USF Libraries
USF Digital Collections

Delivering continuing education in health education using self-directed computer-mediated instruction

MISSING IMAGE

Material Information

Title:
Delivering continuing education in health education using self-directed computer-mediated instruction moving from intention to action
Physical Description:
Book
Language:
English
Creator:
Ellery, Jane
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
prefessional preparation
computer-based training
professional development
health promotion
distance learning
Dissertations, Academic -- Public Health -- Doctoral -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Using advanced technologies can help increase the availability of educational offerings; however, the steps taken in this direction must be appropriate for the target population and the specific content taught. As such, understanding factors that lead to health educators' intentions and behavior related to computer-mediated instruction for continuing education is an important step in developing and marketing appropriate computer-mediated instruction programs (Hoffman & Novak, 1994). Using the theory of planned behavior (Ajzen, 1988) this study explored the relationships between health educators' perceived behavioral control, attitudes, and subjective norms related to computer-mediated continuing education programs and their intentions to use, and previous experience with, computer-mediated education. Employing a cross sectional survey design, data were collected from 504 members of the Society for Public Health Education (SOPHE) (40% response rate) using an online survey instrument. Logistic regression was used to investigate the associations between attitudes, subjective norm, perceived behavioral control, and intention related to using computer-mediated continuing education programs and a proxy measure representing their computer-mediated continuing education behavior. Perceived behavioral control and attitudes were found to have significant associations with computer-mediated continuing education behavior, with intention partially mediating the association with perceived behavioral control and fully mediating the association with attitudes. When studying a subset of the group composed of respondents with a positive intention toward computer-mediated continuing education programs, respondent characteristics and barriers identified as distinguishing between individuals with positive and negative behaviors included perceived behavioral control, presence of a license or certification, a lack of programs, a lack of relevant topics for programs, and a lack of technical support for programs. These results suggest that for health education and health promotion professionals to engage in computer-mediated continuing education programs, more programs, especially ones that address topics relevant to their current functioning, need to be created and made readily available. Also, ensuring that appropriate technical support is available to assist participants, and informing potential participants of the availability of this technical assistance, may encourage more health educators and health promotion professionals to follow through on their intentions to participate in computer-mediated programs.
Thesis:
Thesis (Ph.D.)--University of South Florida, 2003.
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 Jane Ellery.
General Note:
Includes vita.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 236 pages.

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 - 001416914
oclc - 52831445
notis - AJJ4766
usfldc doi - E14-SFE0000052
usfldc handle - e14.52
System ID:
SFS0024748:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Delivering Continuing Education in Health Education using Self-Directed Computer-Mediated Instruction: Moving from Intention to Action by Jane Ellery A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Community and Family Health College of Public Health University of South Florida Major Professor: Robert J. McDermott, Ph.D. Kelli R. McCormack Brown, Ph.D. Wayne Westhoff, Ph.D. Ann Barron, Ph.D. Candi Ashley, Ph.D. Date of Approval: July 11, 2003 Keywords: Professional Development, Computer-Based Training, Distance Learning, Health Promotion, Professional Preparation Copyright 2003, Jane Ellery

PAGE 2

i Table of Contents List of Tables ................................................................................................................. ....iii List of Figures .....................................................................................................................v Abstract ....................................................................................................................... .......vi Chapter One: Introduction ..................................................................................................1 Statement of the Problem ........................................................................................3 Purpose of the Study ...............................................................................................5 Research Questions .................................................................................................6 Assumptions ............................................................................................................7 Delimitations ...........................................................................................................8 Limitations ..............................................................................................................9 Definitions ...............................................................................................................9 Importance of the Study ........................................................................................12 Chapter Two: Review of Related Literature .....................................................................14 Introduction to Distance Learning ........................................................................15 Computers as a Distan ce Learning Tool ...............................................................16 Effectiveness of Compute r-Mediated Instruction .................................................18 Guidelines for Distance Deliver y of Educational Programs .................................21 The Theory of Planned Behavior ..........................................................................24 Predicting Intention and Behavior Re lated to Computer Technologies ...............28 Continuing Education for Health Educators .........................................................29 Computer-mediated Instruction for Conti nuing Education in Health Education .32 Online Data Collection .........................................................................................33 Chapter Three: Methods ...................................................................................................39 Research Questions ...............................................................................................39 Population .............................................................................................................40 Research Design, Sample, a nd Study Administration ..........................................41 Survey Development and Pilot Testing ................................................................43 Data Collection .....................................................................................................46 Operationalization of Variables ............................................................................47 Data Analysis ........................................................................................................49 Univariate and bivariate analyses .............................................................49 Identifying subscale items .........................................................................50 Multivariate analyses ................................................................................52

PAGE 3

ii Potential Problems, Threats, and Ethical Considerations .....................................56 Summary ...............................................................................................................58 Chapter Four: Results .......................................................................................................59 Results from the In-Depth Interviews ...................................................................59 Review Panel Findings .........................................................................................61 Field Testing .........................................................................................................64 Quantitative Survey Results ..................................................................................74 Descriptive Statistics .................................................................................78 Factor Analysis .........................................................................................95 Composite Score Development and Re searcher Defined Variables .......100 Between Groups Comparisons ................................................................104 Multivariate Models ................................................................................117 Chapter Five: Discussion ...............................................................................................127 Summary .............................................................................................................128 Practical Implications ..........................................................................................134 Study Considerations ..........................................................................................137 Limitations ..............................................................................................137 Strengths .................................................................................................140 Lessons Learned Related to Online Data Collection ..........................................141 Future Directions ................................................................................................143 References .......................................................................................................................147 Appendices ......................................................................................................................155 Appendix A. A Structural Model of the Theory of Planned Behavior. .............156 Appendix B. A Structural Model of the Theory of Reasoned Action. ..............157 Appendix C. Listing of Hea lth Education Competencies ..................................158 Appendix D. Letters of Approval from the Institutional Review Boards. .........164 Appendix E. Pre-noti ce Electronic Message .....................................................169 Appendix F. Electronic Message Requesting Participation ...............................170 Appendix G. Follow-up Notification 1 ..............................................................171 Appendix H. Follow-up Notification 2 ..............................................................173 Appendix I. Review Panel Members .................................................................174 Appendix J. Operationalization of Independent Variables ................................175 Appendix K. Study Questionnaire .....................................................................180 Appendix M. Compiled Field Notes from In-Depth Interviews ........................192 Appendix N. Review Panel Packet for Dissertation ..........................................197 Appendix O. Results from Review Panel ..........................................................214 Appendix P. Scree Plot for Eige nvalues from Factor Analysis .........................222 Appendix Q. Table of Cross Tabulations and Chi-Square Tests for Behavior ..223 About the Author ..................................................................................................End Page

PAGE 4

iii List of Tables Table 1. Test-Retest Reliability Per-Item Correlations: Perception Variables (Spearman Rank Correlation Coefficients) ........................................66 Table 2. Test-Retest Reliability Per-Item Correlations: Barriers (Spearman Rank Correlation Coefficients) .........................................................69 Table 3. Test-Retest Reliability Per-Item Correlations: Other Items (Spearman Rank Correlation Coefficients) .........................................................71 Table 4. Comparison of Special Inte rest Group and Caucus Membership among Response Groups .....................................................................................75 Table 5. Professional Characteristics of Respondents. .....................................................80 Table 6. Computer and Internet Use .................................................................................85 Table 7. Means and Standard Deviat ions of Respondents Perceptions Related to Attitude, Subjective Norm, and Perceived Behavioral Control Toward Computer-Mediated Instruction. ..............................................88 Table 8. Means and Standard Deviat ions for Barriers Associated with Using Computer-Mediated Instruction for Continuing Education. ....................92 Table 9. Means and Standard Deviati ons of Respondents' Current Interest in Participating in Continuing Education Opportunities .....................................95 Table 10. Questionnaire Items and Co rresponding Factor Loadings from the Rotated Factor Pattern Matrix and Factor Structure Matrix (N=410) ...............................................................................................................97 Table 11. Frequency and Percent for Recoded Variable Representing Intention ............................................................................................................102 Table 12. Frequency and Percent for the Researcher-Developed Variable Representing Behavior ......................................................................................103 Table 13. Frequency and Percent for the Researcher-Developed Variable Representing Continuing Education Behavior ..................................................104

PAGE 5

iv Table 14. Summary of Significant Bi variate Associations: Respondent Characteristics by Independent Variables .........................................................110 Table 15. Respondent Characteristic s by Continuing Education Behavior ....................111 Table 16. The Effect of Attitude, S ubjective Norm, Perceived Behavioral Control, Intention and the Control Vari ables on Continuing Education Behavior (Model 1) ..........................................................................118 Table 17. The Effect of Attitude, S ubjective Norm, Perceived Behavioral Control, and the Control Variables on Continuing Education Behavior (Model 2) ...........................................................................................120 Table 18. The Effect of Attitude, S ubjective Norm, Perceived Behavioral Control, and the Control Variables on Continuing Education Behavior for Respondents with a Positive Intention (Model 3) .......................122 Table 19. Review Panel Members I nvited to Review Survey Tool ................................174 Table 20. Study Variables and the Statements and Format for their Operationalization .............................................................................................175 Table 21. Respondent Characteristics by Behavior ........................................................223

PAGE 6

v List of Figures Figure 1. The Relationship of the Theo ry of Planned Behavior Constructs in this Study ......................................................................................................132 Figure 2. A Structural Model of the Theory of Planned Behavior .................................156 Figure 3. A Structural Model of the Theory of Reasoned Action. .................................157 Figure 4. Scree Plot of Eigenvalues from Factor Analysis .............................................222

PAGE 7

vi Delivering Continuing Educati on in Health Education Usin g Self-Directed Computer Mediated Instruction: Moving from Intention to Action Jane Ellery ABSTRACT Using advanced technologies can help in crease the availability of educational offerings; however, the steps taken in this di rection must be appropriate for the target population and the specific content taught. As such, understanding factors that lead to health educators intentions and behavior related to computer-mediated instruction for continuing education is an important step in developing and marketing appropriate computer-mediated instruction programs (Hof fman & Novak, 1994). Using the theory of planned behavior (Ajzen, 1988) this study e xplored the relationships between health educators perceived behavioral control, attitudes, and subjective norms related to computer-mediated continuing education progr ams and their intentions to use, and previous experience with, computer-mediated education. Employing a cross sectional survey de sign, data were collected from 504 members of the Society for Public Health E ducation (SOPHE) (40% response rate) using an online survey instrument. Logistic regre ssion was used to investigate the associations

PAGE 8

vii between attitudes, subjective no rm, perceived behavioral contro l, and intention related to using computer-mediated continuing education programs and a proxy measure representing their computer-mediated c ontinuing education behavior. Perceived behavioral control and attitu des were found to have significant associations with computer-mediated continuing education behavi or, with intention partially mediating the association with perceived be havioral control and fully mediating the association with attitudes. When studying a subset of the group composed of respondents with a positive intention toward computer-mediated c ontinuing education programs, respondent characteristics and barriers id entified as distinguishing betw een individuals with positive and negative behaviors included perceived behavioral control, presence of a license or certification, a lack of programs, a lack of relevant topics for programs, and a lack of technical support for programs. These results suggest that for health edu cation and health promotion professionals to engage in computer-mediated conti nuing education programs, more programs, especially ones that address topics relevant to their current functioning, need to be created and made readily available. Also, ensuring th at appropriate technica l support is available to assist participants, and informing potential participants of the availability of this technical assistance, may encourage more health educators and health promotion professionals to follow through on their intentions to participate in computer-mediated programs.

PAGE 9

1 Chapter One: Introduction Traditionally, education has consisted of bringing students to sources of knowledge. However, an increased need for ongoing educational training and the emergence of computer and Internet technol ogies have spawned in creased interest in developing new educational opportunities. These new, less traditional educational opportunities are delivered outside a classroom setting and are centered on bringing the sources of knowledge to students (Barley, 1999). Alt hough traditional approaches remain the mainstay of Americas educationa l offerings, rapid adva nces in technology experienced in the past three decades have led to expanded opportuni ties for the delivery of educational programs. Many of these e xpanded opportunities help overcome some of the time and distance related barriers impos ed by classroom learning environments. Computer technology is one medium that can be considered when developing new educational opportunities. Literature relate d to the use of computers and the Internet in health education is limited, and health professionals have been slower to embrace Internet and computer technologies than business and financial fields (Eng, 2001). However, research in other disciplines sugge sts computers and Inte rnet-based distance education programs are able to promote stud ent-centered learning and reach individuals who may not be able to attend courses be ing offered in traditional formats (U.S. Department of Health and Human Services [US DHHS], 1997; Web-based Education

PAGE 10

2 Commission, 2000). Additionally, well-designed computer-mediated instructional programs may be effective educational tool s (Barron, 1998; Steckler, Farel, Bontempi, Umble, Polhamus, & Trestler, 2001; Umble, Cervero, & Yang, 2000; Wenger, Holloway, & Garton, 1999). Teaching high-quality courses to working health professionals using modern computer and web-based technologies is possible (Steckler, et al., 2001, p. 745). Numerous studies in management and info rmation sciences have used the theory of planned behavior to look at the relationships between attitudes, subjective norms, perceived behavioral control, behavioral intentions, and beha viors related to information technologies (Davis, 1989; Davis, Bagozzi, & Warshaw, 1989; Mathieson, 1991; Morris & Venkatesh, 2000; Taylor & Todd, 1995a; Taylor & Todd, 1995b). Additionally, Davis (1989) used the theory of planned behavior as the basis for development of a domainspecific model, the technology acceptance model. Research suggests that the theory of planned behavior provides a content-free theoretical framework (Ajzen, 2001) that predicts behavior as well as, and sometimes better than, models designed for specific content domains (Ajzen, 2001; Davis, Bagozzi, & Warshaw 1989; Mathieson, 1991; Taylor & Todd, 1995a). When comparing the theory of planne d behavior and the technology acceptance model, Mathieson (1991) suggests that the technology acceptance model provides a quick and inexpensive way to gather genera l information about individuals perceptions of a system (p. 187), recommending its use to identify general levels of satisfaction. However, the theory of planned behavior delivers more specific information, giving more insight into why an individual or group might be dissatisfied (p. 187).

PAGE 11

3 Additionally, Mathieson argues that the theory of planne d behavior provides more specific information that can better guid e development (p.1 73). Based on these findings, the theory of planned behavior should be useful in exploring health educators intention to use and actual usage of computer and Intern et technologies for continuing education and in identifying sa lient features related to bot h the users and the mode of delivery that should be considered as com puter-mediated educational opportunities are developed. In addition to providing expanded opport unities for the delivery of educational programs, advances in Internet and computer technologies provide researchers with new research tools. Although limitations exist in the delivery of online surveys, researchers using carefully designed tools and methods to collect information from populations that regularly use computers and the Internet can have good success with Internet-based survey data collection (Dillman, 2000). Statement of the Problem Today our Nation faces a widening gap be tween the challenges to improve the health of Americans and the capacity of th e public health workforce to meet those challenges (US DHHS, 1998, p. 1). In respons e to this concern, the Steering Committee of the Public Health Functions Project commissioned the Subcommittee on Public Health Workforce, Training, and Education in Sept ember 1994. Part of this subcommittees charge was to address issues related to the training and educational needs of public health practitioners. The resulting re port, a culmination of the effo rts of many well-established public health professionals, states:

PAGE 12

4 The public health workforce requires up-to -date knowledge and skills to deliver quality essential public health services. To meet the training and continuing education needs of an evolving workfor ce, a clearer understanding is required concerning the functions and composition of the public health workforce both now and in the future Furthermore, becau se this is a geographically dispersed and demographically diverse workforce, new strategies for presenting efficient and effective training must be developed. (US DHHS, p. 3) Researchers and expert pane ls have worked to identify current public health workforce training competencies and contin uing education competencies for health educators (Allegrante, Moon, Auld, & Ge bbie, 2001; Gebbie & Hwang, 1998; National Commission for Health Education Creden tialing, Inc., 1996, 1999; OCarroll & the Public Health Informatics Competency Working Group, 2002; US DHHS, 1997). These competencies provide valuable insight in to content areas to be addressed during continuing education and workforce training programs. However, continuing education programs have been shown to be most effectiv e when tailored to suit the specific needs of the professionals participating in the trai ning (Sweeissen & Tilgner, 2000). Continuing education training differs from pre-service trai ning in that participan ts typically have onthe-job experience and have sp ecific job-related tasks they are interested in improving. A panel may be able to address global issues re lated to health education job skill needs, but the wants and needs identified by currently employed health educators for workforce training are also important to consider.

PAGE 13

5 Using advanced technologies to provide instruction for public health educators has been identified by resear chers and expert panels as being important (US DHHS, 1997); however, the steps taken in this dir ection must be appropriate for the taget population and the specific c ontent taught. Unfortunatel y, although educational media have expanded in variety, information on h ealth educators preferences for types of education media and perception toward us ing computer-mediate d instruction for continuing education remains limited. Continui ng education should be individualized to the practitioners receiving the training (Sweri ssen & Tilgner, 2000), and the effectiveness of computer-mediated instruction may be discipline-specific (Dominguez & Ridley, 2001). As such, understanding factors that lead to health educators intentions and behavior related to computer-mediated in struction for continuing education is an important step in developing, implementi ng, and marketing appropriate computermediated instruction programs (Hoffman & Novak, 1994). Also important is identifying the attributes that contribute to moving health educators who have a positive intention to perform the behavior into action. Little evid ence-based research is available to draw from as health educators develop and use computer-mediated instruction for workforce training, and the process must be given great consideration to ensure movement in the right direction. As Ehrmann (1995) notes, I f youre headed in the wrong direction, technology wont help you get to the right place (p. 21). Purpose of the Study The purpose of this study is to exam ine the relationships between health educators perceived behavioral control, attitudes, and subjective norms related to

PAGE 14

6 computer-mediated continuing education progr ams and their intentions to use, and previous experience with, computer-mediate d education. Such information may be useful in helping to design and implement st rategies for the use of computers and the Internet in the delivery of continuing education programs for health educators. Research Questions The research questions inves tigated in this study include: 1. What is the association between health educators perceived behavioral control related to using computer-mediated c ontinuing education programs and their behavior related to computer-mediated education? 2. What is the association between health educators attitudes related to using computer-mediated continuing education pr ograms and their behavior related to computer-mediated education? 3. What is the association between health educators subjective norms related to using computer-mediated continuing e ducation programs and their behavior related to computer-mediated education? 4. Do health educators intentions to us e computer-mediated continuing education programs mediate the association between perceived behavi oral control, attitudes, and subjective norms and their behavior related to computer-mediated education? 5. For individuals with a positive intention to use computer-mediated instruction, what characteristics studied help different iate between those who have previously used this learning medium and those who have not?

PAGE 15

7 Assumptions Assumptions related to this study include: 1. Health educators who report both past experience with computer-mediated instruction and a future intention to participate in computer-mediated programs are considered to have a high level of perceived behavioral control for computer-mediated instruction. 2. Health educators adopt computer and In ternet technologies prior to adopting the use of computer-mediated inst ruction for continuing education. 3. Health educators invited to participat e in the survey have access to the Internet. 4. Health educators responding to the survey report their responses to the survey instrument accurately. 5. Health educators who are not able to pa rticipate in an Internet-based survey due to a lack of access to a computer or the Internet or a l ack of the skills necessary for survey participation are unlikely to access computer-mediated instruction for continuing education. 6. The health educators who submit survey responses are th e individuals who were invited to participate. 7. Electronic mail messages that are not re turned as undeliverable are delivered to the intended health educator and are opened and read by that individual. 8. A self-report questionnaire is an appropr iate tool for studyi ng the research question.

PAGE 16

8 9. An Internet-based survey format is appropriate for obtaining information related to the self-re port of behaviors. 10. Computer-mediated instruction is eff ective across topics of professional development and in the delivery of con tinuing education programs for health educators. 11. Computer-mediated instruction is desira ble for health educator professional development and the delivery of continuing education programs. 12. Computer-mediated instruction is feasib le for health educator professional development and the delivery of continuing education programs. Delimitations The following delimitations under res earcher control are imposed: 1. Subjects in this study are volunteers. 2. Participants are health educators who are listed in the 2002-2003 SOPHE membership directory and who have a valid electronic mailing address. 3. The study excludes health educators w ho are current members of SOPHE but are not able to participate in an Intern et-based survey due to a lack of access to a computer or the Internet or a lack of the skills necessary for survey participation. 4. The study excludes health educators w ho are current members of SOPHE but are involved in the development or ov ersight of this re search project. 5. Members of SOPHE who ar e retired, whether or not they have a valid electronic mailing address.

PAGE 17

9 Limitations Issues related to this study that may l ead to limitations in its findings include: 1. The nature of the study and the use of a convenience sample limit the generalizability of the findings of the study. 2. Health educators who volunteer to partic ipate in surveys may be different than those who do not participate in surveys. 3. Health educators who participate in In ternet-based surveys may be different than those who do not participate in Internet-based surveys. 4. Study findings may be biased toward hea lth educators with higher levels of technology experience since the data ar e being collected using an online survey tool. 5. Study findings are specific to compute r-mediated instruction as defined for this project and may not be applicab le to other types of distance and distributed learning opportunities. Definitions 1. Asynchronous Asynchronous refers to computer communication that is not restricted to a specific time. 2. Attitude An attitude is a disposition to respond favorably or unfavorably to an object, person, inst itution, or event (Ajzen, 1988, p. 4). 3. Computer platform Com puter platform refers to the type of computer being used (e.g., Macintosh, PC) and th e software packages and versions

PAGE 18

10 (e.g., Internet Explorer, version 3.5 ; Internet Explorer, version 6.0; Netscape, version 6.0) inst alled on that computer. 4. Computer-mediated instruction An educational program delivered with the assistance of a computer. This co uld be a program that is taught over the Internet, delivered on a CD-R OM or DVD, located on a computer (such as seen with a kiosk), or using multiple computer-based delivery formats. For this study, compute r-mediated refers to self-paced instructional programs. 5. Continuing education A course or program that has been approved for university credit, continuing educatio n credit, or some other form of certification and is offered for indivi duals currently employed in health promotion positions. 6. Distance education Distance education is an umbrella term referring to educational programs where the instru ctor/facilitator and the student are geographically separated. 7. Experience Experience refers to an individuals past involvement in an activity, and not level of expertise. A person who is denoted as being experienced is one who has previously participated in a computermediated educational program. 8. Health educators For this study, hea lth educators refer to individuals who were listed in the 2002-2003 SOPHE membership directory.

