type scale. The anchors on the scale were “I keep this to myself many times a day” (5), “I keep this to myself a few times a day” (4), “I keep this to myself a few times a week” (3), “I keep this to myself a few times a month” (2), “I never keep this to myself” (1) and “I never feel this” (0). In this subscale, the “I never feel this” the “0” score was to be treated as a missing value because if the research posited that if a participant never felt that emotion then it is impossible to suppress that emotion. Initial data analysis showed that when a “0” score was coded as a missing value, the number of participant responses available for analysis dropped from 183 to 81. Therefore, it was decided to retain the “0” value rather than recode it as a missing value. Although the DEELS scale had been just recently published, it had undergone rigorous validity testing. Convergent validity was established when the DEELS results were found to be significantly, positively related to both the Morris and Feldman

PAGE 61

53 dissonance subscale (correlations ranged from .21 to .31) and the Brotheridge and Lee’s surface acting scale (correlations ranged from .22 to .44). Discriminant validity was established by comparing the DEELS subscales to the Morris and Feldman’s duration dimension. In this comparison, each of the correlations between the DEELS subscales and the Morris and Feldman’s duration dimension were found to be non-significant. Criterion-related validity was assessed by examining the relationships between the faking and suppression subscales of the DEELS with a measure of emotional exhaustion. Again, the researchers found that five factors of the DEELS (faking positive, faking negative, suppressing positive, suppressing negative and genuinely expressing negative were significantly positively related to emotional exhaustion. Since both the faking and suppressing factors of the DEELS were found to have criterion related validity, it was anticipated that these two subscales would be aggregated for a total measure of emotional labor in the proposed study. However, prior to aggregation, the disattenuated correlation was calculated for these two scales and was found to be .46. Due to the low disattenuated correlation, a correlation matrix was prepared (see Appendix D) so that a visual inspection could be made between the variables of interest with the aggregated measure (Emotional Dissonance) as well as the two sub-scales, Faking Emotion and Suppressing Emotion. As can be seen in Appendix C, the strength and direction of the correlations remained the same for all variables, with the exception of Extraversion and Faking Emotion, whether the sub-scales remained separate or were aggregated. Therefore, due to the lack of change within the correlation matrix and because it was the recommendation of Tews and Glomb (2002), the developers of the DEELS, to aggregate

PAGE 62

54 the scales because emotional dissonance is composed of both faking and suppressing emotions, it was decided that these scales would be aggregated to create the emotional dissonance variable. Job Satisfaction Satisfaction with the nature of the work was measured by four items from the Job Satisfaction Survey (Spector, 1985). This subscale required the participants to indicate on a one to six Likert-type scale, their agreement with the statement presented (see Appendix E). See Table 2 for computed alpha and comparison. Intention to Quit Intention to quit was measured with three statements that required the participants to indicate on a one to five Likert-type scale, their agreement with the statement presented (see Appendix F). This measure was based on the Wiesberg and Sagie (1999) measure of intention to quit. The three questions were summed for an overall measure of intention to quit. See Table 2 for computed alpha and comparison. Performance Performance was measured by percentage of organizational goal obtained. Every department had a monthly goal that the debt collectors were to obtain. If the debt collectors exceeded the monthly goal, they were rewarded with a commission. If they did not reach the goal, the collectors were only paid their hourly wage for the time earned and, if they were with the organization for more than three months, were placed on notice of not reaching organizational goals which could result in dismissal. During the first month of employment, performance was found to be unreliable, therefore the decision

PAGE 63

55 was made to eliminate any participants who did not have a full two months worth of performance data. Two months of performance data for each participant were averaged, resulting in the performance data analyzed. The mean, standard deviation and alpha are reported in Table 2. This performance structure wherein collectors are given a specific goal that indicates a specific amount of debt they are expected to collect for the month, is typical within the collection industry. In addition, it is also typical within the collection industry to award a bonus to any collector who exceeds the performance goal. Procedure The researcher met with subgroups of the participant pool to inform the participants of the study. The subgroups ranged from groups of five to groups of 20. During that time, the participants were encouraged to participate but were informed that it was not required and that participation would not affect their current jobs in any manner. The participants were informed of the importance of responding to the survey questions in an honest manner. In addition, the participants were told of the coding structure utilized by the researcher. Participants were assured that the researcher was the only person who knew of the coding structure. The researcher outlined the security practices in place for the use of the coding structure and of the planned destruction of the coding sheet upon completion of the statistical analyses. At that time, all participants who wished to participate completed a Consent Form (Appendix G). Upon completion of the Consent form, the researcher gave all participants their particular code so that they could put the code onto the measures they completed.

PAGE 64

56 The participants were asked to complete a demographic sheet (Appendix H), the independent measures (Helpfulness and IPIP scales) as well as three of the dependent measures: Emotional Labor Scale, Job Satisfaction Survey and Intention to Quit. Once the participants had completed their measures, they left them in box at the rear of the room that was provided by the researcher. In addition, the participants were encouraged to take an additional copy of the Consent Form that they had signed. The performance data were obtained directly from the host organization. The presentation of the measures was not randomized or varied prior to having the participants complete them. Some researchers have argued that the order in which the measures were presented may have influenced the ensuing measures, which is also called a “priming” effect. However, O’Reilly and Caldwell (1979) found this not to be true when they experimentally manipulated the presentation of measures and found that “there were no significant variations attributable to the ordering of the questionnaire parts” (p. 160). In addition, Campion (1988) also experimentally manipulated the study’s measures presentation by using two alternate forms of a questionnaire to determine whether there was a priming effect. One form of the questionnaire presented the job design questions before the outcome measures. The alternate form switched the presentation so that the outcome measures were was presented before the job design questions. The results of this study revealed that there was a small mean difference in the job satisfaction outcome only (i.e., there was a difference in only one of four outcome measures).

PAGE 65

57 The current study is largely concerned with correlations among the measures rather than means. There is no evidence that the correlations among measures such as those used in the current study are affected by the order of scale presentation. Results Data Analysis All of the alphas reported above were computed using only participants who had answered every question on the appropriate measure. The number of such participants per scale can be seen in Table 2. Before I calculated the correlation matrix and path analysis, I decided to average the items within measures for extraversion, agreeableness, conscientiousness, anger, helpfulness, emotional dissonance, job satisfaction, and intention to quit to maximize the number of scale scores obtained. For example, if a participant did not answer one of the 10 questions asked for any personality measure, the average of the remaining responses for that party would serve as the response on that scale for that participant. In exploring the data, it was decided that this system would not compromise the data analyses because missing data accounted for a small percentage of the total data obtained. There was no participant who completed less than 80% of any measure. The frequency of the missing data is displayed in Table 3.

PAGE 66

58 Table 3 Frequency and Percentage of Missing Data Measure Total Number of Data Points Total Number of Missing Data Percentage Extraversion 1830 11 .6% Agreeableness 1830 15 .8% Conscientiousness 1830 13 .7% Anger 1830 13 .7% Emotional Labor 5124 29 .5% Job Satisfaction 732 1 .1% Intention to Quit 549 0 0% With a total participant pool of 183, the initial step was to compute a correlation matrix of all nine variables of interest. The correlations and descriptive statistics for the nine scales are shown in Table 4. Next, as a means of double checking the upcoming LISREL results and for better understanding of the underlying path analysis concepts, four separate regressions were computed. First, emotional labor was regressed on extraversion, anger, agreeableness, conscientiousness and helpfulness. Second, nature of work (a facet of job satisfaction) was regressed on emotional labor. Third, intention to quit was regressed on satisfaction with the nature of work. The final analysis involved regressing performance onto satisfaction with the nature of work. Because the last three regression equations involved only one variable being regressed onto another single variable, the beta weights obtained in these regressions were compared to the zero-order correlations (found in Table 4) to ensure that they were the same. The results of these regressions can be found in Table 5.

PAGE 67

59 Table 4 Means, Standard Deviations and Correlations between all Variables of Interest Variable N Mean SD Extra Agree Consc Anger HelpED JS ITQ Perf Extra 183 3.52 .69 1.00 Agree 183 3.93 .62.19** 1.00 Consc 183 3.85 .63-.03 .34** 1.00 Anger 183 2.47 .75.13 -.33**-.31** 1.00 Help 183 3.56 .54 .21* .39**.30**-.31** 1.00 ED 183 2.11 .67 -.01 -.17* -.25**.29** -.07 1.00 JS 183 4.45 1.22-.09 .03 .26** -.13 .10-.18* 1.00 ITQ 183 2.55 1.06.04 -.04 -.16* .12 .04.14*-.63** 1.00 Perf 183 129.86 57.80.00 -.08 .01 .10 .03 .01 .18**-.20**1.00 EXTRA = Extraversion, Agree = Agreeableness, Consc = Conscientiousness, Help = Helpfulness, ED = Emotional Dissonance, JS = Job Satisfaction, ITQ = Intention to Quit, Perf = Performance *p<.05 ** p<.01

PAGE 68

60 Table 5 Regression Analyses Dependent Variables: Emotional Labor Independent Variables R2 F Beta t .12 4.93** Extraversion -.06 -.80 Agreeableness -.05 -.68 Anger .25 3.16** Conscientiousness -.18 -2.34* Helpfulness .10 1.20 Job Satisfaction (Nature of Work) .03 6.07* Emotional Labor -.18 -2.46** Intention to Quit .39 116.49** Job Satisfaction (Nature of Work) -.63 -10.79** Intention to Quit .04 7.20** Performance -.20 -2.68** The data for all 183 participants were then entered into LISREL program for path analyses. With a correlation matrix, path coefficients resemble correlation coefficients and are easier to interpret. With the covariance matrix, path coefficients can range far above or below the 1.0, making interpretation a little more difficult. However, some goodness-of-fit tests are based on assumptions only met by the covariance matrix and not the correlation matrix. For these reasons, the correlation matrix path analysis was used to

PAGE 69

61 estimate path coefficients and the covariance matrix path analysis was used to test the model’s fit. The correlation path analyses results were checked against the correlation matrix (Table 4) and the regressions (Table 5) and found to be in alignment, thus giving confidence in the results. In order to show the path model coefficients and their relationship to the hypotheses, Figure 5 depicts the model with the associated path coefficients. For purposes of clarity, the relations among the variables were explored first to determine support or non-support of the hypotheses and then the overall fit of the model was examined. Results are presented in this fashion because, despite overall model fit, the relations between the antecedents and consequences of emotional dissonance (that is, specific parts of the model) are still interesting because this type of research has not been performed in the past.

PAGE 70

62 Figure 5 – Path Model including Path Coefficients Extraversion N euroticis m (Anger) Agreeableness Conscientiousness Helpfulness Emotional Dissonance Job Satisfaction (Nature of Work) Intention to Quit Performance -.18* -.06 .25* -.06 -.18* .10 .18* -.63*

PAGE 71

63 Tests of Hypotheses with Path Analysis As can be seen in Figure 5, the path coefficient from Extraversion to Emotional Labor was -.06, but not significant. Therefore, Hypothesis 1 was not supported. The path coefficient from Anger to Emotional Labor was significant at .25 but in the opposite direction of that hypothesized and, therefore, Hypothesis 2 was not supported. The path coefficient from Agreeableness to Emotional Labor (-.06) was not significant, and was in the opposite direction of that hypothesized. Therefore, Hypothesis 3 was not supported. The path coefficient from Conscientiousness to Emotional Labor (-.18) was significant and supports Hypothesis 4. The path coefficient from Helpfulness to Emotional Labor (.10) was not significant and therefore Hypothesis 5 was not supported. The path coefficient from Emotional Labor to Job Satisfaction (Nature of Work) was significant in the predicted direction, thus supporting Hypothesis 6. Hypothesis 7 predicted a negative path coefficient from Job Satisfaction (Nature of Work) to Intention to Quit, and this hypothesis was supported with a significant path coefficient of -.63. Finally, the path coefficient from Job Satisfaction (Nature of Work) and Performance was significant at .18, thus supporting Hypothesis 8. Once the path analysis was completed, the model was tested for fit. All goodness of fit testing was done utilizing the covariance matrix path analysis results. The first overall test for fit was the chi-square ( "2). When analyzing fit via "2, using the covariance matrix is one of the four assumptions that Pedhazur (1997) stated is necessary in order to have credible interpretation of results. The "2 for this model was 166.95 (p<.00) which indicated that this model varied significantly from the hypothetical perfect fit model.

PAGE 72

64 Although the chi-square did not support the fit of the model, chi-square is widely criticized as being very sensitive to sample size and rarely supports the fit of the hypothesized model. The next method of determining goodness of fit was the GFI. The GFI represents a standardized overall measure of fit that is based on the observed and reproduced values of the covariance matrix. The GFI can range from 0, which would represent no fit to 1, which would represent a perfect fit. Because the GFI tends to run higher than other fit measurements, the standard convention for interpreting the GFI is to set a cut-off of .90 or above which would represent a good fit. The GFI for the proposed model was .88, just barely below the cut-off point set for supporting model fit. Next, the Bentler-Bonett Normed Fit Index (NFI) was evaluated. The NFI assesses the incremental fit of models by comparing the proposed model to a null model. The NFI can range from 0 to 1, where a NFI of .50 would mean that the proposed model improves fit by 50% compared to the null model. The standard cut-off value for the NFI is .90 and that cut-off was used. The NFI for the hypothesized model was .43, far below the cut-off set prior to analyzing the data. Also calculated was the Bentler-Bonett Nonnormed Fit Index (NNFI). The NNFI is very similar to the NFI except that there is an adjustment for degrees of freedom. Unlike the NFI, the NNFI can result in values that are outside of the 0-1 range. A negative NNFI indicates that the null model is a better fit than the proposed model. Conversely, a positive NNFI reflects that the proposed model is a better fit than the null model. Standard convention for the interpretation of the NNFI is the same as the NFI

PAGE 73

65 with a cut-off of .90, which is the cut-off that was used for this study. The NNFI for the proposed model was .081, again below the cut-off set prior to analyzing the data. Finally, the standardized root mean square residual (SRMR) was evaluated. The Standardized RMR takes the observed covariance matrix and compares it to the covariance matrix based on the model. Differences between the observed and predicted values are squared, averaged and rooted. The result may range from 0 which would represent a perfect fit to 1 which would represent no fit. It was anticipated that a low SRMR would exist thus adding further support for the proposed model. The SRMR for the model was .26, an indication of poor fit. SRMR values greater than .10 are generally considered large. Possible reasons for the lack of fit of this particular model are addressed in the discussion. Gender Differences To explore whether there are gender differences in the antecedents and consequences of emotional labor, two additional path models were produced, one for females and one for males. Correlation matrices were produced as well for both the females and the males. See Table 6 for the female matrix and Table 7 for the male matrix. The female model with path coefficients can be seen in Figure 6. The male model with path coefficients can be seen in Figure 7. As can be seen in these two figures, there were gender differences in the path coefficients.