PAGE 19

11 9. Intention Intention is an indication of how hard people are willing to try, of how much of an effort they are planning to exert, in order to perform a behavior (Ajzen, 1991, p. 181). 10. Perceived behavioral control Perceived behavioral contro l refers to the perceived ease or difficulty of perfor ming the behavior and it is assumed to reflect past experience as we ll as anticipated impediments and obstacles (Ajzen, 1988, p. 132). 11. Professional development Professional development refers to programs that expand knowledge and skills rela ted to current employment or anticipated future employment. 12. Platform Platform refers to the hardware and software associated with the functioning of a computer (see computer platform). 13. Server A server is a computer syst em formatted to function as a conduit for accessing and exchanging information over the Internet. 14. Subjective norm Subjective norm is a social factor that refers to the perceived social pressure to perfor m or not perform the behavior (Ajzen, 1991, p. 188). 15. Self-directed A self-directed course or program is one that allows you to proceed at your own pace. This course or program may or may not have a general time line associated with its progression (this would include a class that required assign ments to be turned in or specified dates and a general time when all work must be completed).

PAGE 20

12 16. Synchronous Synchronous refers to computer communication that occurs simultaneously for all individuals involved. 17. Technology Technology refers to a pr ocess that reduces the uncertainty of achieving a desired outcome. 18. Upload Upload refers to moving files to a server to allow them to be accessed over the Internet. Importance of the Study Reports have alluded to the changing cont ext in which health educators practice to make important contributions to improving public health (Allegrante, et al., 2001; US DHHS, 1997). Professional development a nd continuing education are important contributors to health educator s successfully functioning in this changing environment. Since the health education workforce is ge ographically dispersed and demographically diverse, emerging technologies may be helpful in expanding the current continuing education offerings (US DHHS, 1997). Self-paced, computer-mediated educational programs may be helpful in removing some of the time and distance barriers health educators encounter when looking for pr ofessional development opportunities. Health professionals have been slower to embrace Internet and computer technologies than business and financial fi elds (Eng, 2001). However, a majority of health educators have been shown to use the Internet and com puters (Brown, Ellery, Perlmutter, in press; Ellery, Brown, & Perlmutter, 2001; Hanks, Barnes, Merrill, & Neiger, 2000). This high usage level suggests that health educat ors may be ready to adopt computer-mediated educational programs as a form of professional development.

PAGE 21

13 As such, understanding how health educators want to engage in their professional development pursuits could improve the delive ry of continuing education opportunities to health educators. Information from this study can contri bute to the development, implementation, and marketing of computer-mediated continuing education in health education. The Public Health Workforce report encourages the use of distance learning for workforce development because of the potential to acce lerate and expand educational opportunities (US DHHS, 1997). The theoretical advantages of Internet and other computer-mediated instruction programs for continuing educati on are numerous. However, the current availability of these types of programs in health educa tion is limited. Ajzens (1988, 1991) theory of planned behavior and his recommendations for using the theory to develop behavioral interventions provide an intuitive framework for garnering a better understanding of the attitudes, social influen ces, and perceived barriers related to health educators use of computer-mediated educati on. Combining the findings from this study with published recommendations related to th e design of computer and Internet-based instruction programs and previously identified health education competencies will assist educators and software programmers in develo ping and delivering effective, meaningful computer-mediated educational programs. In addition, information from this study will be helpful for organizations and individuals exploring continuing education and distancebased workforce training opportunities.

PAGE 22

14 Chapter Two: Review of Related Literature In this section, an introduc tion to distance learning is provided, an overview of the theoretical framework and constructs associated with the theory of planned behavior is offered, and information providing insight into the training needs of health educators is summarized. This information can lead to an improved understand ing of the issues involved in using computer-mediated inst ruction to provide continuing education programs for health educators. In addition to studies about the use of computer-mediated instruction, litera ture focusing on the effectiveness of Internet, and other computermediated educational opportunities, con tinuing education, and the use of distance learning for continuing educati on are reviewed. Finally, info rmation related to collecting survey information using an online data collection instrument is synthesized. A search of multiple, interdisciplinary, databases available through the virtual library at the University of South Florida (Tam pa, Florida) was used to identify articles. This approach was utilized because of th e limited amount of printed health education literature related to computer-mediated in struction, the multidisciplinary nature of computer-mediated instruction for educationa l purposes, and the use of the theory of planned behavior to study similar, technol ogy-based questions in other disciplines. Organization of this review is categorized into sections including: (a) introduction to distance learning, (b) computers as a distance learni ng tool, (c) effectiveness of

PAGE 23

15 computer-mediated instruction, (d) guidelin es for distance delivery of educational programs, (e) the theory of planned behavi or, (f) predicting inte ntions and behaviors related to computer technologies, (g) continuing educat ion for health educators, (h) computer-mediated instruction for continuing education of health educators; and, (i) online data collection. Introduction to Distance Learning Individuals involved in dist ance education have a wide range of options that Willis (1995) effectively classifies into f our major groups, including print materials, audio tools, instructional vide o tools, and computer applica tions. Print materials are the original distance-learning medium and pr ovide the foundation from which all other educational delivery systems have evolved. Audio tools include the telephone, audio conferencing, short-wave radio, tapes and ra dio. The third group, often referred to as instructional video tools, encompasses slides and other still images, films, videotapes, live movie images and video conferencing. The final group, computer applications, incorporates the electronic pres entation of information within computer applications and includes computer-ass isted instruction, computer-managed instruction, computermediated instruction, electronic mail, fax, computer conferencing, and World-Wide Web applications. This last group, computer applic ations, and its use for distance delivery of continuing education programs to health education professionals was the focus of this study.

PAGE 24

16 Computers as a Distance Learning Tool The rapid expansion of computer tec hnology and the Internet have opened many instructional opportunities for distance educa tion. Some possibilities include using electronic mail for one-to-one correspondence between instructors and students or between students; online coll aboration, including both Inte rnet chat and computer conferencing; accessing web-based resources; computer-mediated educational software packages; and web-based edu cational programs (Blonna & Sh apiro, 2001; Florida Center for Instructional Technology [FCIT], 1999; Willis, 1995). Highlighting potential adva ntages and disadvantages of computer-mediated instruction provides an important backdrop for the discussion of its use as a continuing education tool for health educators. Th is list includes the ad vantages of offering computer-mediated instruction now as compar ed to the past, and the advantages of computer-mediated instruction as compared to traditional educational offerings. Potential advantages identified in the literature include: 1. Multimedia instructional capabilities; 2. Ability to facilitate self-paced learning; 3. High levels of interactivity; 4. Ability to provide written records of discussions and instructions; 5. Constantly emerging innovations in computer technology; 6. Improving access to this technology; 7. Delivery of programs to differe nt groups at varied times;

PAGE 25

17 8. Involvement of participants in activ ities that increase interaction while learning; 9. Reduction of need to schedule group cl asses for individuals with tight schedules; 10. Ability to reach individuals in diverse, widespread, rural, and hard-toreach locations; 11. Availability of synchronous and as ynchronous options for learning; 12. Assurance that the delivered message is consistent; and 13. Opportunity for multiple demonstrat ion and practice opportunities for reinforcing comprehension of concepts (Coleman, Sims, & Threfall, 1998; FCIT, 1999; Willis, 1995). Some potential limitations include: 1. The cost of network development; 2. Rapidly changing computer technology; 3. Varied computer operating systems and software applications; 4. Widespread computer illiteracy; 5. Computer viruses that can be tr ansferred with assignments and discussions; 6. The need for substantial planning on the part of the instructor; 7. The high level of motivation and computer proficiency required by individuals (both instructors and students) participating; 8. A heavy reliance on written communication;

PAGE 26

18 9. Availability of instructors to answer questions; 10. The difficulty of customization of pre-packaged programs for specific educational needs; 11. Decreased possibility for actual, hands-on training; 12. Decreased interpersonal communicati on and social inte raction; and 13. A lack of non-verbal communication (Coleman, et al., 1998; FCIT, 1999; Smith, Smith, & Boone, 2000; Willis, 1995). Effectiveness of Comput er-Mediated Instruction Although academicians and educational agencies have devoted energy and resources to evaluating the effectivene ss of computer-mediated instruction and developing guidelines for dist ance learning environments, mu ch remains to be learned about these areas. At the request of the Am erican Federation of Teachers (AFT) and the National Education Association (NEA), Phipps and Merisotis (1999) presented a review of research on the effectivene ss of distance learning in highe r education in April of 1999. For this review, they examined past issues of The Journal of Research on Computing in Education The American Journal of Distance E ducation, The Journal of Distance Education, The Journal of Computer-media ted Instruction, Research in Distance Education publications produced by the Univers ity of Maryland Institute for Distance Education, publications produced by the Am erican Center for the Study of Distance Education, and Ed: The Official Publication of the U.S. Distance Learning Association Although they were interested primarily in studies about th e effectiveness of distance

PAGE 27

19 education, they discovered that course and program design for distance learning commanded the most attention in the liter ature (Phipps & Merisotis, 1999, p. 35). Most of the available studies suggest a favorable comparison between distancebased classes and traditional classroom settings (Russell, cited in Young, 2000). However, Phipps and Merisotis (1999) questi on the quality of the re search studies. It should be emphasized that the re view provided evidence of the f act that there is a relative paucity of true, original research dedicated to explaining or predicting phenomena related to distance learning (Phi pps & Merisotis, 1999, p. 15). Additionally, the underlying assumption when comparing traditional instruct ion and distance delivery is that each of the mediums is constant acro ss all content areas and all st udents. This assumption is flawed because student attributes and instructio nal attributes are igno red (Lockee, cited in Carnevale, 2001). The three broad measures of effectiveness most often examined are student outcomes (such as grades), student attitudes about learning from a distance, and student satisfaction with th e distance learning experience (Phipps & Merisotis, 1999). Key shortcomings of existing research in clude lack of control for extraneous variables, lack of a random selection process for subj ects, questionable validation and reliability of test instruments measuring stude nt outcomes and attitudes, and inadequate control for attitudes a nd feelings of students and faculty (Phillips & Merisotis, 1999). All learners cannot be expected to have simila r levels of computer self-efficacy and academic self-efficacy, and one of these types of effi cacy may have a greater predictive usefulness in computer-mediated instru ction (Joo, Bong, & Choi, 2000). Generalization of the findings from these distance learning program s also may be limited because students who

PAGE 28

20 choose distance learning formats may be more motivated to learn than students in traditional classes (Sonner, 1999). Finally, activities used to measure student outcomes are similar to some instructional practices us ed in distance education programs. As such, studies suggesting improved results from distance-based programs actually may be measuring an improved ability to perform an activity as opposed to improved knowledge and skill (Sonner). Well-designed computer-mediated educati on programs do not appear to have a negative effect on student achievement (Wenger, et al., 1999). However, the discipline in which the programs are being delivered ma y make a difference when studying the effectiveness of computer-mediated inst ruction (Dominguez & Ridley, 2001). Few studies have examined the effectiveness of continuing education programs for public health professionals, and even fewer have looked at the effectiveness of computermediated instruction and other distance learni ng mediums for the delivery of continuing education programs (Umble, et al., 2000). Literature related to computer-mediated instruction for health educators is limited. Findings related to delivering continuing education to health professionals suggest that teaching high-qua lity, distance-based courses to working health professionals is possible (Steckler, et al., 2001). Moreover, designers of continuing education programs should consider that hea lth professionals are interested in practical co urses of short duration, deliv ered by experienced peers, reflecting the type of activitie s in which they are involved (Swerissen & Tilgner, 2000). Although not specifically referring to dist ance-based programs, Swerissen and Tilgner suggest that the most eff ective programs are tailored to ward the specific needs of

PAGE 29

21 different professionals involve d in health promotion. Whereas Internet-based, and other computer-mediated distance education classe s have the opportunity to promote studentcentered learning and reach individuals who may not be able to attend courses being offered in a traditional format (US DHH S, 1998; Web-based Education Commission, 2000), additional insight into th e effectiveness of computer-m ediated learning for health educators and the continuing education needs of these professionals is warranted. Guidelines for Distance Deli very of Educational Programs Two reports released in 2000 that provid e guidelines and benchmarks to use for distance delivery of education programs include the Institute for Higher Education Policy (IHEP) report, Quality On the Line: Benchmarks for Success in Internet-Based Education (Phipps & Merisotis, 2000), and the Ameri can Federation of Teachers (AFT) report, Going the Distance: AFT guidelines fo r good practice in distance education (2000). Research on design, operation, and administration of distance education programs is plentiful (Phipps & Merisotis, 2000). The IH EP and AFT reports are the synthesis of countless professionals in the field of dist ance education reviewing previous relevant findings. A consolidation of the recommendations from these two reports provides useful guidelines for Internet-based, and other co mputer-mediated distance learning programs. Although these reports are de signed predominately for higher education course work delivery, many of the recommendations can be extrapolated to other areas, such as the delivery of continuing education to h ealth educators through distance education. Guidelines for distance education pertinent to continuing education for health educators based on the IHEP and AFT reports include:

PAGE 30

22 1. Learning outcomes and required standard s, not technology av ailability, should form the basis for course development, design, and delivery; 2. Individuals presiding over distance-based classes should have control of the content of the course and the materials that will be used during instruction; 3. Individuals planning to pr ovide distance-learning pr ograms should be properly prepared to meet the requiremen ts of teaching in this format; 4. Courses should be designed to maximi ze the potential of the distance-learning environment; 5. Students entering into distance education training should be fully aware of the specific requirements for the class they are entering (such as computer skills needed, software and hardware require d, participation in discussions, and the ability to submit assignments electronically); 6. Students should be encouraged to anal yze, synthesize and evaluate course information as part of their program; 7. Instructors and students should main tain close personal interaction; 8. Class size should encourage a high degree of interactivity; 9. Instructional materials and methods should be reviewed periodically to ensure they continue to meet the learning needs and program standards; 10. Constructive feedback on assignments and inquiries should be provided to students in a timely manner; 11. Students should have access to suffici ent library resources and external information to allow them to be successful in their program;

PAGE 31

23 12. Expectations about assignment comp letion should be agreed upon by both the instructor and the students; 13. Students should have access to techni cal assistance throughout their course; 14. The programs educational effectivene ss and teaching/learning process should be assessed through multiple channels; 15. Intended learning outcomes should be reviewed regularly; and 16. The material covered in a distance-base d class should be consistent with the information that would be presented in a similar, classroom-based course. The guidelines that have been summarized to this point are related to distance learning in general and are important to note when developing self-paced Internet-based educational opportunities. A dditional consideration should be given to recommendations specific to computer and Internet delivery of educational programs. Barron (1998) suggests that the development of web-base d instruction opportuni ties should include: 1. A thorough media analysis and feasibili ty study to determine if web-based instruction is the appropri ate instructional choice; 2. A design strategy develope d around course objectives; 3. An analysis of the platforms and the target audience; 4. Meaningful interactions; 5. Consideration of visual guidelines; 6. A differentiation between intern al and external hyperlinks; 7. Limited page lengths; 8. A minimal use of audio, video, and plug-ins; and,

PAGE 32

24 9. Encouragement for collaboration. The Theory of Planned Behavior The theory of planned behavior, an exte nsion of the theory of reasoned action, draws from contemporary attitude theories (Ajzen, 1988; Ajzen & Fishbein, 1980; Fishbein & Ajzen, 1975). The original theory, the theory of reasoned action, posits that attitudes toward a behavior a nd subjective norms related to that behavior can be used to predict and understand an individuals intent ion to perform a behavior. An important assumption of this theory is that most socially relevant behavior s are under volitional control. Consistent with this assumption, a persons intention related to performing a behavior is the immediate determinant of the action (Ajzen & Fishbein). As such, attitudes and subjective norms related to a be havior can be used to understand and predict behaviors that are under volitional control. When dealing with behaviors that are under volitional control, a reasonable expectation is that individuals will carry out their intentions. In these situations, intentions and behaviors are highly correlated. Unfortuna tely, many behaviors are not under volitional control. For these behavior s, the degree of successful performance depends on factors outside an individuals cont rol as well as behavior al intentions (Ajzen, 1988). The theory of planned behavior (Aj zen, 1988) includes a pe rceived behavioral control component to help account for factors that affect the ability to translate intentions into action but are outside an individuals co ntrol. The degree of perceived control a person has over a given behavior many be influenced by both inte rnal and external factors (Ajzen, 1988). Internal factors aff ecting perceived behavioral control include

PAGE 33

25 information, skills, abilities, emotions, a nd compulsions. Opportunities and dependence on others are external factors that infl uence perception of c ontrol (Ajzen, 1988). In the theory of planned behavior, a central component is an individuals intention to perform the behavior of intere st. This feature is consistent with the theory of reasoned action (Ajzen, 1988). The theory of planned be havior differs from the theory of reasoned action because it postulates three determinants of intentions. The first two, attitude toward the behavior and subjective norm, are consistent with the original theory. The third and novel determinant is the degree of perceived behavi oral control (Ajzen, 1988). Perceived behavioral control re fers to the perceived ease or difficulty of performing the behavior and it is assumed to reflect past experience as well as anticipated impediments and obstacles (Ajzen, 1988, p. 132). Perceived behavioral control has two im portant features. First, perceived behavioral control has motivational implications for intentions. Individuals who believe they have the resources, oppor tunities, knowledge, and abi lity to perform a certain behavior are likely to have str ong behavioral intenti ons to engage in th at behavior (Ajzen, 1988). Conversely, individuals who perceive that they do not ha ve the resources or skills needed to carry out the behavior are unlikely to form a positive intenti on. This feature of perceived behavioral control has an impact on behavior that is me diated by behavioral intentions. The second feature of percei ved behavioral control is that it can have a direct effect on behavior. The theory of planned behavior posits that perceived behavioral control helps predict behavior independently from behavioral intention because it may

PAGE 34

26 reflect actual control with a degree of accu racy (Ajzen, 1988). The actual lack of opportunities and resources to engage in a beha vior impact an individuals ability to perform the behavior and may be measured partially by th e perception of behavioral control. To summarize the two features, pe rceived behavioral c ontrol can influence behavior indirectly through inte ntions and directly as a proxy m easure of actual control. Perceived behavioral control is ba sed on past experiences, second-hand information related to observing and discussi ng behavior, and other f actors that increase or decrease the perceived difficulty of perfor ming the task (Ajzen, 1988). Interestingly, the link between perceived behavioral control and behavior is only expected to emerge when agreement exists between perception of control and an individuals actual control over a behavior (Ajzen, 1988). Additionally, the theory of planned behavior reduces to the theory of reasoned acti on when behavioral control approaches its maximum and issues of control are not among an individua ls important considerations (Ajzen, 1988, p. 136). Unfortunately, attitudes, subjective norms, and perceived behavioral control are latent, hypothetical constructs that must be inferred from a variety of observable responses (Ajzen, 1988, 1991). Ajzen (1988, 1991) argues that salient beliefs are the antecedents ultimately determining indivi duals intentions and actions. Whereas individuals can hold a great many beliefs, th ey can only attend to a relatively small number at any given point in time. Ajzen (1991) states that, It is these salient beliefs that are considered to be the prevailing dete rminants of a persons intentions and actions (p. 189). Three kinds of salien t beliefs are identifie d in the theory of planned behavior:

PAGE 35

27 behavioral beliefs, normative beliefs, and control beliefs (Ajze n, 1991). Respectively, these beliefs constitute the underlying determ inants of attitudes, subjective norms, and perception of control toward a behavior (Ajzen, 1991). In addition to beliefs, attitudes, and s ubjective norms, theoretical discussions related to understanding and predicting behavior also incl ude the importance of factors such as demographic variables and personality traits. These factors, referred to by Ajzen and Fishbein (1980) as external variables, pl ay a moderating role. Instead of impacting behavioral intention directly, external variables influence the relative importance that attitudes, subjective norms, and perceived be havioral control cont ribute to behavioral intentions (Ajzen & Fishbein). Ajzen (1991) identifies thr ee conditions that must be met for the theory of planned behavior to be accurate in predicti ng intended behavior. These elements include: 1. Intentions and perceptions of control must be as sessed in relation to the particular behavior of inte rest, and the specified context must be the same as that in which the behavior is to occur (Ajzen, 1991, p. 185). 2. Intentions and perceived behavioral contro l must remain stable in the interval between their assessment and observa tion of the behavior (Ajzen, 1991, p. 185). 3. Prediction of the behavior from percei ved behavioral control should improve to the extent that percepti ons of behavioral control re alistically reflect actual control (Ajzen, 1991, p. 185).