PAGE 74

66 Figure 6 – Path Model including Path Coefficients for Females Extraversion N euroticis m (Anger) Agreeableness Conscientiousness Helpfulness Emotional Dissonance Job Satisfaction (Nature of Work) Intention to Quit Performance .09 -.08 .20 -.07 -.06 .05 .19 -.63*

PAGE 75

67 Figure 7 – Path Model including Path Coefficients for Males Extraversion N euroticis m (Anger) Agreeableness Conscientiousness Helpfulness Emotional Dissonance Job Satisfaction (Nature of Work) Intention to Quit Performance .28 -.03 .34* .0 1 -.31* .16 .21 -.61*

PAGE 76

68 Table 6 Female Means, Standard Deviations and Correlations between all Variables of Interest Variable N Mean SD Extra Agree Consc Anger Help ED JS ITQ Perf Extra 103 3.50 .73 1.00 Agree 103 4.04 .62 .17 1.00 Consc 103 3.94 .62-.07 .33** 1.00 Anger 103 2.53 .75.15 -.36**-.32** 1.00 Help 103 3.59 .59 .18 .43**.31**-.34** 1.00 ED 103 2.07 .68 -.05 -.15 -.13 .22** -.08 1.00 JS 103 4.61 1.21-.13 .01 .23** -.17 .10 -.09 1.00 ITQ 103 2.46 1.07.01 -.09 -.10 .13 .02 .16 -.63** 1.00 Perf 103 125.18 50.51.07 -.05 .01 .08 .06 .03 .19 -.20*1.00 EXTRA = Extraversion, Agree = Agreeableness, Consc = Conscientiousness, Help = Helpfulness, ED = Emotional Dissonance, JS = Job Satisfaction, ITQ = Intention to Quit, Perf = Performance *p<.05 ** p<.01

PAGE 77

69 Table 7 Male Means, Standard Deviations and Correlations between all Variables of Interest Variable N Mean SD Extra Agree Consc Anger HelpED JS ITQ Perf Extra 80 3.55 .65 1.00 Agree 80 3.80 .61 .24* 1.00 Consc 80 3.72 .62.05 .30** 1.00 Anger 80 2.38 .74.11 -.34**-.36** 1.00 Help 80 3.54 .46 .26* .31**.28**-.30** 1.00 ED 80 2.17 .65 .03 -.18 -.40** .41** -.04 1.00 JS 80 4.26 1.21-.02 .01 .26* -.12 .10 -.28* 1.00 ITQ 80 2.67 1.04.06 .05 -.20 .12 .09 .11 -.61** 1.00 Perf 80 135.88 65.86 -.07 -.09 .04 .15 .00 -.02 .21 -.22*1.00 EXTRA = Extraversion, Agree = Agreeableness, Consc = Conscientiousness, Help = Helpfulness, ED = Emotional Dissonance, JS = Job Satisfaction, ITQ = Intention to Quit, Perf = Performance *p<.05 ** p<.01

PAGE 78

70 First, to determine whether there were overall gender differences on any of the individual variables, t-tests were computed. The t-tests showed significant differences for three variables: conscientiousness, agreeableness and job satisfaction. In all three cases, the females scored significantly higher than the males. Turning to the path models, several differences were found as well. To begin with, for the females, there were no significant path coefficients for any of the five antecedents to emotional dissonance. The males, however, had significant path coefficients from both anger (.34) and conscientiousness (-.31) to emotional dissonance. To determine whether those differences were significant, testing was done with two ANCOVAs. In the ANCOVAs, all the personality variables were entered as independent variables and emotional dissonance was entered as the dependent variable. In addition, in the first ANCOVA, an interaction term was added to determine whether there was an interaction between gender and conscientiousness. In the second ANCOVA, an interaction term was added to determine whether there was an interaction between gender and anger. A significant interaction term would signify that the male and female path coefficients noted above were different. Unfortunately, this testing resulted in nonsignificant results for both conscientiousness and anger. With respect to the relationship between emotional dissonance and job satisfaction, the male path coefficient (-.28) was significant whereas the lower female path coefficient (-.09) was not significant. Because the relationship between emotional dissonance and job satisfaction was one-to-one (a single path), testing for differences in

PAGE 79

71 path coefficients by gender was completed by testing the equality of two independent correlations. The test for the difference was not statistically significant (z = -1.30, ns). The path coefficient from satisfaction with nature of work and intention to quit remained approximately the same for both females (-.63) and males (-.61) and both male and female path coefficients were significant. Both the female path coefficient (.19) and the male path coefficient (.21) from satisfaction with nature of work to performance was positive, but neither path coefficients was significant. Goodness of fit indices were also examined for the models based on male and female data. These fit indices followed the same pattern as the main proposed model with significant chi-square results for both male and female path models, and GFIs, NFIs and NNFIs were less than the cut-off of .90 for both male and female models. Finally, the SRMR of .31 for the male model and .25 for the female model were both large. Post-Hoc Gender Analyses Due to the gender differences found in this study and detailed above, post-hoc analyses were performed to determine whether there were gender differences on any of the demographic variables. The results of these analyses can be found in Table 8. The only demographic variable that differed as function of gender was tenure, with the females having significantly longer tenure than the males.

PAGE 80

72 Table 8 Post-Hoc Gender Demographic Comparisons Males Females t Test Variable N Mean SD N Mean SD df t Value Age 80 28.38.44103 27.7610.17 181 -1.05 Race1 80 2.95.55101 2.97.57 179 -0.24 Experience2 80 21.1825.40103 29.3837.68 181 -1.67 Tenure3 80 2.61.14103 2.261.14 181 1.98* 11=Asian, 2=Black, 3=Caucasian, 4=Hispanic, 5=Native American, 6=Other 2 Experience was measured in months 3 Tenure was measured in months p<.05 Discussion The primary goal of this research was to test a model of the antecedents and consequences of emotional dissonance. It was hypothesized that certain personality characteristics may serve as antecedents by either lessening or heightening the amount of experienced emotional dissonance experienced. More specifically, it was posited that having personality characteristics that assist in aligning the experienced emotions with the required emotional displays would result in lower emotional dissonance. Emotional dissonance was hypothesized to result in several different consequences. Lower emotional dissonance was predicted to be associated with higher satisfaction with the nature of the work and consequently higher job performance and lower intention to quit. Satisfaction with the nature of the work, intention to quit and performance would thus be considered consequences of emotional dissonance. A path model was developed to summarize and test the plausibility of the hypothesized relations. Unfortunately, the proposed model was not supported well. In

PAGE 81

73 general, it appeared that the part of the path model concerned with the consequences of emotional dissonance was more supportive of the hypotheses than was the part of the path model concerning antecedents. One reason for the lack of firm support may be that the model does not accurately reflect the flow of the variables. More specifically, it is possible that the personality factors do not have only indirect effects on the proposed consequences of job satisfaction, performance and intention to quit. Quite possibly, these factors have direct effects on the consequences of interest. In addition, emotional dissonance may not show its influence solely through job satisfaction; emotional dissonance may also have direct effects on both job satisfaction and intention to quit. Turning to the correlation matrix to explain the lack of fit of the model, one can see several interesting relationships. First, conscientiousness was significantly related to job satisfaction (r = .26, p <.01) and to intention to quit (r = -.16, p < .05). Based on these findings, it would appear that conscientiousness has a direct relationship with both job satisfaction and intention to quit. Interestingly, though, conscientiousness had virtually no relationship with performance (r = .01, ns). In addition, the Helpfulness measure was significantly correlated with all the personality facets. More specifically, the relationship between Helpfulness and Agreeableness (r = .39, p<.01) was contrary to the prior findings of Penner and Fritzsche (1993b) who found a nonsignificant correlation between Helpfulness and Agreeableness (r = .11, ns). The current finding is what Penner and Fritzsche (1993b) expected to find,

PAGE 82

74 but did not. Theoretically, Helpfulness and Agreeableness should be correlated because they both share the personality facet Altruism. In reviewing the hypotheses, some of the results are rather surprising. The results showed that Anger was positively related to emotional dissonance rather than negatively as predicted. This result means that collectors who feel more anger, also tend to experience the most emotional dissonance. This relationship may reflect the host organization’s culture rather than debt collectors in general. At this host organization, the preferred interpersonal style was to remain calm and courteous to all debtors. In the host organization’s opinion, it was likely that the debtor was being contacted by multiple debt collection companies and therefore had a choice as to whom he or she would pay. Further, it was the host organization’s position that the nicer the collector was to the debtor, the more likely collection would be. It is possible, therefore, that angry collectors were unable to express their anger to their clients and thus experienced considerable dissonance. Such a process could explain the unexpected finding. The direction of the relationship between Agreeableness and emotional dissonance is in the opposite direction as that hypothesized and Agreeableness was, in fact, negatively related to emotional dissonance. In addition, the positive relationship of Helpfulness with emotional dissonance was also opposite that hypothesized in this study. Even more surprising was that, although there was a significant positive correlation between Helpfulness and Agreeableness, their adjusted relationships, as evidenced by their path coefficients, with emotional dissonance were opposite. This finding indicates

PAGE 83

75 that a suppressor variable, a variable that masks the relationship between other variables, is present. This finding needs to be replicated. Fortunately, this study was able to show that two personality traits, Anger and Conscientiousness, do appear to be related to emotional dissonance in debt collectors. These findings both confirm and extend the current literature. Tews and Glomb (2003) also found a negative correlation (r = -.26 p<.05) between conscientiousness and suppressing negative emotions. This study confirms that finding and extends it to faking emotions as well. In addition, Anger as a facet of Neuroticism, has never been explored as an antecedent of emotional dissonance. This finding gives future researchers a starting point for future reference. Overall, it was disappointing that not all personality traits were shown to be significant antecedents of emotional dissonance. One of the reasons that personality may not be an antecedent in an industry known for their high levels of emotional labor may be because of deep acting. Debt collecting can be an unforgiving job, and being on the phone constantly asking people for money may be extremely grueling. However, the money that can be made in this is very alluring. Consequently, good debt collectors may be engaging in deep acting rather than surface acting to get them through the day. If this is true, they would experience little emotional dissonance. Since an emotional dissonance scale was used in this study rather than a scale which measured surface and deep acting, it is impossible to determine whether deep acting affected emotional dissonance in this sample. Future researchers may want to turn their attention to how debt collectors manager their emotional dissonance. Along those lines, future researchers may also want

PAGE 84

76 to look at the relationship between personality and surface and deep acting to determine whether different personality types are more likely to engage in surface acting or deep acting. It is possible that the way debt collectors manage their emotional dissonance (either through deep acting or surface acting) may be serving as a moderator between personality and emotional dissonance. Another possibility as to why personality is not an antecedent for emotional dissonance in debt collectors is based on the James-Lange Theory (Myers, 1986). According to this theory, emotion follows bodily responses. Simply put, the body acts or reacts and emotion follows. If this is true, then debt collectors would be angry with a debtor and their body would then follow suit and give off physiological signs that it is angry. This theory even extends to facial gestures such as smiling which would then generate physiological responses that tell the human brain that it is happy. Therefore, based on this theory, personality would be irrelevant because, once the debt collector initiates the behavior, the body responds accordingly. Following this line of thought, if the body is responding accordingly and giving the brain physiological messages to confirm the behavior, there would be a decrease in the incongruence between felt and displayed emotion. In this scenario, displayed emotion would dictate experienced emotion. Although this is a possible explanation of experienced emotion, if taken at its face value, this theory would mean that there is no such thing as emotional dissonance. However, due to the findings in this study as well as other studies (e.g. Abraham, 1999; Morris & Feldman, 1996) it is clear that emotional dissonance does exist and, therefore, it is not really plausible that this theory provides a complete explanation of the findings.

PAGE 85

77 An additional area in which this study contributes to the literature is the replication of the negative association between dissonance and job satisfaction (Abraham, 1998). The additional benefit of this study is that a facet of job satisfaction was used rather than a global measure thus extending previous findings to now include that increased emotional dissonance results in decreased satisfaction with the nature of the work. Two other important replications were the relationships between satisfaction with the nature of the work and intention to quit (Tett & Meyer, 1993), and satisfaction with the nature of the work and performance (Iaffaldano & Muchinsky, 1985). The strength of the relationship between satisfaction with the nature of the work and intention to quit was surprising in its strength (r = -.63) because this is a much stronger relationship compared to Spector’s (1985) meta-analyses result of -.32. One possible reason why this relationship was so strong is that turnover is extremely prevalent in the debt collection industry so that the participants in this study had no hesitation to truthfully respond about their desire to leave the organization. The positive relationship between satisfaction with the nature of the work and performance is an extension of previous research showing that there is a small but positive relationship between these two variables. The finding in this study was very similar to that of Iaffaldano and Muchinsky (1985) who found that, overall, the relationship between job satisfaction and performance was small (.146 uncorrected; .17 corrected) but positive. Both of these satisfaction findings are of great interest because

PAGE 86

78 they represent a population that has remained relatively untapped therefore, adding credibility to previously found results. Overall, although the path model failed to support a number of hypotheses, this research still contributes to the study of emotional labor. Although many researchers use debt collectors as an illustration of the large range emotional labor and emotional dissonance can take, this study is the first to examine empirically the antecedents and consequences of emotional dissonance in debt collectors. This research is the start of being able to map out exactly what the differences are between the “heel and toe” (Hochschild, 1983) of emotional labor and emotional dissonance. The main implication from this study is that in debt collection, just as in customer service, emotional dissonance affects job satisfaction, which in turn affects both turnover intentions and performance. Organizations that understand the concept of emotional dissonance and feel that their employees may be experiencing greater levels of emotional dissonance should react quickly to avoid the serious implications that emotional dissonance can have on their organization. Organizations may want to work with their employees so that they could employ deep acting techniques that may decrease the emotional dissonance. Although deep acting may decrease emotional dissonance, it has not yet been proven that it also reduces stress. Another avenue that organizations may want to explore is attempting to find a way so that employees would not have to suppress emotions for long periods of time. One possibility would be to allow employees to vent or release their suppressed emotions outside of their customers/clients’ presence, either in person or via the phone, so that over time, the employees do not feel that they are

PAGE 87

79 continually suppressing emotion. Unfortunately, utilizing a personality tool to select employees who are less predisposed to suffer from the rigors of emotional dissonance does not seem to be the answer at this time. Future researchers should continue the search to find a selection tool that can discern those who are predisposed to experience greater levels of emotional dissonance. In addition, future researchers should not shy away from using debt collectors as a sample to explore the depth and range of emotional dissonance. This study confirmed that consequences of emotional dissonance do not change when the type of emotional labor being performed is changed. Unfortunately, this study was unable to determine specific personality antecedents of emotional dissonance for debt collectors. Although disappointing, it is interesting that personality did not predict emotional dissonance well considering that Tews and Glomb (2003) argued that personality plays a role in emotional labor. More specifically, Tews and Glomb (2003) felt that extraversion and neuroticism are two of the most relevant personality characteristics for the emotional labor domain (p. 18). The inability to replicate the extraversion findings and by finding that a facet of neuroticism works as well, may mean that the most relevant personality characteristics for the emotional labor domain are population sensitive. In other words, by using debt collectors rather than customer service agents, this study may hint at the importance of the type of emotional dissonance being experienced or emotional labor being performed. It is possible that previous findings may not generalize outside of the customer service arena. Future researchers may want to focus on different types of

PAGE 88

80 industries to see how well the previous work on emotional dissonance and emotional labor can be replicated. An additional avenue for future researchers would be to replicate this study in an urban area. This study was conducted using debt collectors from rural mid-western areas. The personality make-up of these types of individuals may vary drastically from those in more urban areas and in areas where there is more competition for jobs within the debt collection company. In addition, in this study, the host organization’s primary method of collecting debts was to be courteous and considerate of the debtor. Rudeness or aggressive language was the exception to the rule rather than the rule in this situation. There are many debt collection companies currently in operation today that do not use this same method of collecting and it would be interesting to see how this study replicates in a different organizational culture. It is quite possible that some of the surprising findings in this study were the result of the organizational culture that may be atypical of the debt collection industry. Unfortunately, since the host organization for this study was a sample of convenience, the unusual organizational culture and collection methodology of this debt collection agency was unknown prior to the study being performed. Although neither the male or female model showed good fit to the data, there were some interesting relations. Previous research has indicated that females may have to perform more emotional labor than males (Deusch, 1990; Hochschild, 1989). However, in this study there were no significant differences in the amount of reported emotional dissonance between males and females.