PAGE 36

28 The efficacy of the theory of planned behavior was evaluated in a meta-analytic review. This review of 185 independent studies found that the theory accounted for 27% of the variance in behavior and 39% of the variance in behavioral intention (Armitage & Conner, 2001). A structural diagram of the underlying constructs of the theory of planned behavior is presente d in Appendix A. This grap hic depicts the relationship between attitudes, subjective norms, perceived behavioral control, be havioral intentions, and behavior that are presented in Ajzens (1988) theory of planned behavior. For comparison, a diagram of the theory of r easoned action is presented in Appendix B. Predicting Intention and Behavior Re lated to Computer Technologies Studies in management and information scie nces have used the theory of planned behavior and other closely related models to predict technology rela ted behavior based on behavioral intentions, attitudes, subjective norms, and perceived behavioral control (Davis, 1989; Davis, et al., 1989; Mathies on, 1991; Morris & Venkatesh, 2000; Taylor & Todd, 1995a; Taylor & Todd, 1995b). These studie s provide important insight as to how health educators might respond to the use of computer technologies, and ultimately, the use of computers and the Inte rnet for continuing education. A workforce-based study of 118 participants involved in the implementation of a new technology and a study of 786 business school students lend support to the usefulness of the theory of planned behavior in predicting compute r-related behavioral outcomes (Morris & Venkatesh, 2000; Taylor & Todd, 1995a, 1995b). Both of these studies demonstrate strong predictive ab ility even though both rely on actual measurement of behavior as opposed to self-r eport measures. Interestingly, Armitage

PAGE 37

29 and Conner (2001) suggest that the theory of planned behavior is more accurate in predicting self-report behaviors than behaviors that are dire ctly observed. Additionally, in both studies the participan ts were exposed to, or trai ned, in using the computer technology being studied. Overall, the theory of planned behavior model provides a good fit in predicting the use of a computing resource center (Taylor & Todd, 1995a, 1995b). Attitude, subjective norm, perceived behavioral control and their antecedent belief conditions have been shown to influence computer-related behaviors significantly (Morris & Venkatesh, 2000; Taylor & Todd, 1995a, 1995b). Intere stingly, Taylor and Todd report (1995a, 1995b) the addition of the perceived behavior al control variable did not appear to increase the prediction of usag e behavior. This finding is co nsistent with Ajzens (1988) argument that the theory of planned behavior reduces to the theory of reasoned action as perceived behavioral control in creases. Since all of the s ubjects had both experience and training using the computing resource cente r, a reasonable expectation is that the behavior would be considered volitional, w ith perceived behavioral control having little impact on behavior. Continuing Education for Health Educators The public health education workforce plays an important role in improving public health (Allegrante, et al., 2001). Although reports have alluded to the changing context in which health educators practice (Allegrante, et al., 2001; US DHHS, 1998), less attention has been give n to how these changes are influencing the competency

PAGE 38

30 requirements and continuing education needs of currently employed health educators (Allegrante, et al., 2001). Entry-level competencies for health e ducators have been identified (National Commission for Health Education Credential ing, Inc., 1996). These competencies are used as the basis for the Certified Health Education Specialist (CHES) exam, as well as for developing the curricula in health e ducation professional preparation programs (Schwartz, ORourke, Eddy, Auld, & Smit h, 1999). Additionally, graduate-level competencies for health educators, conti nuing education competencies for the public health education workforce, core competenci es for public health professionals, core competencies for the current public health workforce, and health informatics competencies have been identified (Allegr ante, Moon, Auld, & Gebbie, 1998; Allegrante, et al., 2001; National Commission for Hea lth Education Cred entialing, Inc., 1999; OCarroll, 2002; US DHHS, 1997). When fo cusing specifically on continuing education and workforce training needs, a consensus pa nel of experts formulated recommendations for public health core competencies (G ebbie & Hwang, 1998). These competencies included: 1. Public health values and acculturation; 2. Epidemiology, quality assurance, and economics; 3. Informatics; 4. Communication; 5. Cultural competency; 6. Team building and organizational effectiveness;

PAGE 39

31 7. Strategic thinking and planning/visioning; 8. Advocacy, politics, and policy development; and, 9. External coalition building and mobilization. To address the needs of health educator s, an expert panel identified continuing education competencies for the currently employed public health education workforce (Allegrante, et al., 1998). These competencies included: 1. Advocacy; 2. Business management and finance; 3. Communication; 4. Community health planning and deve lopment, coalition building, and leadership; 5. Computing and technology; 6. Cultural competency; 7. Evaluation; and 8. Strategic planning. Although the multiple competency suggestions may require future revisions, they offer an initial content base to use in the development of training programs for health educators. A summary of various conti nuing education competencies is found in Appendix C. Although the above projects have identified necessary competencies and probable future needs (Allegrante, et al. 1998, 2001; Gebbie & Hwang, 1998; US DHHS, 1998), little information is available related to the ac tual continuing education practices of health educators. The available information focu ses on outcomes related to specific programs

PAGE 40

32 (Umble, et al., 2000), information generated by expert consensus pa nels comprised of health educators (Allegrante, et al., 2001), or anecdotal evid ence (Goldman, et al., 2002). This information offers little insight into the continuing education wants, needs, and practices as expressed by health educators themselves. Although delivery systems and methods fo r delivering continuing education are identified as an important area to address (Allegrante, et al., 1998, 2001), a literature gap exists regarding how employed health educat ors feel about current continuing education offerings and what they think about their m ode of delivery. Anecdotal evidence suggests that although a core group of people use jour nal self-study articles and videotapes from past meetings, the majority attend professi onal meetings to receive credit and have networking opportunities (M. E. Auld, pe rsonal communication, September 17, 2001). Additionally, electronic discussions suggest programs offering continuing education credit are not readily ava ilable and can be costly (e.g., HEDIR archive, 2001). Computer-mediated Instruction for Con tinuing Education in Health Education Data from two randomized studies, bot h with 57% response rates and with respondents numbering 721 and 226, respectivel y, show that over 80% of health educators have access to computers and the Internet and that a majority report using basic computer and Internet applica tions (Brown, et al., in pres s; Hanks, et al., 2000). These findings suggest health educators are in a favorable position to use computers and the Internet for continuing education. Moreover, expert panels argue that computer-mediated instruction as a form of distance learning has the potential to improve skills in the public health workforce (US DHHS, 1998), including health educators.

PAGE 41

33 Empirical data related to the use of co mputer-mediated instruction for continuing education in health education is limited. A small-scale study collecting input from individuals who had participat ed in an online, qualitative evaluation course provides important insight concerning computer-m ediated instruction among public health professionals. This study demonstrated an improvement in participants skill levels in collecting qualitative data and in participants self-efficacy. Additionally, most participants rated the lessons as valuable a nd the teaching methods satisfactory (Steckler, et al., 2001). The most common difficulty reported was finding time to complete the modules, and participants did not exhibit a signif icant change in knowledge of beliefs about quantitative methods (Steckler, et al.) Challenges identified related to course development included: (1) structuring courses to fit into a professionals busy schedule; and, (2) increasing organizati onal and supervisory support for programs that improve job skills (Steckler, et, al.). Online Data Collection The ability to collect data utilizing com puter and Internet te chnologies has made an important contribution to data collection efforts and provides an additional tool for researchers to collect valuable informa tion. Couper and Nicholls (1998) note: Survey research has undergone many important changes in the last half century. These include the development of applie d probability sampling, the growth of telephone interviewing, new approaches to statistical an alysis, a greater understanding of nonsampling bias and e rror, the evolution of panel survey methods, and the appreciation of cognitive psychology principles in questionnaire

PAGE 42

34 design. Yet, none of these changes may have more far-reaching effects on survey research than the application of computer methods to survey data collection and capture. This transition promises to be a crucial turning point for survey practice (p. 1). Using an online survey tool to collect survey information can be effective when used with appropriate populat ions and when sound administrative methods are followed. As with other forms of self-administered data collection tools, this method has both advantages and disadvantages. Some advantages include: 1. Low cost; 2. Ease and convenience; 3. Speed of delivery; 4. Improved data quality; 5. Improved efficiency of data collection; and 6. The potential to overcome international boundaries (Couper & Nicholls, 1998; Dillman, 2000; McDermott & Sarvela, 1999; Ramos, Sedivi, & Sweet, 1998). When discussing the advantages and di sadvantages of online surveying it is important to look at survey error. As with other forms of self-administered survey delivery, sampling error, nonresponse error, measurement error, and coverage error present major concerns (Dillman, 2000). Samp ling error is any difference between the statistic measured in the sample and the actual parameter of the population from which the statistic has been calc ulated (McDermott & Sarvela, 1999, p. 274). Online surveys

PAGE 43

35 have the potential to reduce or eliminate sampling error by increas ing the number of individuals surveyed with little additional cost. Once an el ectronic collection system has been developed, the per-person cost of adding people is much less than in traditional survey formats (Dillman). However, rese archers should not attempt to survey large volumes of participants if the increase in volume is going to have a negative impact on the survey response rate (Dillman). Nonresponse error occurs when a significant number of people in the survey sample do not respond to the questionnaire and have different characteristics from those who responded (Dillman, 2000, p. 10). Computer and Internet l iteracy along with access to computers and the Internet influence an individuals ability to respond to an online survey (Dillman). As such, individuals within a survey sample who do not have the skills and access necessary to respond are not able to submit information. This does not provide an accurate representation of the population sampled. Also, the appearance of the survey and the addition of elaborate graphics requiring long download times can impact an individuals ability to respond to a survey (Dillman). This event can prevent individuals who have slow Internet access and older computers from responding. The third type of error, measurement error, has many similarities between paperand-pencil and online surveys. Measurement error results from poorly worded questions or questions presented in a way that leads to inaccurate or uninterpretable results (Dillman, 2000). When designing online instrume nts, care should be taken to avoid the use of question structures that have known measurement problems. Also, the design and presentation of the questions should carefully consider the possibility of questions being

PAGE 44

36 displayed in a format that can be misread or have response options not appearing on the screen (Dillman, 2000). Another source of measurement error comes from a method effect. Although important to consider, th e method effect for online data collection should be consistent with other self-administere d surveys. The final source of error, c overage error, presents the greatest concern for online survey instruments. Coverage error occurs when the list from which the sample is drawn does not include all elements of the population (Dillman, 2000, p. 10). Unfortunately, all individuals do not currently have access to computers and the Internet making a survey of all U.S. households imprac tical (Dillman). In ternet-based surveys may have appropriate coverage if care is ta ken to survey groups who use the Web and electronic mail regularly (Dillm an). Additionally, Dillman warns obtaining responses from large numbers of respondents cannot be substituted for meeting the survey requirement of good coverage (p. 355.) a nd recommends using Web techniques to ensure that the survey is limited to invited participants. Researchers also caution that Internet security and anonymity concerns may discourage individuals from answering Inte rnet-based questionnaires (McDermott & Sarvela, 1999; Ramos, et al., 1998). The future of the Internet is difficult to predict, and these concerns may persist or be quickly resolved. However, until these issues are addressed, electronic surveys should be limite d to impersonal, non-threatening content (McDermott & Sarvela, 1999). Dillman (2000) offers 14 principles to consider when constructing Web-based surveys, including:

PAGE 45

37 1. Introduce the questionnaire with a motivational welcome screen that has clear instructions on how to proceed with the survey; 2. Use an identification number to limit access to people in the sample; 3. Choose an interesting first question; 4. Use a conventional format, simila r to paper self -administered questionnaires; 5. Limit use of color so survey is easy to read; 6. Avoid differences in the visual appe arance of questions that result from different computer configurations; 7. Provide good instructions related to necessary computer actions; 8. Use drop-down boxes sparingly; 9. Do not make movement through the survey contingent on providing answers to all questions; 10. Provide clear instructions related to skip patterns; 11. Construct questionnaires that scroll through the questions; 12. Format questions so all possible an swers appear on a single screen; 13. Use symbols to give participants a re presentation of how far they have progressed through the survey; and 14. Limit or avoid the use of question stru ctures with measurement problems. Health educators have been using the Internet to collect Web-based survey for the past few years. Topics th at have been studied include risk taking behavior and temperament (Daley, McDermott, Brown, & Kittleson, 2003; Daley, 2000), health risk

PAGE 46

38 behavior (Pealer & Weiler, 2000), recreati onal drug use by successful adults (White, Nicholson, & Duncan, 2000), and computer us e (Perlmutter, Brown, & Ellery, 2000), to name a few.

PAGE 47

39 Chapter Three: Methods This chapter describes the methods used to conduct this study and is divided into eight sections: (a) research questions; (b) population; (c) research design, sample, and study administration; (d) survey development a nd pilot testing; (e) data collection; (f) operationalization of variables; (g) data analysis; and, (h) potential problems, threats, and ethical considerations. Research Questions The research questions investig ated in this study included: 1. What is the association between hea lth educators perceived behavioral control related to using self-paced, co mputer-mediated continuing education programs and their previous experience with computer-mediated education? 2. What is the association between health educators attitudes related to using self-paced, computer-mediated conti nuing education programs and their previous experience with computer-mediated education? 3. What is the association between health educators subjective norms related to using self-paced, computer-mediated c ontinuing education programs and their previous experience with computer-mediated education? 4. Do health educators intentions to use self-paced, computer-mediated continuing education programs mediat e the association between perceived

PAGE 48

40 behavioral control, attitudes, and subjective norms and their previous experience with computer-mediated education? 5. For individuals with a positive intention to use computer-mediated instruction, what characteristics studied help di fferentiate between those who have previously used this learning medium and those who have not? Population According to the Standard Occupational Clas sification (SOC) at the Bureau of Labor Statistics (2001), health educators are defined as individuals who: 1. Promote, maintain, and improve indivi dual and community health by assisting individuals and communities to adopt healthy behaviors. 2. Collect and analyze data to identif y community needs prior to planning, implementing, monitoring, and evalua ting programs designed to encourage healthy lifestyles, policies and environments. 3. May also serve as a resource to assist individuals, other professionals, or the community, and may administer fiscal re sources for health education programs. Unfortunately, a census listing of these health educators is not available. Instead, a convenience sampling of health educators was used to examine the relationships of the study variables. The convenience sample in cluded members of the Society for Public Health Education (SOPHE) listed in the 2002-2003 publishe d directory (N= 1882). Membership in SOPHE represents health education professionals in private, public, and non-profit sectors. Eligibility for membership requires an individual to possess one of the following characteristics:

PAGE 49

41 1. A graduate or undergraduate degree from a formal health education program. 2. Employment in a health education capacity. 3. Faculty responsibilities in a health education program. 4. An undergraduate degree from a program approved by the SOPHE/American Association for Health Education Baccalaureate Program Approval Committee or a gr aduate school accredited by the Council on Education for Public Hea lth (Society for Public Health Education, No date). Research Design, Sample, and Study Administration This investigation used a cross-sectional survey design, and the procedures followed were reviewed and approved (IRB #100277) by the Office of Research, Division of Research Compliance Institutional Review Boards (IRB) at the University of South Florida. Because this study was conduc ted using an online questionnaire, a waiver of written consent was requested and granted. The approval forms associated with the IRB process can be found in Appendix D. The names and electronic mailing addresse s of potential participants were inspected for accuracy and any missing or inac curate addresses (N=121) were researched using Internet-based database directories (e.g., commercial directories, university directories, organization dire ctories) and updated when th e available information was enough to assure the identity (N=11). The indi viduals on this list were imported into contact management software (Goldmine, Version 5.5; FrontRange Solutions, Colorado

PAGE 50

42 Springs, CO), which allowed for the generati on of personalized electronic messages and assisted in tracking respondents and facilita ting the distribu tion of follow-up reminders to improve response rate. The survey was conducted in accordance with Dillmans (2000) recommendations for online surveys and fo llowed his 14 principles for conducting Web surveys. The researcher and individuals contributing to this project and who are also members of the organization were excluded fr om participating in the study (N=4). Potential participants were sent a pre-notice electronic message. This message provided a brief description of the survey objectives and the date on which they would receive the invitation to complete the surv ey (see Appendix E). Messages that were returned as undeliverable (N= 444) were inspected and electronic mailing addresses updated when possible (N=70). Individuals requesting removal from the study (N=136) were noted in the contact management system and removed from the study. All members remaining (including individuals for whom valid electronic mailing addresses were available, individuals not requesting to be removed from the study, and individuals not serving in the development of the surv ey instrument) comprised the final study population (N=1259). Individuals in the study populat ion were contacted electr onically a second time. This invitation to participate in the study included a personalize d letter, a personal identification number (PIN), and a link to th e survey website (see Appendix F). This invitation to complete the survey was sent within two or three da ys of the pre-notice (three days for individuals whose messages were not returned as undeliverable, and two days for most individuals whose messages were undeliverable but a

PAGE 51

43 replacement/corrected address was located) to incr ease the likelihood that the participants connect the memory of the first contact with the second (Dillman, 2000). Following three business days, a follow-up reminder with a survey link was forwarded to all nonresponders (see Appendix G). This process was repeated after an additional five business days for individuals continuing as non-res ponders (see Appendix H). In total, the survey was open to receive responses for 26 days. Survey Development and Pilot Testing The survey instrument was developed based on recommendations from Ajzen (1988, 1991, 2001), Dillman (2000), and McDermott and Sarvela (1999), and it was pilot tested in accordance with the methods recommended by McDermott and Sarvela. The survey questions were based on the constructs associated with the theoretical framework and related surveys identified in the review of the literature (Mathieson, 1991; Morris & Venkatesh, 2000) and presented by Ajzen (1988, 1991, 2001, 2002a, 2002b). Consistent with the recommendations and operationaliza tion of previous researchers (Ajzen, 1988, 1991; Ajzen and Fishbein 1980; Davis 1989; Mathieson, 1991), questionnaire items related specifically to the use of compute r-mediated instruction fo r continuing education as opposed to general computer usage. Each predictor was assessed directly by asking respondents to make judgments based on a set of scaled questions (Ajzen, 2002b) as opposed to indirectly, or based on responses to questions related to the corresponding beliefs. Multiple items were used to measure each pred ictor to help improve the reliability of a self-reported measure, and the items were interspersed in a non-systematic order (Ajzen, 2002b). Informa tion related to demographic characteristics and general

PAGE 52

44 information important in describing the su rvey respondents was also collected. The survey tool was exposed to face and content va lidation and temporal stability (test-retest reliability) testing. Initially, question items were developed based on information available in the literature. These items were then subjected to questioning through in-depth interviews with eight individuals (N=8). These individuals were asked to respond to general questions about the survey and provide feedback related to the clarity and interpretation of the questions. Additionally, feedback from these interactions was used to determine appropriate response categories for questions identify individuals and organizations important to the subjective norm predictor va riable, determine potential incentives to improve response rates, and decide whether a reference time frame would be utilized. The results of these interviews and a summa ry of the updates to the questionnaire are presented in the results section. When designing a survey, validity is the most important consideration (Bernard, 2000; McDermott & Sarvela, 1999; Pedhazu r & Schmelkin, 1991). McDermott and Sarvela (1999) define validity as the appropriateness, mean ingfulness, and usefulness of specific inferences made from test scores (p. 140). Bernard (2000) suggests that validity is the accuracy and trustworthiness of instruments, data, and findings in research (p. 46). Face validation confirms that, the instrument appears to measure the construct under consideration, and appears to be appr opriate for the audience for which it is intended (McDermott & Sarvela, 1999, p. 141). Content validation es tablishes that the items on the survey are representative of the domain being assessed. Th e next step in the

PAGE 53

45 survey development process was to get fee dback from a panel of experienced health educators, prominent continuing education professionals, individuals familiar with computer-mediated instruction, and Internet su rvey practitioners. This information was used to help establish both face and c ontent validity (Berna rd, 2000; McDermott & Sarvela, 1999). The individuals comprising th is panel and their areas of experience are listed in Appendix I. The results from this review panel and a summary of the updates that were made to the questionnaire are presented in th e results section. The survey was updated based on feedback from the panel and coded into its online format. To ensure that the technology for collecting informati on using this online tool was functioning correctly, multiple individuals tested the online version of the survey instrument. Thirty-five health educator s were invited to participate in field testing of the survey. This group was asked to comp lete the updated survey in its online format. In addition, individuals in this group were asked to answer the survey a second time after a two-week period to determine test-retest relia bility. The results from this field test, including percent agreement values and per-item correlations, and a summary of the updates that were made to the questionnaire are presented in the results section. Percent agreement values and correlations of 0.80 or a bove were used as the criterion that the items on the instrument were stable over time (McDermott & Sarvela, 1999). Items with modest and strong correlations (0.6-1.0) were retained on the final survey tool. Items with correlations less than 0.6 were include d if a compelling case for their theoretical importance could be argued. The initial su rvey contained 5-8 items to measure each

PAGE 54

46 construct. Input from the review panel, feedback from the pilot tester s, and the test-retest results were used to determine which items were retained on the final survey tool. Data Collection The online survey tool formatting was based upon the recommendations established by Dillman (2000). Web program ming software (Dreamweaver, version 4.0; Macromedia, Inc., San Francisco, CA) was used to develop the site using a combination of HTML and JavaScript. Additionally, Perl server-side programming language compiled and stored the survey data (White Carey, & Dailey, 2001). The coding for the instrument followed website development principles to improve the ability of the tool to be viewed across various hardware and soft ware platforms (White, et al., 2001). To reach the survey instrument from the opening screen of the website, participants were asked to enter a personal identification number (PIN) and a password (their first name as it appears on the electronic message they receiv ed). This process restricted access to the survey tool to only invited participants. Data from the online survey were formatted and submitted into a text file. This reprocessed, delimited text f ile was ready for direct import into SAS (version 8.02; SAS Institute Inc, Cary, NC) for st atistical analysis (W hite, et al., 2001). The only identifying characteristics associated with the text file containing the survey results were the PIN and password, and these were removed before anal ysis. Including these fields initially allowed for tracking of responders and identifi cation of duplicate entries. The data files generated by the survey were located on a s ection of the computer server that did not have public access, and these files were passw ord protected to limit data access to the

PAGE 55

47 investigator alone. As a precaution, electroni c mail messages containing survey data also were generated and sent to the researcher. This back-up plan permitted the researcher to recreate the dataset should an unexpected malfunctioning in the technology occur (Daley, et al., 2003). Operationalization of Variables The similarities between the theory of planned behavior and the theory of reasoned action are important in the concep tualization of relationships between the constructs in this study. Fo r individuals perceiving a high level of control over a behavior, the theory of planned behavior redu ces to the theory of reasoned action (Ajzen, 1988). Behaviors in the theory of reasoned action are volitional, and these behaviors have a high correlation with their behavior intentions (Ajzen & Fishbein, 1980). The survey tool for this study co llected information about an individuals previous experience with computer-mediated instruct ion and future intention to pa rticipate in the behavior. Based on the correlation between intentions and behaviors no ted in previous research, a proxy measure was developed for behavior. Pa rticipants reporting previous experience with computer-mediated instructi on and a future intention to participate in this type of a training opportunity were considered to elicit the behavior, and the remaining respondents were classified as not having th e behavior. This be havioral proxy, which was called Behavior, was the depe ndent variable for this study. To determine experience level, respondents were asked multiple questions based on their previous par ticipation in computer-mediated learning opportunities. These participants were asked to an swer yes or no to a list of questions. This list included

PAGE 56

48 participation in: (1) a cla ss/workshop offered on a CD-ROM, (2) a class/workshop offered from a website, (3) a web telecast, (4 ) an Internet-based discussion, (5) or any other type of computer-mediated learni ng opportunity. Another question assessed whether any type of formal certification or cr edit was received base d on participation. Individuals who participated in at least one item from the list was classified as experienced and coded as 1. The remaining re spondents were classified as inexperienced and coded as 0. The same items were used to determine future intentions related to computermediated instruction. Indivi duals reporting a willingness to participate in at least one form of computer-mediated opportunity were classified as having a positive future intention, whereas the remaining individuals we re classified as ha ving a negative future intention. Final coding of the dependen t variable was accomplished by coding an individual as performing the behavior (coded 1) if the individua l was experienced and had a positive future inten tion. All remaining respondent s were considered as not performing the behavior (coded 0). Operationalization of the independent variables (attitude, subjective norm, perceived behavioral control) was based on individual responses to multiple question items. Multiple items were used to help ove rcome the validity and reliability deficiencies of single-item measures (A jzen, 2002b; Pedhazur & Schmelkin, 1991). Each question was scored on a Likert-type scale. A lis ting of the independent variables and the corresponding statements are found in Appendix J. The initial item list was reduced and

PAGE 57

49 revised during the planning and pilot testing. The final version of survey tool is found in Appendix K. Data Analysis The analyses performed on the data colle cted in this project used the SAS software program (version 8.2; SAS Institute Inc., Cary, NC), and the analysis plan involved univariate, bivariate, and multivariate analyses. Univariate and bivariate analyses Univariate statistics were used to report demographic information about the sample and were useful in da ta screening and exploring the shape of the data (Hatcher & Stepanski, 1994). The continuous variables were described using means and standard deviations, and frequency distri butions were used for categorical variables. The items that represented attitude, s ubjective norm, and perceived behavioral control, the year respondent began using the Internet, the res pondent age, and the year highest degree was attained are continuous vari ables and were described using means and standard deviations. The remaining variables were reported as frequencies. Associations between variables were examined by submitting variable pairs for analysis using cross-tabulations with chi-squa re tests and analysis of variance (ANOVA). Chi-square tests were used to compute the association between the categorical variable pairs (Hatcher & St epanski, 1994). These variables included Intention, Behavior, Continuing Education Behavior and the res pondent characteristic variables of Gender, Age, License/Certification, Location if Ta king Class, Professional Identification, Professional Role, Current Employer, and Hi ghest Level of Education. Analysis of