PAGE 89

81 In this study, the relationship between satisfaction and dissonance appeared stronger for males than for females. One possible explanation for this difference is that females expect to perform emotionally taxing work and therefore experience less distress than do males. It is possible that the smaller correlation between emotional dissonance and job satisfaction that is seen in the female participants may be the result of females being able to manage their emotions better, therefore not affecting separate emotions such as how satisfied they are with their work. If this is true, then Grandey (2000) was correct when she stated that “men may need more training to manage emotions when dealing with customers” (p. 106). Along those lines, another interesting finding was that none of the path coefficients from the antecedents to emotional dissonance were significant for the females, whereas two of the path coefficients from personality (anger and conscientiousness) to emotional dissonance were significant for males. This finding may mean that males and females have different antecedents of emotional dissonance. These antecedents may differ across gender in other areas than personality, such as emotional management style, social support or organizational culture. Different types of antecedents for males and females is an area that future researchers should explore. As to whether females are more predisposed to suppress anger, this study would indicate that this is not the case. Although not the main thrust of this study, the measures used in this study allowed this question to be looked at in greater detail. An independent t-test was computed on item 13 of the Emotional Suppression portion of the DEELS. Item 13 specifically asked the participants to indicate how often they felt they needed to

PAGE 90

82 suppress anger. The results show that there were no significant differences between males and females. To contrast, the results showed that the female path coefficient from anger to emotional dissonance was not significant whereas the male path coefficient was. Although the difference was not significant, the trend in the data may suggest that as males score higher on the anger trait, they experience greater levels of emotional dissonance. This finding could support Sharkin’s (1993) theory that male anger is seen as something that needs to be controlled. Therefore, males may be more likely to attend to their anger. Taken together with the Milovchevich, Howells, Drew and Day’s (2001) interpretation that females learn to control anger experience and expression and that that gender roles are taught and validated so that males experience greater intensity of anger and outward behavioral responses, it would be much more difficult for males (or those categorized as exhibiting male gender roles) to suppress anger. Due to this difficulty in suppression, males would experience more emotional dissonance as their anger increased. Another interesting gender difference relates to the difference in path coefficients from conscientiousness to emotional dissonance. Again, the males had a significant negative path coefficient (-.31) while the females has a nonsignificant (-.06) path coefficient. This finding can be interpreted that for males, the less conscientious a male is, the more emotional dissonance they experience. It is possible that the highly conscientious take some comfort from the role prescribed by the job when dealing with awkward personal situations because of their sense of duty. That is, they may feel less need to suppress their feelings because their job justifies or requires their behavior.

PAGE 91

83 Again, this finding may simply mean that this personality facet is a better predictor of emotional dissonance for males than it is for females. There may be different antecedents for females than there are for males. Future researchers should explore this issue. It should be noted that any of the gender differences in this study should be interpreted with caution. It is possible that the differences that were found in this study could be the result of extraneous variables which were not controlled for in this study. The significant correlation between tenure and gender which was found in this study is an example of one of these extraneous variables. One possible explanation for this relationship would support the findings in this study in that the females may stay longer with an organization because their job satisfaction is not affected by emotional dissonance. This could also explain the significant difference between males and females and their ratings of job satisfaction with females indicating significantly higher [t(181) = -1.97, p< .05] levels of job satisfaction. However, these relationships could be due to a host of other reasons such as flex time, benefits and organizational culture. While this study found very interesting results with regard to gender differences, this is an area that needs to be investigated further by future researchers. Future researchers may want to explore whether the known consequences of emotional dissonance differ by gender and whether there are moderator or mediator effects. In summary, this research was a preliminary attempt to ascertain whether certain personality traits could be viewed as antecedents to emotional dissonance and whether the known consequences of emotional dissonance would replicate in the debt collection industry. Although only two personality facets, Anger and Conscientiousness, were

PAGE 92

84 found to be related to emotional dissonance, this does not mean that other aspects of personality are not antecedents of emotional dissonance. Future research should focus not only on which factors are antecedents of emotional dissonance but also on what the relationships are between those personality factors and other outcome variables. It is promising to see that the consequences of emotional dissonance have replicated in an arena that represented the opposite end of emotional dissonance from customer service representatives. An added bonus of this research was to find that gender may have a greater effect on certain outcome variables than was previously thought.

PAGE 93

85 References Abraham, R. (1998). Emotional dissonance in organizations: Antecedents, consequences, and moderators. Genetic, Social, and General Psychology Monographs, 124 (2), 229-247. Abraham, R. (1999). The impact of emotional dissonance on organizational commitment and intention to turnover. Journal of Psychology 133(4), 441-455. Arnold, H.J. & Feldman, D.C. (1982). A multivariate analysis of the determinants of job turnover. Journal of Applied Psychology, 67 350-360. Ashforth, B.E. & Humphrey, R.H. (1993). Emotional labor in service roles: The influence of identity. Academy of Management Review,18, 1, 88-116. Averill, J.R. (1983). Studies on anger and aggression. American Psychologist, 38 1145-1160 Barrick, M.R. & Mount, M.K. (1991). The big five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44 1-26. Biaggio, M.K. (1989). Sex differences in behavioral reactions to provocation of anger. Psychological Reports, 64 23-26. Birnbaum, D.W., Nosanchuk, T.A., & Croll, W.L. (1980). Children’s stereotypes about sex differences in emotionality. Sex Roles 6, 435-443. Bluedorn, A.C. (1982). A unified model of turnover from organizations. Human Relations, 35 135-153. Borgatta, E.F. (1964). The structure of personality characteristics. Behavioral Science, 12 8-17.

PAGE 94

86 Brebner, J. (2003). Gender and emotions. Personality and Individual Differences, 34, 387-394. Brotheridge, C.M. (April, 2003). Predicting emotional labor given situational demands and personality. Paper presented at the 18th annual meeting of the Society for Industrial and Organizational Psychology, Orlando, Florida. Buntaine, R.L. & Costenbader, V.K. (1997). Self-reported differences in the experience and expression of anger between girls and boys. Sex Roles, 36, (9-10), 625637. Campion, M.A. (1988). Interdisciplinary approaches to job design: A constructive replication with extensions. Journal of Applied Psychology. Vol 73 (3), 467-481. Costa, P.T. & McCrae, R.R. (1988). From catalog to classification Murray’s needs and the five-factor model. Journal of Personality & Social Psychology, 55, 258265. Costa, P.T., & McCrae, R.R. (1992). NEO PI-R Professional manual Odessa, FL: Psychological Assessment Resources. Cote, S. & Morgan, L.M. (2002). A longitudinal analysis of the association between emotion regulation, job satisfaction, and intentions to quit. Journal of Organizational Behavior, 32 947-962. Deutsch, F.M. (1990). Status, sex, and smiling: The effect of role on smiling in men and women. Personality and Social Psychology Bulletin 16, 531-540.