PAGE 58

50 variance with one between-gr oup factor (ANOVA) was used to compute the association between the scale-scored variables and the ca tegorical variables (Hatcher & Stepanski, 1994). The scale-scored variables include d Attitude, Subjective Norm, Perceived Behavioral Control and the categorical respondent characteristic variables included Gender, Age, License/Certifica tion, Location if Taking Class, Professional Identification, Professional Role, Current Employer, and Hi ghest Level of Education. ANOVA also was used to compute the associations be tween the categorical variables Intention, Behavior, Continuing Education Behavior and the scale-scored variables Attitude, Subjective Norm, Perceived Behavioral Contro l, while chi-square tests were used to compute the association between Behavior a nd the variables used to develop Intention (all categorical variables). Finally, subsets of observations with various levels of reported Intention were analyzed to determine associ ations between the respondent characteristic variables, barriers, and Behavior. Identifying subscale items Subscales representing latent constructs (or factors) were determined based on items identified through factor analysis. Expl oratory factor analysis using a principal axis method identified the underlying factor structure. Squared multiple correlations between variables were used as prior comm unality estimates (Hatcher, 1994). A promax rotation was used to rotate the final soluti on prior to interpretation. Hatcher suggests rotating the solution to make interpretati on easier, and a promax rotation was used because the factors were corre lated. Criteria for determini ng the number of meaningful factors to retain included eigenvalue one, a scree test, the proportion of variance

PAGE 59

51 accounted, and interpretabili ty criteria (Hatcher, 1994). The assumptions underlying exploratory factor analysis include: (1) interval-level measurement of variables, (2) random sampling, (3) a linear relationship between variables, and (4) normal distribution of observed variables (Hatcher, 1994). However, factor analysis is robust in the face of violations (M.S. Forthofer, personal co mmunication, March 24, 2002). Hatcher (1994) suggests that the number of observations subm itted for analysis should be at least five times the number of variables. The number of responses from this study exceeded this recommendation. Subscales were created for factors with at least three items having a significant loading of 0.50 or greater on that item and that also failed to reach significant loadings on any other factors (Hatcher, 1994). Cronbachs alpha was used to test the internal consistency of each subscale created. Because of multicollinearity concerns associated with regression analysis and because the it ems in each subscale were shown to be unidimensional through factor analysis, the it ems of each subscale were combined into a total score for entry into the multivariate anal ysis models (Pedhazur & Schmelkin, 1991). Although differential weights may be applied to the separate items (as in assigning weights from a factor analysis in producing factor scores), for most purposes, unit weighting (i.e., merely summing the separate re sponses, thus weighting each of the items equally) has been shown to produce satisfact ory results (Pedhazur & Schmelkin, p. 125). Since the test items were developed to m easure a similar perception but in multiple formats, unit weighting for summing pur poses was employed. To improve the association of the total score with the rating scale used, an average of all subscale items

PAGE 60

52 was implemented to represent the total score. These total scores, representing the factors or constructs, were used for multivariate modeling. Multivariate analyses Logistic regression was used for multiv ariate modeling since this procedure describes the relationship between a dichot omous dependent variable and a set of predictor variables having eith er categorical or continuous responses (Stokes, Davis, & Koch, 1995). Logistic regression enabled th e researcher to over come many of the restrictive assumptions impos ed by ordinary least square s (OLS) regression. The dependent variable, Continuing Educa tion Behavior, was operationalized as a dichotomous variable with the two res ponse categories repres enting performing the behavior and lack of performance of the behavior. The procedure for recoding the data to arrive at these two responses is described in detail elsewhere in this document and was based on respondents answers to questions a bout past experience and future intention related to using computer-mediated instruc tion for continuing education. The models were based on the probability of performing the behavior. The multiple independent, or predictor, variables were scored at both the categorical and continuous levels. Categorical variables were dummy coded for en try into the logistic models when more than two response categories were present. Once the full model was determined, diagnostics were run to ensure that the assumptions of l ogistic regression were not violated. Box-Tidwell transformations were used to assess linearity between the continuously scored variables and logit-P, a nd tolerance values were used to assess multicollinearity. However, this procedure is relatively robust in the face of violations,

PAGE 61

53 provided an adequate sample size has b een studied (M.S. Forthofer, personal communication, March 24, 2002). Preliminary logistic modeling was performe d to assess the relative contribution of the control variables. This procedure used the odds ratios and 95% confidence intervals and likelihood ratio tests (Pedhazur, 1997). Th e contribution of the control variables was assessed by regressing the control variables ag ainst the dependent variable and examining the odds ratios and 95% confidence intervals. When the 95% confidence interval around the odds ratio did not include the value of 1.0 the variable was considered a useful predictor in the logistic model. Entering the control variables into the model first allowed for testing the contribution of the predictor variables over and above the contribution of the control variables. Additionally, likeli hood ratio tests were used to determine if particular independent variables were more important than others. This procedure was accomplished by determining the difference betw een the -2 log like lihood of two models, one of which is nested in the other. This difference has an approximate chi-square distribution with the degrees of freedom equal to the difference in the number of parameters in the two models. In each calculation, the first -2 log likelihood was obtained from the full model, and the sec ond from the model reduced by the variable under investigation. Since the difference in predictors was one in each calculation, the degrees of freedom equaled one. A table of chi-square distribution was used (alpha=.05) to determine if dropping the variable significantly reduced the model fit. The research questions were answered using three models. Model 1, the full model, was represented by the equation: be havior = control variables + attitude +

PAGE 62

54 subjective norm + perceived behavioral control + intention. Mode l 2 was represented by the equation: behavior = control variables + attitude + subjective norm + perceived behavioral control. Model 3 was used to look at the subset of indi viduals with a positive intention to participate in computer-mediated education. This model was the same as the equation in Model 2; however, its application was to a subset of the total dataset where intention was positive. The first study question is: What is the association between health educators perceived behavioral control related to us ing computer-mediated continuing education programs and their behavior related to computer-mediated education? Model 2 was used to test the hypothesis that an associati on exists between perceived behavioral control and behavior related to computer-mediated education. If the 95% confidence interval around the odds ratio for the spec ified predictor variable in the model did not include the value 1.0, then that variable was considered to be associated with the behavior. The second study question is: What is the association between health educators attitudes related to using computer-media ted continuing education programs and their behavior related to com puter-mediated education? Again, Model 2 was used to test the hypothesis that an association ex ists between attitudes and be havior related to computermediated education. If the 95% confidence interval around the odds ratio for the specified predictor variable in the model did not include the value 1.0, then that variable was considered to be associated with the behavior. The third study question is: What is the association between health educators subjective norms related to using computer -mediated continuing education programs and

PAGE 63

55 their behavior related to computer-mediated education? Again, Model 2 was used to test the hypothesis that an association exists betw een subjective norm and behavior related to computer-mediated education. If the 95% confidence interval around the odds ratio for the specified predictor variab le in the model did not incl ude the value 1.0, then that variable was considered to be a ssociated with the behavior. The fourth study question is: Do health educators inte ntions to use computermediated continuing education programs me diate the association between perceived behavioral control, attitudes, and subjective norms and their behavior related to computer-mediated education? This mediating effect was tested by making comparisons between the values associated with the pr edictor variables in Model 1 and Model 2. Significant odds ratios (determined based on the 95% confidence interval around the odds ratio) identified in Model 2 but found to no longer be significant in Model 1 were considered to be completely mediated by inten tions. Significant odds ratios identified in Model 2 that decreased, but c ontinued to remain significan t were considered to be partially mediated by intentions. The fifth study question is: For individuals with a pos itive intention to use computer-mediated instruction, what characte ristics studied help differentiate between those who have previously used this learning medium and those who have not? Model 3 was used to test the hypotheses associated with applying this m odel to a subset of individuals who expressed a positive inte ntion toward using computer-mediated instruction. These hypotheses include: (1) An association exists between perceived behavioral control and behavior related to computer-mediated education in the subgroup;

PAGE 64

56 (2) An association exists between attitudes and behavior related to computer-mediated education in this subgroup; (3) An association exists between subjective norms and behavior related to computer-mediated educatio n in this subgroup; and (4) An association exists between the control variables and behavior in this subgroup. If the 95% confidence interval on the odds ratio did not include the value 1.0, th en the variable was considered to be associated with behavior. In addition to the multivariate model, the bivariate analyses performed on the subset of individuals having a positive intention to participate in computer-mediated education we re examined to identify characteristics differentiating individuals who have and have not previously participated in computermediated instruction. Potential Problems, Threats, and Ethical Considerations The use of Internet and other comput er-mediated technologies pose potential problems to this study. Message delivery, we bsite availability, loss of data, and the functionality of the data collection tool across multiple computer platforms were all issues that warranted considera tion. Frequent backing up of a ll files related to the project took place, and different comput er and server locations for va rious aspects of the project were utilized to help decrease the impact of any technological malfunctioning. To assure the integrity of the data colle cted, an electronic mail version of each respondents survey answers was sent to the investigator so the da ta could be housed in two different locations and in two different electronic formats. Data from the electronic messages could be used to recreate the data set for analysis if necessa ry. Consideration for th e ability of the tool

PAGE 65

57 to appear similarly across multiple computer platforms was taken into account when coding the tool for online use. Instrumentation posed the greatest threat to the validity of this study, and coverage was the biggest measurement concer n. Since the respondent s were required to complete the survey online, individuals who were more comfortable with online technologies may have been more likely to pa rticipate. Coverage was impacted by the nature of the sample within the population. Unfortunately, the convenience sample selected for this study provided limited covera ge of the health edu cation population as a whole. However, SOPHE may consist of th e broadest representation of employment venues from among the universe of nationa l health education organizations. Protecting the identity of the respon dents was the most important ethical consideration related to this study. As with all surveys that utilize a multiple contact administration procedure that eliminates i ndividuals who have already responded, this survey did not promise anonymity to respondent s. Additionally, the electronic format of the survey increased the likelihood of a doubl e submission. A PIN was used to locate and evaluate duplicate entries, as well as to facilitate non-resp ondent notification. However, the connectors between the responde nts answers to the survey items as recorded for analysis and their identities were removed prior to analysis. Procedures to assure the confidentiality of responses were followed. The possible negative impact of not being able to assure anonymity was redu ced because the survey did not request any personal health information or information co nsidered highly sensitive. Pealer and

PAGE 66

58 Weiler (2000) found response rates consistent w ith traditional survey methods even when anonymity was not given to the subjects part icipating in a risk behavior survey. Summary Data collected during this study were explored using univariate and bivariate procedures. Factor analysis was employed to develop subscales based on the identified factors. Following this data reduction exer cise, multivariate analysis using logistic regression was used to identify statistically significant relationships and to answer the research questions..

PAGE 67

59 Chapter Four: Results The results from this study will be presented in four sections, including results from in-depth interviews, review panel fi ndings, field testing, and quantitative survey results. Results from the In-Depth Interviews The initial data collection for this study included the qualitative results from 8 indepth interviews. These interviews were completed during a fou r-week period between January 8 and March 15 in 2003. Both individua ls who had previously participated in computer-mediated instruction (N=4) a nd individuals who had not previously participated in computer-mediated instructi on (N=4) participated in interviews that utilized a interview guide and lasted a pproximately one hour. The semi-structured interview guide and the compiled comments from these interviews are available in Appendix L, and Appendix M, respectively. During this effort, individuals were asked to respond to questions to determine if the definitions presented at the beginni ng of the survey were understandable and appropriate, to identify information related to the constructs that s hould be included in survey items, to comment on the inclusion of time frames with some questions, to identify barriers, and to s uggest incentives to improve pa rticipation. In itially, four individuals were interviewed. The survey instrument was then updated based on the

PAGE 68

60 information obtained, and the remaining four pa rticipants were interviewed. This second group of interviews did not introduce any new ideas related to the study topics to the project, so the qualitative inquiry was consid ered to have reached saturation. Additional updates to the questionnaire items were completed following the second set of interviews. Responses from the individuals interviewe d suggested that th e definitions were appropriate for this study. Also, the participan ts did not feel time qualifiers were needed for the questions related to experience. Se veral individuals interv iewed suggested that nothing would increase the likeli hood of them taking this surv ey other than being sent a message over the Internet invi ting them to participate. Other individuals interviewed, however, suggested they may be more likely to answer the questionnaire if the message come from a recognized sender, if the surv ey was easy to access and complete, if an organization from which they belonged and had a vested interest encouraged them to take the survey, if they received a decrease on thei r next membership fee, or if they were offered a summary of results. The individuals interviewed were asked if they were more likely to respond if their participation entered them in a chance to ha ve their registration fee paid at the next professional conference. Although a couple of the individuals interviewed responded maybe after prompti ng, most reported that they did not think this was an effective suggestion. From these incentives, the suggestion that a participant receive a reduced membership at next re newal was eliminated because it was well beyond the project budget. Also, although the researcher attempted to secure the endorsement of SOPHE and ear ly communication with the organization was favorable, ultimately, support could not be obtained by the date the survey was distributed. The

PAGE 69

61 participant incentives employed for the study included making the survey easy to access and complete, an offer of a summary of the re sults, and an offer to be sent the list of online learning opportunities comp iled from survey respondents. Review Panel Findings The next phase of the study consisted of inviting individuals who were currently working in areas related to the study to co mment on the face and content validity of the items that had been developed for each c onstruct. This panel was comprised of individuals who are practicing professionals engaged in work related to the study focus (N=12), and individuals serving on the di ssertation committee (N=5). Of these 17 individuals, one responded that her current schedule precluded her completing this task during the time specified, nine returned st ructured questionnaire s, and seven did not respond. When asked their areas of experience related to this project, three respondents reported having experience in computer-med iated instruction, th ree reported having experience in continuing education, 6 reported having experience in health education, and six reported having experience in Internet survey delivery. During this process, respondents used a st ructured review form (Appendix N) to rate the questionnaire statements. The documen t used to collect information from these individuals was of a rich text file with form fields, and it wa s delivered as an electronic mail attachment. Key issues addressed incl uded each statements ability to measure the construct, the ability of a composite score comprised of multiple stat ements to represent the construct, and identificat ion of barriers to computer-m ediated instruction use among health educators. In addition to scored responses, respondents were invited to make

PAGE 70

62 qualitative comments. Both the quantit ative data and respondents comments are available in Appendix O. Updates to the survey question items we re completed based on the narrative feedback received from reviewers. At th e suggestion of one reviewer, an item from Perceived Behavioral Control was removed due to its vagueness. One item listed initially under the Attitude construct, was moved to the Perceived Behavioral Control construct at the suggestion of another review er. Although this item had an average score less than good (2.0 on a scale ranging from 0 to 3), it wa s retained in the study and underwent field testing. The remaining items from the constructs Perceived Behavioral Control, Attitude, and Intenti ons that had an average scor e less than good (2.0 on a scale ranging from 0 to 3) were remove d from future versions of the survey questionnaire. For the construct Subject ive Norm, items were updated based on the comments received from reviewers and retained, even though the average scores for three of the six items were lower than good. These items remained on the questionnaire because substantial updates had been made to the wording of the items and they were deemed necessary to assure three or more items emerged from the field testing process. A few reviewers expressed concern that the respondents might not understand what is necessary to complete computer-med iated instruction, and therefore, would be unable to answer the questions. To address this issue, some links to examples of selfdirected computer-mediated continuing education programs that individuals could view were added to the introduction. Additi onally, a few reviewers commented on the technical nature of the ques tion items. In response, th e language on the survey was

PAGE 71

63 changed. Self-directed was changed to se lf-paced, and computer-mediated instruction was shortened to computers. Both of these term combinations were defined in the introduction. Three experimental items and three additional barriers were added to the questionnaire based on the feedback from th e review panel. The experimental items included: 1. If offered the chance to take a self-paced course that was of interest to me and that was delivered using a computer, I am confident that I would be able to complete it; 2. Prior to participating in this surve y, I was aware that self-paced programs delivered using computers were an option for continuing my education; and, 3. I have tried to find opport unities to participate in continuing education programs that are self-paced and delivered using computers. The barrier items added included: 1. Lack of professional reward s for continuing education; 2. Lack of programs; and, 3. Continuing education is not importa nt for health education/health promotion. No barriers from the list sent to reviewers were removed because each barrier listed was selected by at least four of the reviewers.

PAGE 72

64 Reviewers were asked to rate the ability of three or more items combined as a composite score to represent each construc t. Replies varied greatly, with reported average scores ranging from 3.4 to 3.8 on a 6 poi nt scale (0 to 5 with 5 representing strongly agree). This score s uggested that whereas respondents were more likely to agree with the statement than they were to disagr ee, they were not in strong agreement. An additional update to the project that was made based on reviewers comments was the operational definition of the term e xperienced. Several reviewers commented that they did not feel that an individual who had particip ated in a single computermediated course should be considered e xperienced. The variable experience was developed to refer to prior part icipation rather than level of expertise. A person who is denoted as being experienced is one who ha s previous experience participating in a computer-mediated program, rather than so meone who has acquired a level of expertise in computer-mediated instruction. Field Testing Following updates to the survey based on in-depth interviews and feedback from the review panel, the survey questionnaire wa s tested in its online format and assessed for test-retest reliability. This procedure consis ted of soliciting 35 individuals in various health positions who were not listed as me mbers of SOPHE in th e 2002-2003 directory. Initially, 19 individuals completed the surv ey. Following a twoweek period, a second solicitation was delivered to th e 19 individuals who had comp leted the survey initially, and 16 completed the survey for a second tim e. Respondents reported that the survey took them less than 15 minutes to complete.

PAGE 73

65 Percent agreement between positive and negative respons es was calculated for the questions presented in section one. Twenty items had a strong agreement (80% or greater), 2 items had a modera te agreement (60%-79%), and 1 item had weak agreement (40%-59%). Per item correlations across th e entire response scale for the same items ranged from .34 to .84. Of these items, two had a strong correlati on (.80-1.0), 11 had a modest correlation (.60-.79), and 9 had a weak correlation (.40-.59). A complete listing of the individual correlations and percent agreement values is presented in Table 1. This list is presented in descending order of percent agreement.

PAGE 74

66 Table 1. Test-Retest Reliability Per-Item Correlations: Perception Variables (Spearman Rank Correlation Coefficients) Questionnaire Item Percent Agreement Correlation I intend to use self-paced computer-mediated instruction as a continuing education tool. 100 .75 Using computers to deliver self-paced continuing education programs is an effective option for professional development. 100 .73 Self-paced continuing education delivered using computers is an effective way for me to learn. 100 .65 If I had access to self-paced continuing education programs delivered using computers it would improve my ability to participate in professional development programs. 100 .56 I have the financial resources that are needed to participate in self-paced continuing education programs delivered using computers. 100 .54 If offered the chance to take a self-paced course that was of interest to me and that was delivered using a computer, I am confident that I would be able to complete it. 100 .52 I have the skills that are needed to participate in self-paced continuing education programs delivered using computers. 100 .45 Individuals who are important to me think using computers for self-paced continuing education is a good idea. 94 .84 I would be a good candidate for self-paced continuing education classes delivered using computers. 94 .78 Individuals who influence my behavior think using selfpaced computer-mediated instruction for continuing education is a good idea. 94 .75 Delivering continuing education programs using self-paced computer-mediated instruction is a good idea. 94 .74 Continued on the next page

PAGE 75

67 Table 1 (Continued) Questionnaire Item Percent Agreement Correlation I am willing to take self-paced continuing education courses delivered using computers. 94 .68 I have the equipment that I would need to participate in selfpaced continuing education programs delivered using computers. 94 .64 I would use self-paced educational programs delivered using computers if I had access to them. 94 .59 During the next 12 months, I will take a self-paced educational program that is delivered using a computer. 88 .84 I have access to self-paced continuing education opportunities delivered using computers. 88 .76 Professional organizations important in my field encourage the use of computers to deliver self-paced continuing education programs. 88 .72 Individuals who I respect encourage the use of self-paced continuing education programs delivered using computers. 81 .69 It would be easy for me to take a self-paced continuing education program delivered using a computer. 81 .52 There are opportunities available for me to continue my education using self-paced programs delivered using computers. 81 .47 My managers and supervisors encourage the use of selfpaced continuing education programs delivered using computer-mediated instruction. 75 .58 My colleagues encourage participation in self-paced continuing education programs delivered using computers. 67 .34 I have the time that is needed to participate in self-paced continuing education programs de livered using computers. 56 .46

PAGE 76

68 Percent agreement between positive and negative respons es was calculated for the barriers measured by the questionnaire. Twenty two items had a strong agreement (80% or higher) and 3 items had a moderate agreem ent (60%-79%). The per item correlations using the entire response scale for these same items ranged from -.10 to 1.0. Of these 25 items, five had a strong correlation (.80-1.0), 10 had a modest correlation (.60-.79), and four had a weak correlation (.40-.59). A complete listing of the individual percent agreement values and correlati ons is presented in Table 2.

PAGE 77

69 Table 2. Test-Retest Reliability Per-Item Correlations: Barriers (Spearman Rank Correlation Coefficients) Questionnaire Item Percent Agreement Correlation Lack Internet skills needed 100 1.0 Lack computer skills needed 100 .38 Not comfortable using computers 94 1.0 No work time release for continuing education 94 .97 Expense 94 .88 Not a good way for me to learn 94 .70 No interest in using computer-mediated learning 94 .44 Lack of discipline to complete self-directed programs 94 .35 The technology is intimidating 94 .15 Continuing education is not important for health education/health promotion 94 -.10 Courses currently available are difficult to use 88 .85 Lack of technical support 88 .76 Lack of immediate feedback during programs 88 .64 Not motivated to continue education 88 .32 Lack of professional rewards for continuing education 81 .75 Lack of access to information about programs 81 .72 Limits professional networking 81 .70 Promotes isolation 81 .67 Lack of relevant topics for programs 81 .62 Continued on the next page

PAGE 78

70 Table 2 (Continued) Questionnaire Item Percent Agreement Correlation No opportunity to apply training 81 .61 Few incentives for continuing education 81 .55 Poor Internet access/connection 81 .41 Lack of time 75 .60 Lack of programs 69 .49 Continuing education is not important in my current position 69 .37 Per item correlations for the remaining questions in section 2 and the general information items ranged from -.07 to 1.0. Of these 37 items, 11 had a strong correlation (.80-1.0), 18 had a modest correlation (.60-.79), and five had a weak correlation (.40-.59). A complete listing of the individual correlations is presented in Table 3.

PAGE 79

71 Table 3. Test-Retest Reliability Per-Item Correlations: Other Items (Spearman Rank Correlation Coefficients) Questionnaire Item Correlation Willing to participate: A program delivered from a CD-ROM / DVD .83 A program located on a computer (such as seen with a kiosk) .76 A program using multiple computer-mediated delivery methods, such as a CD-ROM and the Internet .71 A program offered over the Internet .68 Previous participation: A program offered over the Internet .86 A program delivered from a CD-ROM / DVD .78 A program located on a computer (such as seen with a kiosk) .74 A program using multiple computer-mediated delivery methods, such as a CD-ROM and the Internet .65 Prior to participating in this surv ey, were you aware that you could use self-paced programs delivered using co mputers for continuing education? 1.0 Have you ever tried to locate co ntinuing education programs being delivered using a computer? .78 How many classes or training progr ams have you taken that were delivered using a computer? .59 Have you ever completed a class or training program delivered using a computer for which you (or any of th e participants) received continuing education credit, university credit or some type of certification? .52 Have you ever started a class or training program delivered using a computer that you were unable to finish? .53 Continued on the next page

PAGE 80

72 Table 3 (Continued) Questionnaire Item Correlation Participated in the past: Interactive audio (tel ephone conference) .87 Live satellite conference .86 Computer-mediated (CD-ROM, Internet, computer) program .75 Videoconference .68 Print-based (journal review) .63 In-person conference .59 Videocassette self-study .43 Audiocassette self-study -.07 Interest: Computer-mediated (CD-ROM, Internet, computer) program .72 Print-based (journal review) .70 In-person conference .70 Live satellite conference .70 Interactive audio (tel ephone conference) .70 Videoconference .66 Audiocassette self-study .28 Videocassette self-study .13 Do you have a license or certificati on that requires you to earn continuing education credit? 1.0 What year did you first start using a computer? .85 Continued on the next page

PAGE 81

73 Table 3 (Continued) Questionnaire Item Correlation What year did you first start using the Internet (i.e. e-mail or Web searches)? .65 Do you have access to a computer at home? 1.0 Do you have access to a computer at work? 1.0 How do you connect to the Internet from home? 1.0 How do you connect to the Internet from work? 1.0 If you took a self-paced continuing e ducation course using a computer, from where would you be most likely to do this? .71 The nature of the survey and the re lative newness of computer-mediated instruction may account for some of the vari ation in responses between time one and time two. Interestingly, a few of the individuals in this group commented that they had not thought about using computer-media ted instruction prior to taki ng the survey or that they had been so excited about looking into the oppor tunities presented in the definitions that they were afraid their perceptions had changed between the initial and the follow up tests. Updates were made to the survey instrument based on feedback from this group. The most notable change was in the question re lated to barriers. The response scale was modified by changing the term influence to b arrier, and the items were reordered.

PAGE 82

74 Quantitative Survey Results The original sample group contained 1883 indi viduals. Of these individuals, four were removed because of their involvement in the study. Additionally, 110 individuals had e-mail addresses that were not pub lished in the SOPHE directory and not successfully located using online databases. Messages from 374 individuals were returned as non deliverable, and attempts to update their addresses were unsuccessful. Additionally, 136 individuals who received messages asked to be removed from the study. These factors resulted in a final samp le size of 1259. Data collection using the final sample resulted in 504 survey responses indicating a 40% survey response rate. The number of surveys returned exceeded the 373 minimum sample size calculated based on an effect size of .13 and a power of .80 (Kraemer & Thiemann, 1987). A comparison of the responders and non-res ponders based on their chosen Special Interest Group or Caucus membership associ ated with their SOPHE membership showed only small variations. When looking at caucu s membership, responders were less likely to have selected membership in either the University Facult y or Student and New Professionals caucuses. Representation with in each special interest group was similar between the two groups. Comparison between the groups, including t hose that completed the survey, those with error messages, t hose whose addresses were not found, those asking to be removed, and those who di d not respond, is presented in Table 4.

PAGE 83

75 Table 4. Comparison of Special Intere st Group and Caucus Membership among Response Groups Membership Completed Survey Non Responder Delivery Error Requested Removal Address Not Found Special Interest Group Anthropology and Public Health 3 4 4 2 2 Children, Adolescent and School Health 9 10 10 13 5 Community Health Education 47 46 54 38 57 International and Cross-Cultural Health 4 5 4 6 4 Medical Care and Patient Education 8 9 10 17 7 Social Marketing and Health Communications 12 11 8 11 8 Worksite Health Education 4 3 4 2 8 Caucus University Faculty 18 22 8 15 7 Student and New Professionals 9 12 10 4 9 Note: Values do not add to 100% because res pondents were able to specify one Special Interest Group and/or one Caucus.

PAGE 84

76 Various undeliverable messages related to th e notifications that did not reach their intended recipients were returned to the rese archer from the different servers processing the electronic mail delivery. The majority of e rror messages stated that either the host or users were not recognized. Additional messages that were reported included: rejected for policy reasons, not listed in public name and address book, too ma ny messages received, spam-ware detected, mailbox full/quota violation, and access denied. The multiple invitation format and persistence of participants interested in participating in the study presented opportunities for identifying so me potential reasons for non-response. Although the survey page was pretested and piloted among multiple individuals using different co mputers and different software configurations, a few early responders reported difficulty moving from the definitions page to the survey page. Communication with these indivi duals quickly identified that the button at the bottom of the definitions page did not appear in an early version of the Internet Web browser, Netscape Navigator (version 4.7; Netscape Co mmunication Corporation, Mountain View, CA). A coding update was designed to addr ess this issue; howe ver, administrative updates were being made to the server housi ng the survey and the update was unable to be uploaded until this process was finalized. Of the early responders who reported this problem by sending an electronic message to the researcher (N=10), all but one successfully completed the survey. Unfort unately, the number of participants who encountered this error but did not report it a nd were unable to complete the survey cannot be determined. Additional error messages that were reported included a transfer interrupted error and a runtime error. Partic ipants reporting these errors (N=2) were able

PAGE 85

77 to complete the survey during subsequent attempts. Again, the total number of potential respondents who encountered this error and di d not subsequently complete the survey cannot be determined. Communication with some participants s uggested other potential reasons for nonresponse. Automatic electronic mail proces sing might have impacted the response rates and the characteristics of responders. For exam ple, one individual responded to the final message noting that the earlier messa ges had been sorted into the mass mailing/unrecognized file and that she did not regularly read these messages. Reports from other non-responders suggested that the su rvey was too long (N=4) or that the lack of a mid-point response option prevented th em from being able to respond to the questions appropriately (N=2). Additionally, a message from a potential responder stated that a major project and two weeks of vacation prevented he r from completing the survey in the specified time frame, suggesting that the short time frame fo r data collection (26 days) may have impacted some indivi duals ability to respond. Once the administrative process was comple ted and the survey was taken offline, the text file generated by the survey page was imported into a spreadsheet (Microsoft Excel 2002; Microsoft Corporation, Seattle, WA). The data were inspected to ensure that the missing and non-missing values had been properly recorded and to identify and remove any duplicate entries. The missing valu es and other data entr ies were consistent with appropriately recorded entries, and no duplicate values were identified. Personal identifiers were removed, and the data were then imported into a statistical analysis software package (SAS, version 8.02; SAS Institute Inc., Cary, NC) for analysis.

PAGE 86

78 Input from the entire set of respondent s was inspected using frequencies for categorical data and means for continuous variables. After inspection, the individuals who reported their professiona l role or current employme nt as retired (N=4) were removed from additional analysis. Birth year was recoded to reflect age categories based on respondents age in 2003, and the Total Nu mber of professional organizations was calculated. The analysis of data received from the respondents completing the quantitative questionnaire and who were not retired is organized into six sections including, descriptive statistics, factor analysis, co mposite score development and researcher defined variables, between groups analysis multivariate models, and answering the research questions. Descriptive Statistics Initially, the data were analyzed using fr equencies for categorical data and means for continuous variables. Ov erall, less than 2% of data responses were missing. The general demographic information indicated that 82% (N=402) of respondents were female and 18% (N=91) were male. When looking at age, 13% (N=64) were under age 30, 12% (N=59) between 30 and 34, 15% (N=72) between 35 and 39, 12% (N=57) between 40 and 44, 15% (N=71) between 45 and 49, 16% (N=76) between 50 and 54, 12% (N=59) between 55 and 59, and 6% (N=27) were 60 or older. In addition to personal characteristics, severa l questions related to professional characteristics were included on the questionnaire. These questions assessed Level of Education, Professional Identity, Professional Role, Current Em ployer, and Number of Professional

PAGE 87

79 Organizations. The frequencies and percenta ge of responders associated with each of these survey items can be found in Table 5.

PAGE 88

80 Table 5. Professional Charac teristics of Respondents. Variable Frequency Percent Highest Level of Education (N=499) Less than bachelors degree 1 0 Bachelors degree 35 7 Some graduate school 19 4 Masters degree 248 50 Doctoral candidate 27 5 Doctoral/professional degree 166 33 Other 3 1 Professional Identity (N=499) Community/public health educator 205 41 University/college teaching faculty 109 22 Health education researcher 48 10 University/college administrative faculty 24 5 School health educator 15 3 Patient health educator 13 3 Nurse 11 2 Worksite health educator 10 2 University/college stude nt service provider 2 0 School nurse 1 0 Other 62 12 Continued on the next page

PAGE 89

81 Table 5 (Continued) Variable Frequency Percent Professional Role (N=496) Primarily administrative 98 20 Administrative and service delivery 80 16 Primarily teaching 68 14 Administrative and teaching 68 14 Research 59 12 Primarily service delivery 58 12 Service delivery and teaching 20 4 Not currently working in health field 13 3 Other 32 6 Current Employer (N=497) College/university 200 40 Local health department 46 9 Federal government agency 43 9 Non-profit health education organization 39 8 State health department 37 7 Hospital 34 7 Insurance company/HMO 15 3 Continued on the next page

PAGE 90

82 Table 5 (Continued) Variable Frequency Percent Non-hospital health care facility 14 3 Self employed 12 2 Business, industry or organized labor 7 1 Private research organization 6 1 K-12 school 4 1 Voluntary health agency 4 1 Professional association 4 1 State education department 3 1 Health planning agency 2 0 Private foundation 1 0 Other/not working in health area 26 5 Number of Professional Or ganizations (N=500) 0 28 6 1 295 59 2 107 21 3 or more 70 14

PAGE 91

83 In addition to reporting general professional characteristics, participants were asked to select the professional organizations in which they hold membership. Five percent (N=26) reported not currently being a member of any professi onal organizations. For those who noted professional memberships, 91% (N=457) held membership in Society for Public Health Education, 60% (N=302) in American Public Health Association, 23% (N=116) in American Association for Health Education, 15% (N= 73) in Eta Sigma Gamma, 12% (N=58) in Ameri can School Health Association, 3% (N=15) in Association of State and Territorial Direct ors of Health Promotion and Public Health Education, 2% (N=11) in American Colle ge Health Association, 2% (N=11) in American College of Sports Medicine, and 2% (N=10) in Society of State and Territorial Directors of Health Promotion and Public He alth Education. In th is section, responses may add to more than 100% because partic ipants may be members of more than one organization. Respondents also reported memb ership in organizations not listed, such as American Diabetes Associat ion, American Evaluation Asso ciation, Intern ational Union for Health Promotion and Education, and local and regiona l organizations. When asked to provide information about th eir general computer and Internet use, 96% (N=479) reported having access to a com puter at home, and 99% reported access to a computer at work (N=490). Internet access was available to 94% (N=465) of respondents at home, whereas 99% (N=490) were able to access the Internet from work. Additional information collected from th e questionnaire included when respondents started using a computer and the Internet, how respondents access the Internet from home and from work, and where respondents would be most likely to take a computer-mediated

PAGE 92

84 program. The frequencies and percentage of respondents associated with each of these survey items are available in Table 6.

PAGE 93

85 Table 6. Computer and Internet Use Variable Frequency Percent Year started using computer (N=495) 2001-present 1 0 1998-2000 5 1 1995-1997 26 5 1992-1994 66 13 1989-1991 99 20 1986-1988 128 26 Before 1986 170 34 Year started using Internet (N=495) 2001-present 5 1 1998-2000 39 8 1995-1997 169 34 1992-1994 165 33 1989-1991 73 15 1986-1988 26 5 Before 1986 18 4 Continued on the next page

PAGE 94

86 Table 6 (Continued) Variable Frequency Percent Access to Internet from home (N=496) Dial in/phone line 259 52 Cable/DSL 202 41 No Internet at home 31 6 Other/Not sure 4 1 Access to Internet from work (N=496) Continuous access 465 94 Dial in/phone line 16 3 Other 5 1 Not sure 4 1 No Internet at work 6 1 Location for taking Computer-mediated program (N=497) Both home and work 206 41 Work 138 28 Home 133 27 Other 2 0 Would not participate 18 4

PAGE 95

87 Respondents were asked to report their pe rceptions regarding Attitude, Subjective Norm, and Perceived Behavioral Control rela ted to using compute r-mediated instruction for continuing education. This information was collected on a four-point Likert-type scale ranging from 0, representing strongly di sagree, to 3, representing strongly agree. The items with which respondents reported th e strongest agreement included statements indicating that they had the skills (mean = 2.62, SD = .56) and equipment (mean = 2.45, SD = .58) needed to participate and, that if offered the chan ce, they were confident they would be able to complete a computer-mediated program (mean = 2.44, SD = .62). Items that respondents reported the lowest agreem ent with included statem ents indicating that managers and supervisors (mean = 1.52, SD = .73), or colleagues (mean = 1.58, SD = .64), encouraged the use of self-paced contin uing education programs using computers. A complete listing of the means and standard deviations for items in this section are presented in Table 7. This information is presented in descending level of agreement, with the item reporting the strongest agreement listed first.

PAGE 96

88 Table 7. Means and Standard Deviations of Respondents Perceptions Related to Attitude, Subjective Norm, and Perceived Behavioral Control Toward ComputerMediated Instruction. Question item Mean SD I have the skills that are needed to participate in self-paced continuing education programs deliv ered using computers. (N=497) 2.62 .56 I have the equipment that I would need to particip ate in self-paced continuing education programs deliv ered using computers. (N=497) 2.45 .58 If offered the chance to take a self-p aced course that was of interest to me and that was delivered using a computer, I am confident that I would be able to complete it. (N=492) 2.44 .62 Using computers to deliver self -paced continuing education programs is an effective option for professional development. (N=497) 2.37 .58 Delivering continuing education programs using self-paced computer-mediated instruction is a good idea. (N=494) 2.31 .59 I would be a good candidate for self-paced continuing education classes delivered usi ng computers. (N=497) 2.20 .81 It would be easy for me to take a self-paced continuing education program delivered using a computer. (N=495) 2.17 .70 If I were able to use self-paced continuing education programs delivered using computers it would improve my ability to participate in professional development programs. (N=500) 2.16 .70 I am willing to take self-paced continuing education courses delivered using computers. (N=492) 2.13 .68 Self-paced continuing education de livered using computers is an effective way for me to learn. (N=494) 2.08 .71 I would use self-paced educatio nal programs delivered using computers if they were available to me. (N=488) 2.06 .75 Continued on the next page

PAGE 97

89 Table 7 (Continued) Question item Mean SD Individuals who are impo rtant to me think usin g computers for selfpaced continuing education is a good idea. (N=468) 1.99 .61 Individuals who I respect enc ourage the use of self-paced continuing education programs deliv ered using computers. (N=469) 1.88 .64 I have the financial resources that are needed to participate in selfpaced continuing education programs delivered using computers. (N=485) 1.79 .70 Professional organizations important in my field encourage the use of computers to deliver self-paced continuing education programs. (N=482) 1.79 .69 Individuals who influence my be havior think using self-paced computer-mediated instruction for continuing edu cation is a good idea. (N=459) 1.75 .64 Programs or classes using computers to deliver self-paced continuing are available to me. (N=489) 1.72 .77 There are opportunities available fo r me to continue my education using self-paced programs deliver ed using computers. (N=485) 1.67 .70 I have the time that is needed to participate in self-paced continuing education programs delivered using computers. (N=493) 1.62 .77 My colleagues encourage particip ation in self-paced continuing education programs delivered using computers. (N=460) 1.58 .64 My managers and supervisors en courage the use of self-paced continuing education programs delivered using computer-mediated instruction. (N=462) 1.52 .73 Intention I intend to use self-paced, computer-mediated instruction as a continuing education tool (N=493) 1.81 .74 During the next 12 months, I will take a self-paced educational program that is delivered using a computer. (N=495) 1.50 .79

PAGE 98

90 Participants also were aske d about past experience and future intention related to using self-paced computer-mediated instru ction. When asked about willingness to participate in different type s of computer-mediated instru ction, 86% (N=425) responded favorably to programs delivered from a CD-ROM/DVD, 88% (N=435) responded favorably to Internet programs, 84% (N=413) responded favorably to programs delivered using multiple methods, and 45% (N=217) responded favorably to programs located on a computer, such as those seen at a computer kiosk. Responses in this group totals more than 100% because participants may have re ported a willingness to participate in programs from more than one category. When looking at past experience, 64 % (N=313) reported having previously participated in a program delivered over the Internet and 43% (N=213) had participated in a program delivered using a CD-ROM/ DVD. Twenty-eight percent (N=139) had participated in a program that used mu ltiple delivery methods, whereas only 21% (N=102) had used a program specifically located on a computer. Again, respondents may have participated in more than one type of program. Interestingly, 11% (N=54) reported not being aware that they could use comput er-based, self-paced education programs for continuing education, and 47% (N=232) reported having previ ously tried to locate these types of programs. When asked about rece iving credit for their computer-mediated education efforts, 45% of respondents (N=221) reported receiving credit, whereas 34% (N=171) did not receive credit and 21% (N =103) had not taken a computer delivered course. Fifteen percent of respondents (N =71) reported starting a class or training

PAGE 99

91 program that they were unable to finish, a nd the most common reason for not being able to finish was a lack of time. The questionnaire also asse ssed potential barriers to using computer-mediated instruction. These items were assessed using a four-point Likert -type scale ranging from 0, signifying the item is not a barrier, to 3, i ndicating the item to be a significant barrier. Lack of time (mean = 1.60, SD = 1.10), lack of programs (mean = 1.35, SD = 1.03), and expense associated with taking continuing education programs (mean = 1.26, SD = 1.06), received the highest scores; on the other hand comfort level, skills, and access to both computers and the Internet were reported to be minor barriers. A complete listing of the barriers studied and their associated means and standard deviations are presented in Table 8.

PAGE 100

92 Table 8. Means and Standard Deviations fo r Barriers Associated with Using ComputerMediated Instruction for Continuing Education. Barrier Mean SD Lack of time (N=491) 1.60 1.10 Lack of programs (N=476) 1.35 1.03 Expense associated with taking continuing education programs (N=489) 1.26 1.06 Lack of interaction with f aculty/instructor (N=490) 1.24 1.04 Lack of relevant topics for programs (N=463) 1.21 .98 Lack of access to informati on about programs (N=492) 1.20 .99 Lack of social inte raction (N=490) 1.17 1.09 Lack of work time release for continuing education (N=489) 1.12 1.07 Lack of professional networking during programs (N=492) 1.06 1.00 Lack of importance placed on continuing education in my current position (N=488) .88 1.03 Lack of professional rewards for continuing education (N=487) .83 .97 Lack of incentives for con tinuing education (N=491) .81 .94 Lack of technical support for programs (N=492) .75 .85 Lack of interest in using computer-mediated learning (N=493) .74 1.03 Lack of immediate feedback during programs (N=482) .72 .87 Not a good way for me to learn (N=483) .68 .96 Lack of opportunity to apply training (N=487) .65 .84 Continued on the next page

PAGE 101

93 Table 8 (Continued) Barrier Mean SD Lack of importance placed on c ontinuing education in the health education/health promotion field (N=488) .59 .87 Lack of motivation to con tinue education (N=492) .48 .75 Lack of discipline to complete se lf-directed programs (N=496) .43 .67 Difficulty level of currently available courses (N=461) .32 .64 Internet access/connection (N=495) .26 .60 Current computer skills (N=497) .21 .51 Current level of comfort with using computers (N=497) .19 .51 Current Internet skills (N=495) .17 .48 Computer access (N=492) .14 .46 Current level of comfort with using the Internet (N=494) .14 .45 Some additional barriers reported by responde nts included: lack of sound accompanying text and figures, computer fatigue or too much time already spent using the computer, and a preference for reading from books when involved in programs that are predominantly knowledge-based. When asked about continuing education in general, 94% of respondents (N=466) had participated in reading professional journals, 90% (N=447) had attended in-person conferences, 37% (N=180) had participated in live satellite confer ences, 12% (N=57) in videocassette self-study, 8% (N=37) in audiocassette self-s tudy, 44% (N=218) in

PAGE 102

94 telephone conferences, 34% (N=166) in vi deoconferences, and 53% (N=261) in computer-mediated programs during the past 12 months. Professional development and continuing education opportuni ties were assessed on a Like rt-type scale ranging from 0, signifying strong interest, and 3, representing no interest. Respondents showed a moderate or strong interest in participating in reading professional journals (mean=.59, SD=.87), in-person conferences (mean=.34, SD=.60), and computer-mediated programs (mean=.96, SD=.97). Additional responses related to interest in continuing education opportunities are presented from lo west to highest in Table 9.

PAGE 103

95 Table 9. Means and Standard Deviations of Respondents' Current Interest in Participating in Continuing Education Opportunities Continuing Education Opportunity Mean SD Audiocassette self-study (N=493) 2.27 .88 Videocassette self-study (N=495) 2.05 .92 Telephone conference (N=495) 1.77 .96 Videoconference (N=486) 1.47 .93 Live satellite conference (N=493) 1.30 .90 Computer-mediated program (N=493) .96 .97 Reading professional journals (N=492) .58 .74 In-person conference (N=490) .34 .60 Factor Analysis Responses to the 21-items on perceptions related to the use of computers for continuing education were subjected to an exploratory factor an alysis using squared multiple correlations as prior communality estimates. The principal factor method was used to extract the factors. The scree pl ot of the eigenvalues (Appendix P), and the theoretical framework suggested a three-factor structure, so three f actors were retained and rotated for final interpretation. Bernard (2000) suggests that variables loading at least .60 on a factor represent that factor, but that variab les between .30 and .59 are also worthy of consideration. For this study, items loading at .50 or higher on a given factor

PAGE 104

96 and loading less than .50 on th e remaining factors were cons idered to represent a single factor. Using these criteria, 12 items loaded on factor one, five items loaded on factor two, and two items loaded on factor three. Based on the literature and theory used to develop the survey, factor one was identified as Attitude, factor two as Subjective Norm, and factor three as Perceived Behavior Contro l. A complete listing of the items and the factor loadings for each of the thr ee factors is presented in Table 10.