PAGE 95

87 Diefendorff, J.M. & Gosserand, R.H. (2003). Understanding the emotional labor process: A control theory perspective. Journal of Organizational Behavior, 24 945-959. Dyer, L., & Theriault, R. (1976). The determinants of pay satisfaction. Journal of Applied Psychology, 61 596-604. Freifeld, T. (1993). Nurturance and prosocial personality tendencies Unpublished manuscript, University of South Florida, Tampa, FL. Frijda, N.H., Ortony, A., Sonnemans, J. & Clore, G.L. (1992). The complexity of intensity: Issues concerning the structure of emotion intensity. In M. Clark (Ed.) Review of Personality and Social Psychology 13, 60-89. Newbury Park, CA: Sage. Gianakos, I. (2002). Issues of anger in the workplace: Do gender and gender role matter? Career Development Quarterly, 51, (2), 155-171. Glomb, T.M. & Tews, M.J. (2004). Emotional labor: A conceptualization and scale development. Journal of Vocational Behavior, 64, 1-23. Goldberg, L.R. (1981). Language and individual differences: The search for universals in personality lexicons. In Wheeler L. (Ed.), Review of Personality and Social Psychology, 2, 141-166. Beverly Hills, CA: Sage Publications. Grandey, A.A. (2000). Emotion regulation in the workplace: A new way to conceptualize emotional labor. Journal of Occupational Health Psychology, 5, 95-100. Hackett, R.D., Bycio, P., & Hausdorf, P.A. (1994). Further assessments of Meyer and Allen’s (1991) three-component model of organizational commitment. Journal of Applied Psychology, 79 15-23.

PAGE 96

88 Hackmann, J.R. & Oldham, G.R. (1975). Development of the job diagnostic survey. Journal of Applied Psychology, 60 159-170. Hochschild, A.R. (1983). The managed heart: Commercialization of human feeling Berkeley, CA: University of California Press. Hoffman, L.W. (1972). Early childhood experiences and women’s achievement motives. Journal of Social Issues 28, 129-155. Hollenbeck, J.R., & Williams, C.R. (1986). Turnover functionality versus turnover frequency: A note on work attitudes and organizational effectiveness. Journal of Applied Psychology, 71 606-611. Iaffaldano, M.T. & Muchinsky, P.M. (1985). Job satisfaction and job performance: A meta-analysis. Psychological Bulletin, 97 251-273. Ironson, G.H., Smith, P.C., Brannick, M.T., Gibson, W.M. & Paul, K.B. (1989). Constitution of a job in general scale: A comparison of global, composite, and specific measures. Journal of Applied Psychology, 74 193-200. Johnson, H.M. (2004). The story behind service with a smile: The effects of emotional labor on job satisfaction, emotional exhaustion, and affective well-being. Unpublished master’s thesis, University of South Florida, Tampa. Kirchner, W. K. (1965). Relationships between general and specific attitudes toward work and objective job performance for outdoor advertising salesmen. Journal of Applied Psychology, 49 455-457. Kline, Rex B. (1998). Principles and practice of structural equation modeling NY: Guilford Press.

PAGE 97

89 Kring, A.M., Smith, D.A. & Neale, J.M. (1994). Individual differences in dispositional expressiveness: Development and validation of the emotional expressivity scale. Journal of Personality and Social Psychology, 66, (5) 934-949. Kruml, S.M. & Geddes, D. (2000). Exploring the dimensions of emotional labor. Management Communication Quarterly, 1, (14) 8-49. LaFrance, M. & Banaji, M. (1992). Toward a reconsideration of the genderemotion relationship. In M. Clark (Ed.), Review of Personality and Social Psychology Vol. 14, 178-197. Newbury Park, CA: Sage. Mabe, P.A. & West, S.G. (1982). Validity of self-evaluation of ability: A review and meta-analysis. Journal of Applied Psychology, 67, 280-296. Mathieu, J.E. & Zajac, D.M. (1990). A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment. Psychological Bulletin, 108 171-194. McCrae, R.R. & Costa, P.T. (1985). Updating Norman’s “adequate taxonomy”: Intelligence and personality dimensions in natural language and in questionnaires. Journal of Personality & Social Psychology, 49, 710-721. McCrae, R.R. & Costa, P.T. (1987). Validation of the five-factor model of personality across instruments and observers. Journal of Personality & Social Psychology, 52, 81-90. McCrae, R.R. & Costa, P.T. (1989). The structure of interpersonal traits: Wiggin’s circumplex and the five-factor model. Journal of Personality & Social Psychology, 56, 586-595.

PAGE 98

90 Milovchevich, D., Howells, K., Drew, N. & Day, A. (2001). Sex and gender role differences in anger: An Australian community study. Personality and Individual Differences, 31, 117-127. Morris, J.A. & Feldman, D.C. (1996). The dimensions, antecedents, and consequences of emotional labor. Academy of Management Review, 21, (4) 986-1010. Morris, J.A. & Feldman, D.C. (1997). Managing emotions in the workplace. Journal of Managerial Issues, 9, (3), 257-275. Mount, M.K., Barrick, M.R. & Stewart, G.L. (1991). Five-factor model of personality and performance in jobs involving interpersonal interactions. Human Performance, 11 145-165. Myers, D.G. (1986). Psychology New York, NY: Worth Publishers, Inc. O'Reilly, C.A. & Caldwell, D.F. (1979). Informational influence as a determinant of perceived task characteristics and job satisfaction Journal of Applied Psychology. 64 (2), 157-165. Peabody D. & Goldberg, L.R. (1989). Some determinants of factor structures from personality-trait descriptors. Journal of Personality & Social Psychology, 75 552567. Pedhazur, E.J. (1997). Multiple regression in behavioral research: Explanation and prediction United States: Thomson Learning, Inc. Penner, L.A., & Fritzsche, B.A. (1993a, June). The measurement of individual differences in prosocial tendencies. Paper presented at the Sixth International Conference on Personality and Social Behavior, Boca Raton, FL.

PAGE 99

91 Penner, L.A., & Fritzsche, B.A. (1993b, August). Measuring the prosocial personality: Four construct validity studies. Paper presented at the annual meeting of the American Psychological Association, Toronto, Canada. Penner, L.A., Fritzsche, B.A., Craiger, J.P. & Freifeld, T.S. (1995). Measuring the prosocial personality. In J. Butcher & C.D. Spielberger (Eds.), Advances in Personality Assessment (Vol. 10). Hillsdale, NJ: Lawrence Erlbaum. Petty, M.M., McGee, G.W. & Cavender, J.W. (1984). A meta-analysis of the relationships between individual job satisfaction and individual performance. Academy of Management Review, 9 712-721. Poindexter, K. (1994). The role of parental nurturance in the development of efficacy and prosocial tendencies. Unpublished manuscript, University of Arkansas, Fayetteville, AR. Rafaeli A., & Sutton, R. (1987). Expression of emotion as part of the work role. Academy of Management Review, 12 23-37. Schnaue, F. (2005, February 3). Service sector growth slows. Washington Times [On-line]. Available: http://washtimes.com/upi-breaking/20050203-113635-7396r.htm. Sharkin, B.S. (1993). Anger and gender: Theory, research, and implications. Journal of Counseling and Development, 71, (4), 386-389. Sharkin, B.S. & Gleso, C.J. (1991). The anger discomfort scale: Beginning reliability and validity data. Measurement and Evaluation in Counseling and Development, 24 61-68.