PAGE 105

97 Table 10. Questionnaire Items and Correspond ing Factor Loadings from the Rotated Factor Pattern Matrix and Fact or Structure Matrix (N=410) Factor Pattern Factor Structure 1 2 3 1 2 3 Questionnaire Item .63* .16 -.23 .69* .43 -.17 If I were able to use self-paced continuing education programs delivered using computers it would improve my ability to participate in profe ssional development programs. .50* -.22 .26 .42 .05 .26 I have the skills that are needed to participate in self-paced continuing education programs delivered using computers. .16 .55* -.03 .42 .62* .05 Individuals who are im portant to me think using computers for self-paced continuing education is a good idea. -.07 .42 .19 .14 .41 .24 Professional organizations important in my field encourage the use of computers to deliver self-paced continuing education programs. .52* .22 -.06 .62* .46 -.01 Using computers to deliver self-paced continuing education programs is an effective option for professional development. .72* .11 -.03 .77* .45 .03 I would be a good candidate for self-paced continuing education classes delivered using computers. .09 .73* .04 .44 .77* .13 Individuals who I re spect encourage the use of self-paced continuing education programs delivered using computers. .70* .13 -.04 .76* .45 .01 Self-paced continuing education delivered using computers is an effective way for me to learn. Continued on the next page

PAGE 106

98 Table 10 (Continued) Factor Pattern Factor Structure 1 2 3 1 2 3 Questionnaire Item -.04 .21 .79* .10 .28 .81* Programs or classes using computers to deliver self-paced continuing education are available to me. .55* -.02 .37 .56* .28 .40 It would be easy for me to take a selfpaced continuing education program delivered using a computer. -.01 .73* .16 .35 .75* .25 My managers and supervisors encourage the use of self-paced continuing education programs delivered using computermediated instruction. .02 .16 .73* .14 .25 .75* There are opportunities available for me to continue my education using self-paced programs delivered using computers. .54* -.02 .03 .53* .24 .06 I have the time that is needed to participate in self-paced continuing education programs de livered using computers. .17 .01 .15 .18 .11 .16 I have the financial resources that are needed to participate in self-paced continuing education programs delivered using computers. .61* .15 -.02 .68* .43 .04 Delivering continuing education programs using self-paced computer-mediated instruction is a good idea. -.06 .73* .10 .30 .72* .18 My colleagues encourage participation in self-paced continuing education programs delivered using computers. Continued on the next page

PAGE 107

99 Table 10 (Continued) Factor Pattern Factor Structure 1 2 3 1 2 3 Questionnaire Item .81* .07 .00 .84* .45 .06 I am willing to take self-paced continuing education courses delivered using computers. .65* -.20 .28 .57* .14 .30 I have the equipment that I would need to participate in self-paced continuing education programs delivered using computers. .05 .75* -.01 .41 .77* .09 Individuals who influence my behavior think using self-paced computer-mediated instruction for continuing education is a good idea. .75* .14 -.10 .81* .49 -.04 I would use self-paced educational programs delivered using computers if I had access to them. .70* -.15 .20 .64* .20 .22 If offered the chance to take a self-paced course that was of interest to me and that was delivered using a computer, I am confident that I would be able to complete it. Values with a significant loading of .50 or greater. The factor pattern loadings represent the unique contribution that each factor makes to the variance of the observed variab les, and the factor structure loadings represent the correla tions between the variables and common factors (Hatcher, 1994). Whereas it is important to review both, as s uggested by Hatcher (1994) the factor pattern loadings were used to make the final determination of items loading on each factor.

PAGE 108

100 Prior to creating subscale scores, scale consistency was assessed by calculating Cronbach coefficient alpha. Internal c onsistency estimates were .90, .85, and .81 for Attitude, Subjective Norm, and Perceived Behavioral Control, respectively. Composite Score Development and Researcher Defined Variables Subscale scores were created for each c onstruct based on the items identified in the factor analysis. Twelve items were iden tified as representing Attitude. Observations with missing values (N= 32) were inspect ed, and items with only one missing value (N=19) were retained. A s ubscale score for observations with all 12 items completed (N=468), or with only one missing value (N = 19), were developed by calculating the average of the available scor es. The Attitude variable (N=487) had a mean of 2.22 and a SD of .46. Five items were identified as representing Subj ective Norm. Observations with missing values (N= 73) were inspect ed, and items with only one missing value (N=31) were retained. Subscale scores for observations with all 5 items completed (N=427), or with only one missing value (N = 31), were developed by calculating the average of the available scores. The Subject ive Norm variable (N =458) had a mean of 1.74 and a SD of .52. Two items were identif ied as representing Perceived Behavioral Control. Subscale scores for observations with both items complete (N=479) were developed by calculating the average of th ese two scores. The Perceived Behavioral Control variable (N=479) had a m ean of 1.69 and a SD of .67. Ajzen (2002b) suggests measuring variable s using multiple questions in different formats to improve the reliability of survey items. Therefore, multiple item formats were developed in this survey to assess Intention, Behavior, and Continuing Education

PAGE 109

101 Behavior. The survey statements used to determine Intention included items 22, 23 and 24a (see Appendix K for a copy of the survey). The internal consistency of these three items was measured using Cronb achs Alpha and found to be .69. To determine the level of Intention each respondent had toward participation in continuing education using computer mediat ed instruction, a two-step process was implemented. First, a perception score was determined by averaging the data from respondents for survey items 22 and 23. This step provided a score, ranging from a minimum of 0 to a maximum of 3, which represented each respondents perception of continuing education using computer-mediate d instruction. The second step in this process was to develop a willingness score from survey item 24a. If respondents reported a willingness to participate in any of the options listed in survey item 24a, they were considered to be willing to participate in continuing education using computer mediated instruction. This procedure provided a score of 0 for all respondents who did not indicate a willingness to participate in any of the options noted in survey item 24a, and a score of 1 for respondents who did indicate a willingness to participate in one or more of the options. The level of Intention was determined by combining the perception score and the willingness score for each respondent. Indivi duals with a willingness score of 0, or a perception score of less than 1, received an a ssigned Intention value of 0, representing No Intention. Respondents who indicated a willingness to par ticipate (i.e., a willingness score of 1), and who had a perception score between 1 and 1.9, were assigned an Intention value of 1, representing Low Inte ntion. Those respondents who indicated a

PAGE 110

102 willingness to participate (i.e., a willingness score of 1) and who had a perception score between 2 and 2.9, were assigned a value of 2, representing a Moderate Intention level. Finally, those respondents who i ndicated a willingness to par ticipate (i.e., a willingness score of 1) and who also had a perception scor e of 3 were assigned an Intention value of 3, representing a High Intention to undertake computer-media ted instruction. The results from this recoding are presented in Table 11. Table 11. Frequency and Percent for R ecoded Variable Representing Intention Variable Frequency Percent Intention (N=488) No intention 45 9 Low intention 228 47 Moderate intention 167 34 High intention 48 10 According to Ajzen (1988), the theory of planned behavior reverts to the theory of reasoned action when a behavior is under th e control of the indi vidual. Therefore, a proxy measure of the respondents behavior was determined based on the respondents previous participation in co mputer-mediated instructional programs, and their future intention to participate in these programs. As such, all respondents identified as having a

PAGE 111

103 moderate or high intention to participate (see Levels of In tention in Table 10), and who reported having previous partic ipation in a computer-media ted program for both survey items 24b and 27, were assigned a Behavior va lue of 1 representing a positive behavior. All remaining non-missing observations we re assigned a Behavior value of 0. The values derived for Behavior are presented in Table 12. Table 12. Frequency and Percent for the Re searcher-Developed Va riable Representing Behavior Variable Frequency Percent Behavior (N=486) No Behavior 311 64 Behavior 175 36 Question 28 was combined with Behavior to determine Continuing Education Behavior. Respondents with a Behavior of 1 and who reported having received credit of some type were coded as 1 for Continui ng Education Behavior. The remaining nonmissing observations were coded as 0. The values for this variable are presented in Table 13.

PAGE 112

104 Table 13. Frequency and Percent for the Re searcher-Developed Va riable Representing Continuing Education Behavior Variable Frequency Percent Continuing Education Behavior (N=482) No Continuing Education Behavior 366 76 Continuing Education Behavior 116 24 Between Groups Comparisons Associations between variables were made by submitting variable pairs for analysis using cross-tabulations with chi-squa re tests and analysis of variance (ANOVA). When looking at the associations related spec ifically to Continui ng Education Behavior, statistically significant differences were se en between responders who had a positive Continuing Education Behavior and those w ho had a negative Continuing Education Behavior for the variables License or Certification ( 2 15.41, df=1, p < .05), Professional Identity ( 2 16.06, df=5, p < .05), Current Employer ( 2 26.24, df=7, p < .05), Attitude (F 1,472 = 69.06; p < .05), Subjective Norm (F 1,446 = 29.12; p < .05), and Perceived Behavioral Control (F 1,463 = 65.28; p < .05). When analyzing the relationship between the independent variables (Intention, Behavi or, and Continuing Education Behavior) and Age and Gender, few statistically significan t associations emerged. The association between Age and the independent variables reached statistical si gnificance only as it related to Attitude. ANOVA revealed that Age had a statistically si gnificant association

PAGE 113

105 with Attitude (F 7, 467 = 2.46; p < .05). Tukeys HSD te st indicated that respondents age 29 and younger and respondents age 35 to 39 had higher Attitude scores than individuals age 60 and over (.38 and .34, respectively, p < .05). The results for Gender were not statistically significance for any associations studied. When looking at the associations be tween License/Certification and the independent variables, statis tically significant findings we re discovered for Attitude, Intention, Behavior, and Continuing Educa tion Behavior. ANOVA revealed that License/Certification had a statistically significant association with Attitude, (F 1, 482 = 16.74; p < .05). Tukeys HSD test indicat ed that respondents with a license or certification had a .18 higher score for Attit ude (p < .05) than respondents who do not hold a license or certification. A chi-square test showed a stat istically significant association between License/Ce rtification and Intention ( 2 18.77, df=3, p < .05), with 51% (N=145) of license or cer tification holders reporting a moderate Intention or higher, whereas only 32% (N=57) of th eir non-licensed counterparts re ported similar Intentions. Significant associations were also noted for License/Certification and Behavior ( 2 9.59, df=1, p < .05) and for License/Certificati on and Continuing Education Behavior ( 2 15.41, df=1, p < .05). In both associations, a gr eater percentage of respondents with a license or certification had positive Behaviors. Participants were asked where they woul d be most likely to take a computermediated program (Take). Responses to this variable showed a sta tistically significant association with both Attitude and Behavi or. ANOVA revealed that Take had a statistically significant association with Attitude (F 2, 464 = 6.57; p < .05). Tukeys HSD

PAGE 114

106 test indicated that individuals most likely to take a course at work had an Attitude score .18 higher than individuals taking a course from home (p < .05) and .14 higher than individuals taking from both home and work (p < .05). The association between Take and Behavior reached statistical significance ( 2 6.20, df=2, p < .05). For respondents reporting they would take a computer-mediated course at home or work, 43% (N=85) had a positive Behavior, whereas 38% of those repor ting they would take the course at work had a positive Behavior, and 29% of those repo rting they would take the course at home had a positive Behavior. Level of Education (Education) had signifi cant associations w ith all independent variables except Continuing E ducation Behavior. ANOVA rev ealed that Education was statistically significantly a ssociated with Attitude (F 4, 479 = 2.73; p < .05), Subjective Norm (F 4,450 = 3.52; p < .05), and Percei ved Behavioral Control (F 4,470 = 2.81; p < .05). Tukeys HSD test showed that the differen ces between both Education and Attitude, and Education and Subjective Norm, were among re spondents with a masters degree and a doctoral degree, with individuals possessi ng a masters degree having a .14 higher Attitude score (p < .05) and a .16 higher Subjec tive Norm score (p < .05). Tukeys HSD test indicated the differences between Educa tion and Perceived Behavioral Control were between respondents who were doctoral candida tes and those who had doctoral degrees. Doctoral candidates had a .42 higher Perceived Behavior al Control score than respondents with a doctoral de gree (p < .05). The associations between Education and Behavior, and Intention and Behavior showed statistical significance ( 2 12.49, df=4, p < .05 and 2 345.14, df=3, p < .05, respectively). A pos itive Behavior was seen in 32% of

PAGE 115

107 respondents holding a bachelors degree, 33% of those with some graduate school, 43% of those with a masters degree, 31% of doctora l candidates, and 26% of those with doctoral or professional degrees. An association be tween Intention and Be havior was expected since the Intention variable was used to derive the Behavior variable. Professional Identity had statistically si gnificant associations with Subjective Norm, Intention, Behavior, and Continuing E ducation Behavior. ANOVA revealed that Professional Identity had a statistically signi ficant association with Subjective Norm (F 9, 396 = 2.74; p < .05). Tukeys HSD test showed school health educators to have a higher Subjective Norm score than both health education researchers (.54, p < .05) and university/college teaching facu lty (.56, p < .05). The associations between Professional Identity and Intention, Behavior, and Continui ng Education Behavior showed statistically significant findings ( 2 22.38, df=5, p < .05; 2 22.62, df=5, p < .05; and 2 16.06, df=5, p < .05, respectively). When looking at Intenti on, university/college ad ministrators had the highest percentage of individuals with a positive Intention (58 %), followed by patient health educators (54%), community/public health educators (49%), school health educators (49%), health education researcher s (33%), and univers ity/college teaching faculty (25%). The association between Pr ofessional Identity and Intention showed that, 50% of university/college administrators, 47% of school health educat ors, 46% of patient health educators, 39% of community/public he alth educators, 29% of health education researchers, and 16% of university/college t eaching faculty had a positive Behavior. For Professional Identity and Continuing Edu cation Behavior, 46% of patient health educators, 39% of university/college admi nistrative faculty, 33% of school health

PAGE 116

108 educators, 24% of community/public health educators, 15% of health education researchers, and 13% of university/college teaching faculty had a positive Continuing Education Behavior. Professional Role had a significant associ ation with Subjective Norm, Perceived Behavioral Control, Intention, and Behavior ANOVA revealed that Professional Role had a statistically significant association with Subjective Norm (F 6,408 = 3.91; p < .05), and Perceived Behavioral Control (F 6,428 = 2.25; p < .05). Tukeys HSD test showed respondents reporting both administrative a nd direct service delivery had a higher Subjective Norm score than respondents repo rting administrative du ties (.20, p < .05) and researchers (.29, p < .05). Also, respondents reporting a primary role of health education service delivery had a higher mean score (.29, p < .05) than re spondents reporting both administrative and teaching responsibilities. The association between Professional Role and Intention was statistically significant ( 2 37.07, df=18, p < .05), w ith direct service providers (61%) and responders with admi nistrative roles (59 %) having the highest percentage of individuals with a positive Intention, and researchers (18%) having the lowest percentage. The association betw een Professional Role and Behavior was statistically significant ( 2 18.52, df=6, p < .05), with 49% of direct service providers and 42% of administrators having a positive Behavior. Finally, Current Employer had statistically significant associations with all independent variables. ANO VA revealed that Current Employer was statistically significantly related to Attitude (F 16, 442 = 2.22; p < .05), Subjective Norm (F 16,414 = 2.87; p < .05), and Perceived Behavioral Control (F 16, 434 = 1.83; p < .05). Tukeys HSD

PAGE 117

109 test showed that respondents working in a hospital setting had higher Subjective Norm scores than respondents working in a college or university setting (.44, p < .05) and respondents working in non-hospital health car e facilities (.73, p < .05). Also, hospital workers had higher Perceived Be havioral Control scores than college or university workers (.47, p < .05), state health departme nt workers (.60, p < .05) and state education department workers (1.45, p < .05). The a ssociations between Current Employer and Intention, Current Employer and Behavior and Current Employer and Continuing Education Behavior all were statistically significant ( 2 34.30, df=7, p < .05; 2 34.85, df=7, p < .05; and 2 26.24, df=7, p < .05, respectively). The highest percentages of respondents with positive Inten tion were seen in hospital em ployees (77%) and insurance companies/managed care organizations (69%), whereas the lowest in local health departments (38%) and universities/colleges (33%). The highest percentages of respondents with positive Behavi or were seen working in hospitals (68%) and federal government agencies (56%), and the lowest in colleges or univers ities (26%) and local health departments (27%). Similar results we re seen for Continuing Education Behavior. The patterns of statistically significant bi variate associations highlighted above and summaries of the cross-tabulations a nd chi-square test results for Continuing Education Behavior are summar ized in Table 15. A summar y of the cross-tabulations and chi-square test results for Beha vior can be found in Appendix Q.

PAGE 118

110 Table 14. Summary of Significant Bivariate Associations: Responde nt Characteristics by Independent Variables Respondent Characteristic Independent Variables Attitude Subjective Norm Perceived Behavioral Control Intention Behavior Continuing Education Behavior Age NS NS NS NS NS Gender NS NS NS NS NS NS License/ Certification NS NS ** ** ** Location for taking CME NS NS NS ** NS Highest Level of Education * ** ** NS Professional Identity NS NS ** ** ** Professional Role NS * ** ** NS Current Employer * ** ** ** Number of Professional Organizations NS NS NS NS NS p < .05 for ANOVA ** p < .05 for Chi-Square

PAGE 119

111 Table 15. Respondent Characteristic s by Continuing Education Behavior Continuing Education Behavior Overall Behavior Absent Behavior Present 2 Age 0.18 29 and Under 13% (N=62) 9% (N=44) 4% (N=18) 30-34 13% (N=59) 10% (N=47) 3% (N=12) 35-39 15% (N=71) 12% (N=57) 3% (N=14) 40-44 12% (N=55) 9% (N=43) 3% (N=12) 45-49 14% (N=66) 9% (N=44) 5% (N=22) 50-54 16% (N=74) 13% (N=62) 3% (N=12) 55-59 12% (N=56) 8% (N=38) 4% (N=18) 60 and over 6% (N=27) 5% (N=22) 1% (N=5) Gender 0.60 Male 19% (N=90) 15% (N=71) 4% (N=19) Female 81% (N=388) 61% (N=291) 20% (N=97) License/ Certification 15.41* Yes 63% (N=301) 44% (N=211) 19% (N=90) No 37% (N=178) 32% (N=153) 5% (N=25) Continued on next page

PAGE 120

112 Table 15 (continued) Continuing Education Behavior Overall Behavior Absent Behavior Present 2 Location for taking CME 3.83 Home 28% (N=130) 21% (N=99) 7% (N=31) Work 29% (N=135) 23% (N=108) 6% (N=27) Both Home and work 43% (N=198) 30% (N=140) 13% (N=58) Highest Level of Education 6.80 Bachelors 7% (N=34) 5% (N=26) 2% (N=8) Some graduate school 4% (N=18) 3% (N=14) 1% (N=4) Masters degree 50% (N=238) 36% (N=170) 14% (N=68) Doctoral candidate 5% (N= 26) 4% (N=20) 1% (N=6) Doctorate or professional degree 34% (N=162) 28% (N=134) 6% (N=28) Continued on next page

PAGE 121

113 Table 15 (Continued) Continuing Education Behavior Overall Behavior Absent Behavior Present 2 Professional Identity 16.06* Community/public health educator 50% (N=201) 38% (N=153) 12% (N=48) School health educator 4% (N=15) 2% (N=10) 1% (N=5) Patient educator 3% (N=13) 2% (N=7) 1% (N=6) Health education researcher 12% (N=47) 10% (N=40) 2% (N=7) Univ/College teaching faculty 26% (N=106) 23% (N=92) 3% (N=14) Univ/college administrative faculty 6% (N=23) 3% (N=14) 2% (N=9) Continued on next page

PAGE 122

114 Table 15 (continued) Continuing Education Behavior Overall Behavior Absent Behavior Present 2 Professional Role 9.63 Administrative 22% (N=94) 15% (N=67) 6% (N=27) Service delivery 13% (N=55) 8% (N=37) 4% (N=18) Teaching 15% (N=66) 12% (N=51) 3% (N=15) Administrative and service delivery 18% (N=80) 14% (N=60) 5% (N=20) Administrative and teaching 15% (N=67) 12% (N=52) 3% (N=15) Service delivery and teaching 4% (N=18) 3% (N=15) 1% (N=3) Research 13% (N=56) 11% (N=50) 1% (N=6) Continued on next page

PAGE 123

115 Table 15 (continued) Continuing Education Behavior Overall Behavior Absent Behavior Present 2 Current Employer 26.24* College or University 47% (N=194) 39% (N=160) 8% (N=34) Hospital 7% (N=31) 3% (N=14) 4% (N=17) Non-hospital health care facility 3% (N=14) 2% (N=10) 1% (N=4) Insurance company/MCO 3% (N=13) 2% (N=7) 1% (N=6) Local health department 11% (N=45) 9% (N=36) 2% (N=9) State health department 9% (N=36) 7% (N=28) 2% (N=8) Federal government agency 10% (N=42) 7% (N=28) 3% (N=14) Non profit health education organization 9% (N=39) 7% (N=29) 2% (N=10) Note Due to rounding, total percent may not eq ual 100. Response items with expected counts less than 5 were removed. p<.05.

PAGE 124

116 Further examination of associations involving Attitude, Subjective Norm, Perceived Behavioral Control, Behavior, and Continuing Education Behavior showed statistically significant findings for Attitude and Behavior (F 1, 475 = 96.67; p < .05), Attitude and Continuing Education Behavior (F 1, 472 = 69.06; p < .05), Subjective Norm and Behavior (F 1, 449 = 58.64; p < .05), Subjective Norm and Continuing Education Behavior (F 1, 446 = 29.12; p < .05), Perceived Behavi oral Control and Behavior (F 1, 467 = 70.65; p < .05), and Perceived Behavioral Control and Continuing Education Behavior (F 1, 463 = 65.28; p < .05). For statistically si gnificant associations, Tukeys HSD test showed that respondents with a positive Behavi or (N=173) had a mean score for Attitude of 2.48, whereas respondents with a negative Behavior (N=304) had a mean Attitude score of 2.08 (p < .05). Also, respondents with a positive Continuing Education Behavior (N=116) had a mean score for Attitude of 2.52, whereas respondents with a negative Continuing Education Behavior (N=358) had a mean Attitude score of 2.13 (p < .05). Tukeys HSD test performed on Subjective Norm showed positive Behavior respondents (N=167) to have a mean Subjective Norm score 0.36 higher (p < .05) than negative Behavior respondents (N=284). Additionall y, positive Continuing Education Behavior respondents (N=114) had a mean Subjectiv e Norm score 0.26 higher (p < .05) than negative Continuing Education Behavior res pondents (N=351). Finally, Tukeys HSD test for Perceived Behavioral Control showed positive Behavior respondents (N=172) and positive Continuing Education Behavior respond ents (N=114) had higher mean Perceived Behavioral Control scores (0.51, and 0.56, respectively) than their negative Behavior counterparts (N= 297, and N=351, respectively).

PAGE 125

117 Multivariate Models The dependent variable (Continuing E ducation Behavior) for the study was measured dichotomously; therefore, logistic regression was employed for multivariate analysis. Initially, each i ndependent variable (Attitude Subjective Norm, Perceived Behavioral Control, and Inten tion) was entered into a logi stic model independently, and for all variables, statistically significant a ssociations were identif ied. Next, a control model was analyzed. This control model included the variables for Gender, Age, License/Certification, and Education Level, and it was found to be statistically significantly associated with Continuing Education Behavi or (Wald Chi-Square = 16.53, p < .05). Finally, Box-Tidwell transformations for continuous variab les and tolerances for all variables were assessed. Natural log terms entered into the logistic equations were not statistically significant; therefore, the assumption of linearity between the logits for the independent variables and the dependent variable was not violated. Also, the tolerance values ranged from .57 to .86 signifying that assumptions related to multicollinearity had not been violated. The research questions were answered us ing three logistic regression models. Model 1, the full model, was represented by th e equation: behavior = control variables + attitude + subjective norm + pe rceived behavioral control + intention. Over all, this model was statistically signi ficant (Wald Chi-Square = 86.12, p < .05), with one control variable (License/Certification) and two i ndependent variables (Perceived Behavioral Control and Intention) having individually stat istically significant odds ratios. Attitude and Subjective Norm were not statistically associated with Continuing Education

PAGE 126

118 Behavior in this model. See Table 16 fo r the model odds ratios and 95% confidence intervals. Table 16. The Effect of A ttitude, Subjective Norm, Perceived Behavioral Control, Intention and the Control Variables on C ontinuing Education Behavior (Model 1) Variable OR 95% CI Significant Control Variables Gender Male 1.00 -Female .69 (.29, 1.61) Age (Birth) 1.00 (.97, 1.03) License/Certification Yes 1.00 -No .46 (.22, .96) Degree Bachelors 1.00 Masters .92 (.35, 2.43) Ph.D. 1.05 (.63, 1.33) Attitude 1.53 (.63, 1.33) Subjective Norm .62 (.29, 1.33) Perceived Behavioral Control 2.24 (1.35, 3.73) Intention 12.80 (6.61, 24.78) Note: indicates p < .05.

PAGE 127

119 Model 2 was represented by the equation: be havior = control variables + attitude + subjective norm + perceived beha vioral control. Overall, this model was statistically significant (Wald Chi-Square = 71.77, p < .05), with one control variable (License/Certification) and tw o independent variables (Attit ude and Perceived Behavioral Control) having individually significant odds ratios. Subjective Norm was not statistically associated with Continuing Educa tion Behavior in this model. See Table 17 for the model odds ratios and 95% confidence intervals.

PAGE 128

120 Table 17. The Effect of Attitu de, Subjective Norm, Perceive d Behavioral Control, and the Control Variables on Continui ng Education Behavior (Model 2) Variable OR 95% CI Significant Control Variables Gender Male 1.00 -Female .76 (.37, 1.57) Age (Birth) .99 (.96, 1.01) License/Certification Yes 1.00 -No .46 (.25, .84) Degree Bachelors 1.00 Masters 1.03 (.44, 2.41) Ph.D. .84 (.32, 2.20) Attitude 6.00 (2.92, 12.32) Subjective Norm 1.24 (.67, 2.26) Perceived Behavioral Control 3.06 (2.03, 4.62) Note: indicates p < .05.

PAGE 129

121 Model 3 was the same as the equation in Model 2; however, its application was applied to a subset of the total dataset where intention was positive. Overall, this model was statistically significant (Wald Chi-Squa re = 21.23, p < .05), with one control variable (License/Certification) and one independent variable (Perceived Behavioral Control) having statistically significan t odds ratios. Attitude a nd Subjective Norm were not statistically associated with Continuing Educa tion Behavior in this model. See Table 18 for the model odds ratios and 95% confidence intervals.

PAGE 130

122 Table 18. The Effect of Attitu de, Subjective Norm, Perceive d Behavioral Control, and the Control Variables on Con tinuing Education Behavior for Respondents with a Positive Intention (Model 3) Variable OR 95% CI Significant Control Variables Gender Male 1.00 -Female .80 (.33, 1.95) Age (Birth) .99 (.96, 1.02) License/Certification Yes 1.00 -No .46 (.22, .95) Degree Bachelors 1.00 Masters 1.18 (.44, 3.19) Ph.D. 1.23 (.38, 3.96) Attitude 1.76 (.69, 4.47) Subjective Norm .54 (.24, 1.20) Perceived Behavioral Control 2.78 (1.66, 4.64) Note: indicates p < .05.

PAGE 131

123 Likelihood ratio tests were used to dete rmine the relative importance of each independent variable in both Model 1 and Model 2. In Model 1, Intention, Perceived Behavioral Control, and Subjective Norm were found to be statistically significant contributors to the predictive power of the m odel. The differences between the -2 log likelihoods were 95.21, 13.27, and 10.53, respectivel y. In Model 2, Perceived Behavioral Control, Attitude, and Subjective Norm were all found to have predictive power. The differences between the -2 log like lihoods were 37.76, 27.22, and 13.87, respectively. The first study question was: What is the association be tween health educators perceived behavioral control related to us ing computer-mediated continuing education programs and their behavior related to computer-mediated education? Model 2 was used to test the hypothesis that an associati on exists between perceived behavioral control and behavior related to computer-mediated education. The 95% confidence interval around the odds ratio for Perceived Behavior al Control (2.03, 4.62) did not include the value 1.0, so Perceived Behavioral Control was considered to be associated with Continuing Education Behavior (odds rati o = 3.06, p < .05). Moreover, individuals reporting positive levels of Perceived Behavioral Control were three times more likely to have a positive Behavior. The second study question was: What is the association between health educators attitudes related to using computer-mediated continuing education programs and their behavior related to computer-mediated education? Model 2 was used to test the hypothesis that an association exists be tween attitudes and behavior related to computer-mediated education. The 95% conf idence interval around the odds ratio for

PAGE 132

124 Attitude (2.92, 12.32) did not include the value 1.0, so Attitude was considered to be associated with Continuing Education Beha vior (odds ratio = 6.00, p < .05). Moreover, individuals reporting a positive Attitude were six times more likely to have a positive Behavior. The third study question was: What is the association be tween health educators subjective norms related to using computer -mediated continuing education programs and their behavior related to computer-mediated education? Model 2 was used to test the hypothesis that an association exists between Subjective No rm and Continuing Education Behavior related to computer-mediated edu cation. The 95% confidence interval around the odds ratio for Subjective Norm (.67, 2.26) included the value 1.0, so Subjective Norm was not considered to be stat istically associated with Continuing Education Behavior. Since the likelihood ratio tests determined Subjective Norm to be a statistically significant contributor to the predictive power of th e logistic model, additional investigation related to Subject ive Norm was undertaken usi ng logistic analysis modeling of Subjective Norm, Subjective Norm and A ttitude, and Subjective Norm and Perceived Behavioral Control, and include d all control variables. A statistically significant odds ratio was identified in the relationship betw een Subjective Norms and Behavior. This odds ratio was reduced in magnitude but remained statistically significant when Perceived Behavioral Control was added to the model and no longer continued to be statistically significant when Attitude was added to the model. This pattern indicates that Subjective Norm is fully mediated thr ough both Perceived Behavioral Control and Attitude, with the strongest mediator being Attitude.

PAGE 133

125 The fourth study question was: Do health educators inte ntions to use computermediated continuing education programs me diate the association between perceived behavioral control, attitudes, and subjective norms and their behavior related to computer-mediated education? This mediating effect was tested by making comparisons between the values associated with the pr edictor variables in Model 1 and Model 2. Statistically significant odds ratios (deter mined based on the 95% confidence interval around the odds ratio) identified in Model 2, but found not to be statistically significant in Model 1, were considered to be completely mediated by intentions. Statistically significant odds ratios identif ied in Model 2 that decrease d, but continued to remain statistically significant, were considered to be partially mediated by intentions. Starting with Perceived Beha vioral Control, the confidence limits for the odds ratio estimates in both Model 2 and Model 1 did not include the value 1.0, so Perceived Behavioral Control was determined to have a statistically signif icant effect in both models. A comparison of the odds ratio in Mo del 2 (3.06) with the odds ratio in Model 1 (2.24) suggests that Perceived Behavioral Control is partia lly mediated by Intention. When looking at Attitude, the confidence lim its for the odds ratio estimates in Model 2 were statistically significant; however, those in Model 1 were not. Therefore, Attitude was considered to be fully mediated by Inte ntion. Finally, the confidence limits for the odds ratio estimates for Subjective Norm were not statistically signifi cant in either Model 1 or Model 2. In addition to its mediating e ffects, the 95% confiden ce interval around the odds ratio for Intention (6.61, 24.78) showed a direct association between Intention and Continuing Education Behavior (odds ratio = 12.80, p < .05). Intention was the largest,

PAGE 134

126 direct predictor of Behavior, with individuals having positiv e Intentions nearly 13 times more likely to have a positive Behavior than individuals having negative behaviors. The fifth study question was: For individuals with a positive intention to use computer-mediated instruction, what characte ristics studied help differentiate between those who have previously used this learning medium and those who have not? Model 3 was used to test this hypot hesis by applying Model 2 to a subset of individuals who expressed a positive Intention ( > 2) toward using computer-mediated instruction (N=215). Among respondents with a positive Intention, Perceived Behavioral Control was the only independent variable to be statistically significant, and License/Certification was the only control variable to show a statis tically significant effect in this model. Further examination of the subset of respondents who had a positive Intention toward computer-mediated instruction (N=215) revealed characteristics that differentiated between individua ls who had previously partic ipated in computer-mediated continuing education from those who had not. Chi-square tests were used to identify associations between the study variables and Continuing Education Behavior. For this group, none of the respondent characteristic variables were statistically associated with Continuing Education Behavior. Chi-squa re tests between Barriers and Continuing Education Behavior were statistically significant for Lack of Programs ( 2 12.36, df=3, p=.01); Lack of Relevant Topics for Programs ( 2 7.96, df=3, p=.05); and Lack of Technical Support for Programs ( 2 10.35, df=3, p=.02).

PAGE 135

127 Chapter Five: Discussion The purpose of this study was to investigate the relationship between health educators perceptions toward computer-media ted instruction in cont inuing education and their use of this continuing education practice. Although th e term health educator has been used throughout the previous chapters, this term was used with the operational definition of individuals who were lis ted in the 2002-2003 SOPHE membership directory. Individuals from diverse backgrounds who study health behavior, health promotion, health education, and other rela ted areas hold membership in the SOPHE organization. Additionally, SOPH E publishes two premier journals, Health Education and Behavior and Health Promotion Practice As such, for the purpose of the discussion, the terms health education and health promotion professionals will be used to more accurately reflect the diversity within the SOPHE membership. Five-hundred and four health education and health promotion professionals responde d to a request to complete an online questionnaire developed by the resear cher to address this issue. As suggested by Ajzen (2002b) all questions re lated to the dependent and independent variables were specific to continuing educa tion using computer-mediated instruction. Logistic regression was used to inves tigate the associations involvi ng Attitudes, Subjective Norm, Perceived Behavioral Control, Intention, a nd Continuing Education Behavior. The four

PAGE 136

128 sections in this chapter incl ude a summary of the research findings, study considerations, practical implications, and futu re directions. Summary Hypotheses related to five research questi ons were tested. Each research question is listed, followed by a summary of the findi ngs. For each question, the results presented controlled for age, gender, license or certification, and education level in multivariate models. W hat is the association between health educators perceived behavioral control related to using computer-mediated conti nuing education programs and their behavior related to computer-mediated education? Health educators levels of perceived behavioral control were associated with their behavior related to using computermediated instruction for continuing educa tion. Compared to individuals with low perceived behavioral control sc ores, those with high scores were three times more likely to have participated previous ly in computer-mediated instru ction and to be willing to repeat the behavior in the future. These da ta suggest that individuals with a positive feeling of control over their ability to use computer-mediate d instruction for continuing education are more likely to participate in con tinuing education through this mode of instruction. What is the association between health educators attitudes related to using computer-mediated continuing education programs and their behavior related to computer-mediated education? Health educators attitudes were associated with their behavior related to using computer-media ted instruction for continuing education.

PAGE 137

129 Compared to individuals with negative attitudes, those with positive attitudes toward computer-mediated continuing education progr ams were six times more likely to have participated previously in computer-mediated instruction and to be willing to repeat the behavior in the future. These data suggest th at individuals with a positive attitude toward using computer-mediated instruction for continuing education are more likely to participate in continuing educati on using this mode of instruction. What is the association between health educators subjective norms related to using computer-mediated continuing educa tion programs and their behavior related to computer-mediated education? Health educators subj ective norm was not directly associated with their behavior related to using computer-media ted instruction for continuing education. Although, health educators interactions with other individuals and with professional organizations does not have a unique effect on continuing education behavior, it may influence thei r attitudes toward participa ting in continuing education using computer-mediated instruction as well as their perception of perceived behavioral control. Subjective Norm is also a significan t contributor to the pr edictive power of the model. As such, understanding the social interactions related to computer-mediated instruction remains important to consider. Do health educators intentions to us e computer-mediated continuing education programs mediate the associati on between perceived behaviora l control, attitudes, and subjective norms and their behavior related to computer-mediated education? The association between health e ducators attitudes toward co mputer-mediated instruction and their behavior related to th e practice was mediated by their intention to participate.

PAGE 138

130 Intention also partially mediated the associa tion between health educators feelings of perceived behavioral control and their beha vior related to using computer-mediated instruction for continuing education. Thes e data suggest that for health educators, perceptions related to attitude toward c ontinuing education using computer-mediated instruction influences their intentions to use this mode of inst ruction for continuing education. This influence impacts behavior indirectly, but attitudes related to continuing education using computer-mediated instruction do not have a direct effect on actual use of this mode of instructio n for continuing education. Conversely, health educators perceived behavioral control had both direct and indirect effects on computer-mediated continuing education behavior. Perceived beha vior control influen ces intention to use computer-mediated instruction for continuing ed ucation which, in turn, affects behavior. In addition, perceived behavioral control has a non-mediated, direct effect on behavior. Overall, intention to participate in com puter-mediated continuing education programs was the strongest predictor of the behavior with individuals having a positive intention being almost 13 times more likely to participat e in continuing education by this mode of instruction. For individuals with a positive intention to use com puter-mediated instruction, what characteristics studied help different iate between those who have previously used this learning medium and those who have not? When studying the subset of individuals with a positive intention toward computer-m ediated instruction for continuing education, health educators perceived behavioral control was found to be associated with their behavior. Compared to individuals in this group who reported low levels of perceived

PAGE 139

131 behavioral control, those with high perceived behavioral cont rol levels were nearly three times more likely to participate in comput er-mediated continuing education programs. These data suggest that for individuals w ho have a positive intention to engage in computer-mediated continuing education program s, those who have positive feelings of perceived behavioral control are more likel y to participate in continuing education through this mode of instruct ion than persons having nega tive feelings of perceived behavioral control. Additional characterist ics distinguishing between users and non-users of computer-mediated continuing education among health educators with a positive intention included the presence of a license or certification, a per ception of a lack of programs or a lack of relevant topics for programs, and the availability of technical support for computer-mediated programs. Overall, the results of this study suggest th at the theory of planned behavior is a good fit for studying computer-mediated inst ruction for continuing education among health educators and health promotion professi onals. A graphical re presentation of this relationship is depicted in the following figure.

PAGE 140

132 Figure 1. The Relationship of the Theory of Planned Behavior Constructs in this Study Subjective Norms for Behavior Attitude Toward Behavior Perceived Behavioral Control for Behavior Intention Usage/Behavior As expected, based on discussions related to this theory (Ajzen, 1988, 2001, 2002b; Davis, et al., 1989; Mathieson, 1991; Taylor & Todd, 1995a), Intention was the strongest predictor of Continuing Education Behavior. Moreover, the effects of Attitude on Continuing Education Behavior were fully mediated by Intention, and Perceived Behavioral Control had both a direct effect and an indirect effect (mediated through Intention) on Continuing Education Behavior. Although Subjective Norm was not found to have a direct effect on Intention, it contributed significantly to the predictive power of the model. The Subjective Norm construct as it relates to the theory of planned behavior model fit has been shown by some researchers investigating computer use to influence behavioral intention (Taylor & Todd, 1995a), whereas other researchers have not found

PAGE 141

133 this relationship (Davis, et al., 1989; Mathie son, 1991). Reasons suggested for previous non-significant findings include weakness in the psychometric pr operties of the scale, the individual nature of the computer-mediated act ivity, a self-report of a behavior instead of a direct observation, and the pres ence of a true organizational setting (Davis, et al., 1989; Taylor & Todd, 1995a). In this study, the psychometric properties for the subjective norm construct were not as strong as the othe r study constructs. The review panel gave lower average scores to the items combined fo r this construct, and two of the items had only moderate percent agreement scores for test-retest reliability. Additionally, the behavior studied was an indivi dual rather than an organi zational behavior, it was selfreported rather than directly observed, and it did not occur in a si ngle organization. One additional consideration for subjective norm not having a statically significant direct effect on intention or behavi or is to consider the diffu sion process. Rogers (1995) suggests that over time, different factors a ffect the decision to adopt an innovation. When considering the diffusion of compute r-mediated continuing education programs for continuing education, subjective norm may not currently have a direct effect, but the contribution of this relationship may evolve over time. When looking at computer equipment a nd skills, although all respondents were required to have access to a computer and the Internet for participat ion in the study, they also reported the strongest agreement with statements related to confidence using computer-mediated continuing education programs and with having the skills and equipment needed to complete computer-med iated programs. Interestingly, these items were originally constructed by the researcher to measure Perceived Behavioral Control;

PAGE 142

134 however, results from the factor analysis load ed these items on the Attitude factor. The loading of these items on the Attitude factor is supported by the decomposed theory of planned behavior model tested by Taylor and Todd (1995a). In this model, the Attitude variable is further broken down to include Ease of Use, Perceived Usefulness, and Compatibility. These variables could be us ed to describe the items discussed above. Practical Implications This research identified areas that might need to be addressed if health education and health promotion professionals are to be persuaded to use computer-mediated instruction for continuing education. These results can be applied to the development and implementation of programs, as well as the marketing of ex isting opportunities. Implications from this project specifica lly target individuals who have a positive intention toward computer-mediated instructi on for continuing education but have never participated in a computer-mediated program. Measuring on a scale from 0, signifying no barrier, to 3, signifying a significant barrier, the highest mean score for a barrier was 1.6. Although this score suggests that no barrier was a strong barrier, the ten top barriers reporte d by the entire group of respondents included a perceived lack of time, a lack of programs, expense associated with programs, lack of instructor interacti on, lack of relevant program topics, lack of access to information about programs, lack of social interaction, lack of release time from formal work or employment, lack of prof essional networking, and lack of importance placed on continuing education in current posit ion. Noting the barriers that distinguish between positive and negative continuing educ ation behavior among individuals with a

PAGE 143

135 positive intention to participate in compute r-mediated continuing education can provide valuable insight. These barriers included a lack of programs, a lack of relevant program topics, and a lack of technical support. Of these three barri ers, a lack of programs and a lack of relevant program topics both are part of the general barrier list and are important to consider. Interestingly, a lthough a lack of technical support was not one of the top ten barriers, it was a barrier that distinguished between individuals with positive and negative behaviors. As such, this barrier may play an important, unique role in moving health educators from intention to action. The results suggest that for health edu cation and health promotion professionals to engage in computer-mediated conti nuing education programs, more programs, especially ones that address topics relevant to their current functioning, need to be created and made readily available. Also, ensuring th at appropriate technica l support is available to assist participants, and informing potential participants of the availability of this technical assistance, may encourage more health educators and health promotion professionals to follow through on their intentions to participate in computer-mediated programs. Interestingly, and not surprisingly, lack of time was listed as the leading barrier to participation in computer-mediated continuing education among respondents. Unfortunately, technology does not add hours to the day. Instead, it provides the flexibility to squeeze educational tasks into al ready tight personal schedules. Jacqueline Woods, Executive Director of the American Association of University Women (AAUW) comments, We need to deal with the time bind that all parents and ol der students face if

PAGE 144

136 we want to make the rhetoric of lifelong learning for the information economy a reality (AAUW, 2001, paragraph 3). As su ch, in addition to developing learning opportunities, advocating for or ganization culture changes th at allow workers to work smarter, and not longer, is imperative. To further expend on this issue, lack of time was the leading barrier for those individuals who were successful in finding time to respond to the survey invitation. Common sense would s uggest that lack of time may have been even more of an issue for individuals w ho were not successful in finding time to complete the survey. The results of this survey suggest that gender does not ha ve a statistically significant impact on computer-mediated contin uing education behavior. However, 82% of the respondents were female, so a social significance may be pres ent. A report from AAUW suggests that sixty percent of online learners are over 25 and female (AAUW, 2001), with working mothers who are interested in continuing their education giving up personal recreation time and sleep to be able to take online classes. These same individuals report numerous bene fits to continuing their e ducation; however, they also express anxiety about balancing the dema nds of work, family, and school (AAUW, 2001). Again, time becomes an important issue to consider. Additionally, identifying and implementing practices to help maintain a positive mental health are key in helping individuals cope with ad ded responsibilities.

PAGE 145

137 Study Considerations Limitations One limitation of the study was the us e of a proxy measure for Continuing Education Behavior. Although the derivati on of Continuing Education Behavior was determined based on the theory of planned behavior and the theory of reasoned action (Ajzen, 1988, 2001, 2002b; Ajzen & Fishbein, 1980) a stronger measure of Continuing Education Behavior would be one in which actua l behavior was assessed at a future point in time. The number of surveys returned exce eded the 373 minimum sample size recommended in the power calculation, with a response rate of 40%. This response rate is consistent with the findings of some on line surveys (Fyfe, Leonard, Gelmi, Tassell, & Strack, 2001; McDonald & Adam, 2003), highe r than achieved in others (Crawford, Couper, & Lamias, 2001; McDonald & Adam; Ranchhod, A. & Zhou, F., 2001; Sax, Gilmartin, & Bryant, 2003), lower than achie ved in a few studies (McDonald & Adam, 2003), and less than anticipated using the multiple follow-up delivery method (Dillman, 2000). Interestingly, the use of electronic ma il sorting strategies by both individuals and organizations may have decreased the number of electronic mail messages that actually reached the intended recipient, a fact which if true would make th e 40% response rate a conservative estimate. Participants in the pilot testing suggested that if the survey came from a recognized individual or organization they may be more likely to complete it.