PAGE 100

92 Smith, P.C., Kendall, L.M. & Hulin, C.L. (1969). Measurement of satisfaction in work and retirement Chicago: Rand McNally. Spector, P.E. (1985). Measurement of human service staff satisfaction: Development of the job satisfaction survey. American Journal of Community Psychology, 13 693-713. Spector, P.E. (1997). Job satisfaction: Application, assessment, causes, and consequences Thousand Oaks, CA: Sage Publications, Inc. Spector, P.E. (2001). Job satisfaction survey [On-line]. Available: http://chuma.cas.usf.edu/~spector/scales/jsspag.html Spielberger, C.D., Johnson, E.H., Russell, S.F., Crane, R.J., Jackobs, G.A. & Worden, T.J. (1985). The experience and expression of anger: Construction and validation of an anger expression scale. In M.A. Chesney & R. H. Rosenman (Eds.), Anger and Hostility in Cardiovascular and Behavioral Disorders (pp. 5-30). Washington, DC: Hemisphere Publishing Corp. Steel, R.P. & Ovalle, N.K. (1984). A review and meta-analysis of research on the relationship between behavioral intentions and employee turnover. Journal of Applied Psychology, 69 673-686. Sutton, R.I. & Rafaeli, A. (1988). Untangling the relationship between displayed emotions and organizational sales. Academy of Management Journal 31, 461-487. Tett, R.P. & Meyer, J.P. (1993). Job satisfaction, organizational commitment, turnover intention, and turnover: Path analyses based on meta-analytic findings. Personnel Psychology 46, 259-290.

PAGE 101

93 Tews, M.J., & Glomb, T.M. (2003). Emotional labor and the five-factor model of personality. Paper presented at the 18th annual meeting of the Society for Industrial and Organizational Psychology, Orlando, Florida. Underwood, M.K., Hurley, J.C., Johanson, C.A. & Mosley, J.E. (1999). An experimental, observational investigation of children’s responses to peer provocation: Developmental and gender differences in middle childhood. Child Development, 70, (6), 1428-1446. Vey, M.A. & Bono, J.E. (2003). Emotional labor: Is it labor for everyone? Paper presented at the 18th annual meeting of the Society for Industrial and Organizational Psychology, Orland, Florida. Weisberg, J., & Sagie, A. (1999). Teachers’ physical, mental and emotional burnout: Impact on intention to quit. Journal of Psychology Interdisciplinary and Applied, 133 (3), 333-340. Weiss, D.J., Dawis, R.V., England, G.W. & Lofquist, L.H. (1967). Manual for the Minnesota Satisfaction Questionnaire (Minnesota Studies in Vocational Rehabilitation, No. 22). University of Minnesota, Minneapolis. Wharton, A.S. (1993). The affective consequences of service work. Work and Occupations, 20 (2), 205-232.

PAGE 102

94 Appendices

PAGE 103

95 Appendix A Part 1 Below are a set of statements which may or may not describe how you make decisions when you have to choose between two courses of action or alternatives when there is no clear right way or wrong way to act. Read each statement and circle the number on the line which corresponds to the choices presented below. Strongly Disagree Disagree Neither Agree nor Disagree Agree Strongly Agree 1. I am usually pretty effective in dealing with emergencies. 1 2 3 4 5 2. I tend to lose control during emergencies. 1 2 3 4 5 3. When I see someone who badly needs help in an emergency, I go to pieces. 1 2 3 4 5 Part 2 Below are several different actions in which people sometimes engage. Read each of them and decide how frequently you have carried it out in the past. Circle the number which best describes your past behavior. Use the scale presented below. Never Once More than Once Often Very Often 1. I have helped carry a stranger's belongings (e.g., books, parcels, etc.). 1 2 3 4 5 2. I have allowed someone to go ahead of me in a line (e.g., supermarket, copying machine, etc.) 1 2 3 4 5 3. I have let a neighbor whom I didn't know too well borrow an item of some value (e.g., tools, a dish, etc.). 1 2 3 4 5 4. I have, before being asked, voluntarily looked after a neighbor's pets or children without being paid for it. 1 2 3 4 5 5. I have offered to help a handicapped or elderly stranger across a street. 1 2 3 4 5

PAGE 104

96 Appendix B Below are phrases describing people's behaviors. Please use the rating scale below to describe how accurately each statement describes you Describe yourself as you generally are now, not as you wish to be in the future. Describe yourself as you honestly see yourself, in relation to other people you know of the same sex as you are, and roughly your same age. So that you can describe yourself in an honest manner, your responses will be kept in absolute confidence. Please read each statement carefully, and then circle the number that most closely represents how well you feel the statement describes you. Very Inaccurate Moderately Inaccurate Neither Accurate nor Inaccurate Moderately Accurate Very Accurate 1. Am the life of the party. 1 2 3 4 5 2. Feel little concern for others. 1 2 3 4 5 3. Am always prepared. 1 2 3 4 5 4. Get angry easily. 1 2 3 4 5 5. Don’t talk a lot. 1 2 3 4 5 6. Am interested in people. 1 2 3 4 5 7. Leave my belongings around. 1 2 3 4 5 8. Get irritated easily. 1 2 3 4 5 9. Feel comfortable around people. 1 2 3 4 5 10. Insult people. 1 2 3 4 5 11. Pay attention to details. 1 2 3 4 5 12 Rarely get irritated. 1 2 3 4 5 13. Keep in the background. 1 2 3 4 5 14. Sympathize with others’ feelings. 1 2 3 4 5 15. Make a mess of things. 1 2 3 4 5 16. Seldom get mad. 1 2 3 4 5 17. Start conversations. 1 2 3 4 5 18. Am not interested in other people’s problems. 1 2 3 4 5 19. Get tasks done right away. 1 2 3 4 5 20. Get upset easily. 1 2 3 4 5 21. Have little to say. 1 2 3 4 5 22. Have a soft heart. 1 2 3 4 5 23. Often forget to put things back in their proper place. 1 2 3 4 5 24. Am not easily annoyed. 1 2 3 4 5 25. Talk to a lot of different people at parties. 1 2 3 4 5 26. Am not really interested in others. 1 2 3 4 5 27. Like order. 1 2 3 4 5 28. Am often in a bad mood. 1 2 3 4 5

PAGE 105

97 29. Don’t like to draw attention to myself. 1 2 3 4 5 30. Take time out for others. 1 2 3 4 5 31. Avoid my duties. 1 2 3 4 5 32. Keep my cool. 1 2 3 4 5 33. Don’t mind being the center of attention. 1 2 3 4 5 34. Feel others’ emotions. 1 2 3 4 5 35. Follow a schedule. 1 2 3 4 5 36. Lose my temper. 1 2 3 4 5 37. Am quiet around strangers. 1 2 3 4 5 38. Make people feel at ease. 1 2 3 4 5 39. Am exacting in my work. 1 2 3 4 5 40. Rarely complain. 1 2 3 4 5

PAGE 106

98 Appendix C In the following sections, we would like to know about the emotions you express to others, such as customers, clients, coworkers, and supervisors, and emotions that you feel but do not express while on the job. That is, we are interested in what you express through your body language, facial expressions, tone of voice, etc. Consider your experiences at work over the past six months. The following sections may seem somewhat similar, so please read the instructions carefully. Section 1 – Expressing Emotions You Do Not Feel In this section, we would like to know how often you express emotions on the job when you really do not feel these emotions. For example, how often do you express feelings of irritation or concern when you really do not feel that way? How often do you express feelings of ___________ on the job when you really do not feel that way? I express this many times a day when I do not feel it I express this a few times a day when I do not feel it I express this a few times a week when I do not feel it I express this a few times a month when I do not feel it I never express this when I do not feel it 1 Irritation 5 4 3 2 1 2 Anxiety 5 4 3 2 1 3 Contentment 5 4 3 2 1 4 Sadness 5 4 3 2 1 5 Concern 5 4 3 2 1 6 Disliking 5 4 3 2 1 7 Aggravation 5 4 3 2 1 8 Fear 5 4 3 2 1 9 Happiness 5 4 3 2 1 10 Distress 5 4 3 2 1 11 Liking 5 4 3 2 1 12 Hate 5 4 3 2 1 13 Anger 5 4 3 2 1 14 Enthusiasm 5 4 3 2 1

PAGE 107

99 Section 2 – Keeping Emotions To Yourself In this section, we would like to know about emotions you do not express on the job but feel like expressing. That is, we are interested in how often you keep certain emotions to yourself because you feel you should not express them on the job. For example, how often do you keep feelings of happiness or liking to yourself when you really feel that way? How often do you keep feelings of ___________ to yourself when you really feel that way? I keep this to myself many times a day I keep this to myself a few times a day I keep this to myself a few times a week I keep this to myself a few times a month I never keep this to myself I never feel this 1 Irritation 5 4 3 2 1 0 2 Anxiety 5 4 3 2 1 0 3 Contentment 5 4 3 2 1 0 4 Sadness 5 4 3 2 1 0 5 Concern 5 4 3 2 1 0 6 Disliking 5 4 3 2 1 0 7 Aggravation 5 4 3 2 1 0 8 Fear 5 4 3 2 1 0 9 Happiness 5 4 3 2 1 0 10 Distress 5 4 3 2 1 0 11 Liking 5 4 3 2 1 0 12 Hate 5 4 3 2 1 0 13 Anger 5 4 3 2 1 0 14 Enthusiasm 5 4 3 2 1 0

PAGE 108

100 Appendix D Means, Standard Deviations and Correlations between all Variables and Emotional Labor, Faking Emotions and Suppressing Emotions Variable N Mean SD Extra Agree Consc Anger Help EL FE SE JS ITQ Perf Extra 183 3.52 .69 1.00 Agree 183 3.93 .62.19** 1.00 Consc 183 3.85 .63-.03 .34** 1.00 Anger 183 2.47 .75.13 -.33**-.31** 1.00 Help 183 7.13 1.07 .21* .39**.30**-.31**1.00 EL 183 2.11 .67 -.01 -.17* -.25**.29**-.07 1.00 FE 183 2.08 .78 .08 -.16* -.26** .24**-.08 .83**1.00 SE 183 2.15 .81-.10 -.14 -.16* .25**-.04 .85**.41** 1.00 JS 183 4.45 1.22-.09 .03 .26** -.13 .10 -.18* -.13 -.17* 1.00 ITQ 183 2.55 1.06.04 -.04 -.16* .12 .04 .14* .14 .10 -.63** 1.00 Perf 183 129.86 57.80.00 -.08 .01 .10 .03 .01 -.03 .05 .18**-.20**1.00 EXTRA = Extraversion, Agree = Agreeableness, Consc = Conscientiousness, Help = Helpfulness, EL = Emotional Labor, FE = Faking Emotions, SE = Suppressing Emotions, JS = Job Satisfaction, ITQ = Intention to Quit, Perf = Performance *p<.05 ** p<.01

PAGE 109

101 Appendix E Below are a set of statements which may or may not describe how you feel about your job. Read each statement and circle the number on the line which comes closest to your opinion of your job. Disagree Very Much Disagree Moderately Disagree Slightly Agree Slightly Agree Moderately Agree Very Much 1. I sometimes feel my job is meaningless. 1 2 3 4 5 6 2. I like doing the things I do at work. 1 2 3 4 5 6 3. I feel a sense of pride in doing my job. 1 2 3 4 5 6 4. My job is enjoyable. 1 2 3 4 5 6

PAGE 110

102 Appendix F Below are a set of statements which may or may not describe how you feel about your job. Read each statement and circle the number on the line which corresponds to the choices presented below. Strongly Disagree Disagree Neither Agree nor Disagree Agree Strongly Agree 1. I have considered leaving my current job. 1 2 3 4 5 2. I think that if I were choosing my place of work again, I would choose the same organization that I work for now. 1 2 3 4 5 3. I think in the near future I will leave working at this organization. 1 2 3 4 5

PAGE 111

103 Appendix G Space below reserved for IRB Stamp – Please leave blank Informed Consent Social and Behavioral Sciences University of South Florida Information for People Who Take Part in Research Studies The following information is being presented to help you decide whether or not you want to take part in a minimal risk research study. Please read this carefully. If you do not understand anything, ask the person in charge of the study. Title of Study: Antecedents and Consequences of Emotional Labor: Understanding the Relationships Among Personality, Emotional Labor, Job Satisfaction, Intention to Quit and Job Performance Principal Investigator: Laurie K. Diamond, M.A. Study Location(s): Debt Collection Organization You are being asked to participate because your job is a debt collector for the host organization. General Information about the Research Study The purpose of this research study is to which personality factors may be beneficial for someone to be a successful debt collector. Plan of Study To participate in this study, you will need to complete several different survey measures. It should take you approximately 15 to 30 minutes to complete the surveys. The surveys will be coded so that your individual responses will remain anonymous. The survey data will be matched to your performance data to help determine what makes a debt collector perform well and stay with the organization. Payment for Participation You will not be paid for your participation in this study Benefits of Being a Part of this Research Study By taking part in this research study, you may increase our overall knowledge of what makes people be good debt collectors.

PAGE 112

104 Risks of Being a Part of this Research Study There are no known risks associated with this study. Confidentiality of Your Records Your privacy and research records will be kept confidential to the extent of the law. Authorized research personnel, employees of the Department of Health and Human Services, and the USF Institutional Review Board may inspect the records from this research project. The results of this study may be published. However, the data obtained from you will be combined with data from others in the publication. The published results will not include your name or any other information that would personally identify you in any way. In order to ensure your anonymity, identifying information will not be placed on any survey data. Prior to the survey, you will be given a code number to place on the materials. The key to the codes will be safely kept off premises, in Florida, and will be destroyed once all the data has been put into a computer program. Volunteering to Be Part of this Research Study Your decision to participate in this research study is completely voluntary. You are free to participate in this research study or to withdraw at any time. There will be no penalty or loss of benefits you are entitled to receive, if you stop taking part in the study. The participation or non-participation with this study will in no way jeopardize your job. Questions and Contacts € If you have any questions about this research study, contact Laurie Diamond at 813/340-5100. € If you have questions about your rights as a person who is taking part in a research study, you may contact the Division of Research Compliance of the University of South Florida at (813) 974-5638. Consent to Take Part in This Research Study By signing this form I agree that: € I have fully read or have had read and explained to me this informed consent form describing this research project. € I have had the opportunity to question one of the persons in charge of this research and have received satisfactory answers. € I understand that I am being asked to participate in research. I understand the risks and benefits, and I freely give my consent to participate in the research project outlined in this form, under the conditions indicated in it.

PAGE 113

105 € I have been given a signed copy of this informed consent form, which is mine to keep. _________________________ __________________________ __________ Signature of Participant Printed Name of Participant Date

PAGE 114

106 Appendix H Demographic Information Code #:_______________________________ Age: ________years Gender: ______Male ______ Female Race: ____Asian ____Black ____Caucasian ____Hispanic ____Native American/ American Indian ____Other How long have you been employed as a debt collector? ______years _____months How long have you been employed by this organization? ______ years ______ months

PAGE 115

107 About the Author Laurie K. Diamond received her bachelor’s degree in Psychology in 1999 and her M.A. in Industrial/Organizational Psychology in 2002 from the University of South Florida. While completing her undergraduate work, Mrs. Diamond was kept busy raising four children and rotated her school schedule around her children’s bus schedule and was able to graduate Magna Cum Laude. While in the Ph.D. program at the University of South Florida, Mrs. Diamond opened her own consulting company, Integra Performance Consulting and worked hand in hand with her husband, Mitch Diamond, to grown their on-line training software company, eLogic Learning. Due to the travel requirements involved with being a consultant, Mrs. Diamond has had the opportunity to extensively travel the United States and is now focusing her attention on international clients. Mrs. Diamond lives in Tampa, Florida with her husband, daughter and two dogs.