PAGE 146

138 Unfortunately, SOPHE was not successfully re cruited to endorse the survey. Such an endorsement may have helped improve the response rate. Whereas agreement on the content validity as assessed by a review panel indicated that the survey items measured th e identified constructs, the agreement was not strong. Additionally, whereas the test-retest reliability scores as measured by percent agreement were adequate, the variability seen in the per-item correlations was greater. This fact suggests that respondents continue d to feel either positively or negatively toward an item; however, the strength of their response varied. This point may be interpreted to mean that the items measured are inherently unstable or the items could be articulated better in writing. Generalization of the study results is limited. A comparison of responders and non-responders from the study population suggested little variati on between the two groups in terms of their selection of a special interest group. Thus, generalization to the study population may be appropriate. Experts disagree over the extent to which SOPHEs membership represents the range of health promotion and education practitioners; whereas some see SOPHE as having the widest range of practitioner venues, other experts view SOPHE as so mewhat unique, even elite among health education professional groups. Thus, one must be cautions in concluding that similar results would be obtained if the entire univers e of health educators had been surveyed. The results of this study may be biased to ward health educators with more formal or extensive professional prep aration, and toward individu als who have access to the Internet. Individuals who are eligible for membership a nd who choose to be a member of

PAGE 147

139 SOPHE often have advanced degrees. When considering health ed ucators perceptions and practices related to computer-media ted instruction in continuing education, individuals with advanced de grees may be different from individuals with bachelors degrees. Also, within this group, only indivi duals with Internet access were able to respond. Although individuals who do not ha ve Internet access may be unlikely to participate in computer-mediated instructi on, the perceptions and practices of health educators who have Internet access may not be representative of those who do not have access. Respondents were subjected to a forced -response scale. The qualitative interviews suggested that most responders were likely to select a midpoint response, but when prompted to think more in-depth a bout the question, they provided a non-midpoint response. Bernard (2000) suggests that ther e is no best format wh en talking about odd and even number of response categories. He writes further to say that whereas 10% of respondents probably do fall in a midpoint res ponse category, 30% of respondents select this option if available (Bernard, 2000). Using a middle point also can pose difficulties for analysis and interpretati on of results (Pedhazur & Schmelkin, 1991). Therefore, the midpoint response category was removed from the response scale. Whereas this action forced respondents to think about the ques tion, two individuals not ified the researcher that they selected not to complete the surv ey because they refused to be forced into answering questions to which th ey did not have an answer. Three additional respondents reported that their missing data were due to their not being able to answer the question,

PAGE 148

140 and one respondent reported that they did not think that their responses accurately reflected their perceptions because they re ally felt neutrally about the item. One final area of limitations includes th e response categories for some of the demographic items. Whereas the items measuring Professional Role, Professional Identity, and Current Employer may provide some interesting insi ght, caution should be used in their interpretation as the response categories were neither mutually exclusive nor exhaustive. Health educators could benefit from continuing to refine these items and to construct a better understanding of health education practice. Strengths The construction of the survey relied on ex isting literature, in-depth interviews, input from learned professionals, field testing, and comments from the individuals involved in each phase. Also, the development of the items was driven by the theory of planned behavior. Davis, et al. (1989), suggested that the ability to generalize beyond the study population may be enhanced because rese archers, such as Ajzen and Fishbein (1980), have reported extensive experience applying this theory to many different populations under many different contexts. Th is attention to the survey development process improved the design of the questionnai re and minimized the influence of method effect. The electronic delivery of the questionna ire improved data entry and helped guide the research process. First, the data entr y was completed by the respondents, so no data entry errors were introduced during the resear ch process. Also, multiple respondents sent electronic messages or commented at the end of the survey that the questionnaire was

PAGE 149

141 easy to complete and well organized. No repor ts were received expressing concern over the organization of the questionnaire. Lessons Learned Related to Online Data Collection The ability to collect data utilizing com puter and Internet te chnologies has made an important contribution to data collecti on efforts; however, recommended methods to follow continue to emerge. This study fo llowed the recommendation of Dillman (2000). Although most of the strategies outlined by Dillman (2000) were easy to employ and influenced survey delivery, a few deviations to the plan followed during this study may improve the easy of delivery w ithout adversely affecting st udy results. First, the importance of password protecting the website to assure that the respondents were those individuals invited to partic ipate may not be necessary if respondent information is collected for the purpose of follow-up removal. The methods used for this study required participants to ente r identifying information twice. Although website coding permits passing from one web screen to another, this requires the use of cookies, and computers can be configured to block cookies. As such, this pract ice can introduce many difficulties, including respondents choosing not to proceed with responding to the survey. Newer coding practices and the ability to mo re easily interact with databases could improve the practice of password protecting a survey questionnaire without introducing additional problems from passing potential respondents among different websites. The timing of follow-up notifications also provides valuable insight into the online survey process. Dillman (2000) suggest s that the initial surv ey invitation follow within two to three days of the pre-notification; however, he is less specific about the

PAGE 150

142 timing of the remaining follow up messages. For this project, the pre-notification was delivered on a Wednesday, with the initial invitation delivered on the following Monday. This timing allowed the first follow-up reminder to be sent within two to three days, and still in the same week. Interestingly, a fe w individuals responded to the first follow-up reminder with a message alerting the researcher that they were planning to complete the survey over the weekend. For future onlin e studies, this researcher will employ timing that delivers the pre-notification message on a Monday and the ini tial invitation on the following Wednesday. The first follow-up remi nder would be delivered on the following Tuesday. The remaining follow-ups would be delivered on different days of the week, with at least on weeke nd between notifications. In addition to the important informa tion obtained through the data collection effort, the online survey administration to colleagues provided a valuable networking experience. Many individuals invited to pa rticipate in the survey responded to the researcher personally, offering support, guidance, and chances for professional discussion. Even individuals who chose not to complete the que stionnaire provided suggestions for future studies and interesti ng life stories. Al though these discussions came during a very busy data collection pe riod, the researcher attempted to respond promptly to help convey that the responses re ceived were very valuable and appreciated. These discussion, reports from th e participants in the field te sting, and the n early 10% of respondents who reported they did not realize that computer-mediated instruction was an option for continuing education prior to taking the survey, s uggests that the delivery of

PAGE 151

143 the survey may have impacted perception related to computer-mediated continuing education. Future Directions This study provides valuable insight a bout the computer-mediated continuing education practices and per ceptions of health educat ors and health promotion professionals. Areas for further exploration should focus on talking to health educators and health promotion professionals to find out what types of learning opportunities would be relevant to their current practice. When talking to health education and health promotion professionals, constructing a better understandi ng of professional development, continuing education, and lifel ong learning from individuals in this group would also be important. The focus among health education and health promotion professionals related to c ontinuing education may be changing from a continuing education model to a continuing professiona l development model (National Commission for Health Education Credentialing, Inc. (N CHEC) (NCHEC, No date). Interestingly, NCHEC recently proposed changing the Continuing Education Contact Hours (CECH) associated with the Certified Health Education Specialist (CHES) certification to Professional Development Units (PDU). This change is not merely a name change but represents a change in philo sophy which contends that the individual is in the best position to determine which activities best f it his/her professional development needs (NCHEC, No Date, Question 2). Therefor e, identifying and u nderstanding general professional development and lifelong lear ning practices also may be important.

PAGE 152

144 The information collected during this research can help answer many additional research questions. One area to investigate fu rther is levels of Intention. Although this study focused on individuals who reported a moderate or high intention, looking more closely at individuals reporting a low inte ntion may lend additional insight into the process. Comparing Intention to particip ate in continuing education and professional development opportunities in general, with intention to participate in continuing education and professional de velopment using computer-media ted instruction, also could uncover important information. Additionally, information related to the online data collection, such as time survey was comple ted or location (home or work) from where survey was taken may provide valuable inform ation into this method of data collection. Another potentially important consideration is stage of readiness. Readiness could be studied using a comb ination of Prochaskas transtheoretical model and Rogers diffusion of innovation theory. These closel y related models address the sequence of stages an individual must pass through during the innovation-d ecision process, or as they adopt a behavior. Looking first at the transtheoretical model, stage one is precontemplation. An individual who is in pr econtemplation is just becoming aware that a problem exists. This corresponds to th e knowledge stage of the innovation-decision process. During this time, mass media and e ducational programs can be used to improve knowledge related to a topic area (Rogers, 1995) The second stage is contemplation, which parallels the persuasion stage in the innovation decision process. During this period, interpersonal communicatio n channels may be the most important contributor to moving individuals along the change continuum. For individuals at this stage, perceived

PAGE 153

145 characteristics of the innovation (relative advantage, compatibility, complexity, trialibility, and observability) become importa nt focal points for i ndividuals encouraging the adoption of computer-mediated instruc tion to consider (Rogers, 1995). Subjective norm, as conceptualized in the theory of planned behavior (Ajzen, 1988), may have an important influence during this stage. The preparation and action stages of the transtheoretical model, w ith the innovation-decision pr ocess correspondents of the decision stage and the implementation stage, also may have implication in encouraging individuals to adopt computer-mediated instru ction. Moreover, inform ation related to the maintenance/confirmation phases may shed impo rtant insight into th e continued use of computer-mediated education programs. Finally, two additional applic ations of this process may be warranted. First, repeating this study with a group of health educators wo rking in the local health departments may provide insight into the per ceptions of the indivi duals most likely to benefit from computer-mediated continuing education. These individuals are often limited in their ability to travel and are looking for oppor tunities to enhance their skills and knowledge. A potential plan for collecting, organizing and dissemina ting this type of information would be to conduct focus groups or individuals interviews with local health workers. When similar themes are expre sses by several individuals, develop computermediated learning experiences to fulfill the n eed. Share these learning opportunities with the individuals requesting the information and evaluate their impact Once developed, these learning modules could be packaged into a searchable database that others could access and query to find learning modules to meet their needs. Over time, multiple

PAGE 154

146 modules would be available, and meani ngful opportunities, as perceived by the individuals participating, could be determin ed through key word searches. Additionally, as computer software and hardware advan ce, the technology skills needed to develop these types of learning oppor tunities would decrease. Al ready, course management software is helping individuals with minima l computer skills deliver web-based and webenhanced learning opportunities. Also, collecting similar information from the same individuals in a time series study could help determine how the predic tive power of each independent variable changes over time. As suggested by the id ea to study computer-mediated instruction from a stages of change perspective, percep tions change over time. As such, monitoring the magnitude of the various associations between the study constr ucts and how these associations change over time may provide valuable insight into the adoption of computer-mediated instruction. Comparing th ese changes to other time-series studies using the theory of planned behavior may id entify similarities that, like the predictive power of the theory of planned behavior hold true under many contexts. A potential outcome from looking at the influences of Perceived Behavioral Control, Attitude, and Subjective Norm on Behavior over time would be that initially, t echnological factors (Perceived Behavioral Control) are the strongest contributor to Intention and ultimately Use. During the next phase, psychological a nd interpersonal factors (Attitudinal Beliefs) become an important contributor. Ultimat ely, sociological factors (Normative Beliefs) become the driving force behind getting nonusers involved in the practice or the behavior.

PAGE 155

147 References Ajzen, I. (1988). Attitudes, personality, and behavior Chicago, IL: The Dorsey Press. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179-211. Ajzen, I. (2001). Nature and operation of attitudes. Annual Review of Psychology, 52, 27-58. Ajzen, I. (2002a). Behavioral interventions based on the theory of planned behavior Retrieved June 23, 2003, from University of Massachusetts, Department of Psychology Web site: http://wwwunix.oit.umass.edu/~aizen /pdf/tpb.intervention.pdf Ajzen, I. (2002b). Constructing a TpB questionnaire: Conceptual and methodological considerations Retrieved June 23, 2003, from University of Massachusetts, Department of Psychology Web site: http://wwwunix.oit.umass.edu/~aizen/pdf/tpb.measurement.pdf Ajzen, I., & Fishbein, M. (1980). Understanding attitudes an d predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall. Allegrante, J. P., Moon, R. W., Aul d, M. E., & Gebbie, K. M. (1998). Preparing currently employed public health educat ors for changes in th e health system. New York, NY: Columbia School of Nursing. Allegrante, J. P., Moon, R. W., Auld, M. E., & Gebbie, K. M. (2001). Continuingeducation needs of the currently employe d public health education workforce. American Journal of Public Health, 91 (8), 1230-1234. American Association of University Women. (2001). The third shift: Women learning online [Online summary]. Available: http://www.aauw.org/research/3rdshift.cfm [July 16, 2003]. American Federation of Teachers. (2000). Going the distance: AFT guidelines for good practice in distance education [Online]. Available: http://www.aft.org/publications/on_campus/oct00/going.html [December 12, 2000].

PAGE 156

148 Armitage, C. J., and Conner, M. (2001). E fficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40 471-499. Barley, S. R. (1999). Computer-mediated distance education: Why and why not. The Education Digest, 65 (2), 55-59. Barron, A. E. (1998). Designing Web-based training. British Journal of Educational Technology, 29 (4), 355-370. Bernard, H. R. (2000). Social research methods Thousand Oaks, CA: Sage Publications Ltd. Blonna, R., & Shapiro, P. J. (2001). Learning at a distance. Health Promotion Practice, 2(3), 198-202. Bureau of Labor Statistics. (2001). Standard Occupational Classification [Online]. Available: http://www.bls.gov/soc/soc_f1j1.htm [October 17, 2001]. Carnevale, D. (2001). What matters in judging distance teaching ? Not how much its like a classroom course [Online]. The Chronicle of Higher Education Available: http://chronicle.com/free/2001/02/2001022101u.htm [September 23, 2001]. Coleman, M., Sims, C., & Threfall, D. (1998). What is the best safety training method? Occupational Hazards, 60 (10), 159-160. Couper, M. P., & Nicholls II, W. L. (1998). The history and development of computer assisted survey information collection methods. In M. P. Couper, R. P. Baker, J. Bethlehem, C. Z. F. Clark, J. Martin, W. L. Nicholls II, & J. M. OReilly (Eds.), Computer assisted survey information collection (pp. 1-22). Ne w York, NY: John Wiley & Sons, Inc. Crawford, S. D., Couper, M. P., Lamias, M. J. (2001). Web surveys. Social Science Computer Review 19 (2), 146-162. Daley, E. M. (2000). The relationship between temper ament type and self-reported health risk-taking behaviors in selected college students Unpublished doctoral dissertation, University of South Florida. Daley, E. M., McDermott, R. J., McCormack Br own, K. R., & Kittleson, M. J. (2003). Conducting web-based survey research : A lesson in Internet designs. American Journal of Health Behavior, 27 (2), 116-124. Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319-339.

PAGE 157

149 Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer technology: A comparison of two theoretical models. Management Science, 35(8), 982-1003. Dillman, D. A. (2000). Mail and Internet surveys: The tailored design method (2 nd ed.). New York, NY: John Wiley & Sons, Inc. Dominguez, P. S, & Ridley, D. (2001). Assessing distance e ducation courses and discipline differences in their effectiveness. Journal of Instructional Psychology, 28 (1), 15-19. Ehrmann, S.C. (1995). Asking the right questions. Change, 27, 20-27. Ellery, J., Brown, K., Perlmutter, P. (2001, February). Motivators, barriers, and overall use of Internet technology among health educators: Survey results Poster session presented at the 12 th annual art and science of health promotion conference, Washington, DC. Ellery, J., McCormack Brown, K., Perlmutter, P. (In press). Motivators, barriers, and overall use of Internet technology among health educators: Survey results [Abstract]. American Journal of Health Promotion Eng, T. R. (2001). The eHealth landscape: A terrain map of emerging information and communication technologies in health and health care Princeton, NJ: The Robert Wood Johnson Foundation. Fishbein, M., & Ajzen, I. (1975). Belief, attitude, inten tion and behavior: An introduction to theory and research Reading, MA: Addison-Wesley. Florida Center for Instru ctional Technology (1999). A Teachers Guide to Distance Learning [Online]. Available: http://fcit.coedu.usf.e du/DISTANCE/DEFAULT.HTM [September 23, 2001]. FrontRange Solutions (2001). Goldmine (Ver sion 5.5) [Computer software]. Colorado Springs, CO: Author. Fyfe, S., Leonard, H., Gelmi, R., Tassell, A., & Strack, R. (2001). Using the Internet to pilot a questionnaire on childhood di sability in Rett syndrome. Child: Care, Health and Development 27 (6), 535-543. Gebbie, K. M., Hwang, I. (1998). Preparing currently employed public health professionals for changes in the health system. New York, NY: Columbia University School of Nursing.

PAGE 158

150 Goldman, K. D., Florence, J., Cox, N. S., Hager, B., Johnson, L., & Ramsey, D. C. (2002). Five practitioners perspectives on the how, when, where, what, and why of continuing education. Health Promotion Practice, 3 (1), 12-17. Hanks, W. A., Barnes, M. D., Merrill, R. M., & Neiger, B. L. (2000). Computer task and application use by professional health e ducators: Implications for professional preparation. Journal of Health Education 31 (6), 314-319. Hatcher, L. (1994). A Step-by-step approach to using th e SAS system for factor analysis and structural equation modeling. Cary, NC: SAS Institute, Inc. Hatcher, L. & Stepanski, E. J. (1994). A Step-by-step approach to using the SAS system for univariate and multivariate statistics Cary, NC: SAS Institute, Inc. Heldrich, J. J. (2000). Nothing but net: American workers and the Internet economy New Brunswick, NJ: John J. Heldrich Ce nter for Workforce Development at Rutgers University. Available: http://heldrich.rutgers .edu/publications/ACFEB45.pdf [September 25, 2001]. Hoffman, D.L., & Novak, T.P (1994). Marketing in hypermed ia computer-mediated environments: Conceptual foundations. Nashville, TN: Vanderbilt University, Owen Graduate School of Ma nagement. Available: http://ecommerce.vanderbilt.e du/cme.conceptual.foundations.html [February 13, 2002]. Joo, Y., Bong, M., & Choi, H. (2000). Self-e fficacy for self-regulated learning, academic self-efficacy, and Internet self-efficacy in Web-based instruction. Educational Technology Research and Development, 48 (2), 5-17. Kraemer, H.C., & Thiemann, S. (1987). How many subjects? Newbury Park, CA: Sage. Kuzma, J.W. (1998). Basic statistics for the health sciences (3 rd ed.). Mountainview, CA: Mayfield Publishing Co. Macromedia (2000). Dreamweaver (Version 4.0 ) [Computer software]. San Francisco, CA: Author. Mathieson, K. (1991). Predicting user intentions: Comparing the technology acceptance model with the theory of planned behavior. Information Systems Research, 2 (3), 173-191. McCormack Brown, K. R., Ellery, J., & Perlmutt er, P. (in press). General characteristics of Internet use among health educators: Implications for the profession. Journal of Prevention and Intervention in the Community.

PAGE 159

151 McDermott, R. J., & Sarvela, P. D. (1999). Health education evaluation and measurement: A practitioners perspective (2 nd ed.). New York, NY: WCB/McGraw-Hill. McDonald, H. & Adam, S. (2003). A comparison of online and postal data collection methods in marketing research. Marketing Intelligence & Planning 21 (2), 85-95. Microsoft Corporation (2002). Microsoft Ex cel 2002 [Computer software]. Seattle, WA: Author. Morris, M. G., & Venkatesh, V. (2000). Age differences in technology adoption decisions: implications for a changing workforce. Personnel Psychology, 53 (2), 375-403. National Commission for Health Education Credentialing. (1996 ). A competency framework for professional development of certified health e ducation specialists New York, NY: Author. National Commission for Health Educa tion Credentialing, Inc (No Date). National Commission for Health Education Credentialing Proposes Major Policy Changes. Available: http://www.nchec.org/news/exops/faqs.htm [July 15, 2003]. National Commission for Health Ed ucation Credentialing. (1999). A competency-based framework for graduate-level health educators. New York, NY: Author. Netscape Communications Corporation (2001) Netscape Communi cator (Version 4.7) [Computer software]. Mountain View, CA: Author. OCarroll PW, & the Public Health Informa tics Competency Working Group. (2002). Informatics Competencies for Public Health Professionals. Seattle, WA: Northwest Center for Public Health Practice. Pealer, L., & Weiler, R. (2000 ). Research notes. Web-based health survey research: A primer. American Journal of Health Behavior, 24 (1), 69-72. Pedhazur, E. J., & Schmelkin, L. P. (1991). Measurement, design, and analysis: An integrated approach Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. Perlmutter, P., McCormack Brown, K ., & Ellery, J. (2000, November). Motivators, barriers and overall use of Internet technology among health educators. Poster session at the 128 th annual meeting of the American Public Health Association, Boston, MA.

PAGE 160

152 Phipps, R., & Merisotis, J. (2000). Quality on the line: benchmarks for success in Internet-based distance education [Online]. Washington, DC: Institute for Higher Education Policy. Available: http://www.ihep.com/quality.pdf [September 19, 2001]. Phipps, R., & Merisotis, J. (1999). Whats the Difference? A Review of Contemporary Research on the Effectiveness of Distance Learning in Higher Education [Online]. Washington, DC: Institute for Higher Education Policy. Available: http://www.ihep.com/difference.pdf [December 7, 2000]. Ramos, M., Sedivi, B. M., & Sweet, E. M. (1998). Computerized self-administered questionnaires. In M. P. Couper, R. P. Baker, J. Bethlehem, C. Z. F. Clark, J. Martin, W. L. Nicholls II, & J. M. OReilly (Eds.), Computer assisted survey information collection (pp. 389-408). New York, NY: John Wiley & Sons, Inc. Ranchhod, A., & Zhou, F. (2001). Comparing respondents of e-mail and mail surveys: Understanding the implications of technology. Marketing Intelligence & Planning 19 (4), 254-262. Rogers, E. M. (1995). Diffusion of Innovations (4 th ed.). New York, NY: The Free Press SAS Institute Inc. (2001). SAS (Version 8.02) [Computer software]. Cary, NC: Author. Sax, L. J., Gilmartin, S. K., & Bryant, A. N. (2003). Assessing response rates and nonresponse bias in web and paper surveys. Research in Higher Education 44(4), 409-432. Schwartz, L., ORourke, T. W., Eddy, J. M., Auld, M. E., & Smith, B. (1999). Use and impact of competencies for entry-level health educators on professional preparation programs. Journal of Health Education 30 209-214. Smith, S. B., Smith, S. J., & Boone, R. (2000) Increasing access to teacher preparation: The effectiveness of traditional instruc tional methods in an online learning environment. Journal of Special Education Technology, 15 (2), 37-46. Society for Public Health Education (No date). SOPHE bylaws. Available: http://www.sophe.org/about/bylaws.html [September 23, 2001]. Sonner, B. S., (1999). Success in the capstone business course assessing the effectiveness of distance learning. Journal of Education for Business, 74 (4), 243-247. Steckler, A., Farel, A., Bontempi, J. B., Umble, K., Polamus, B., & Trester, A. (2001). Can health professionals l earn qualitative ev aluation methods on the World Wide Web? A case example. Health Education Research, 16 (6), 735-745.

PAGE 161

153 Stokes, M. E. Davis, C. S., & Koch, G.G. (1995). Categorical data analysis using the SAS system Cary, NC: SAS Institute, Inc. Studach, J. R. (2001). An assessment of the computi ng habits of health promotion professionals. Unpublished masters thesis, Amer ican University. Available: http://www.american.edu/academic.depts/ cas/health/nchf/pubsmyismasters.html [September 25, 2001]. Swerissen, H, & Tilgner, L. (2000). A workfo rce survey of health promotion education and training needs in th e state of Victoria. Australian and New Zealand Journal of Public Health, 24 (4), 407-412. Taylor, S., & Todd, P. A. (1995a). Understa nding information technology usage: A test of competing models. Information Systems Research, 6 (2), 144-176. Taylor, S., & Todd, P. A. (1995b). Assessing IT usage: The roll of prior experience. MIS Quarterly, 19, 561-570. Umble, K. E., Cervero, R. M., Yang, B. (2000) Effects of traditional classroom and distance continuing education: A theory-driven evaluation of a vaccinepreventable diseases course. American Journal of Public Health, 90 (8), 12181224. U.S. Department of Health and Human Serv ices, Public Health Service. (1998). The public health workforce: An agenda for the 21 st century [Online]. Available: http://www.health.gov /phfunctions/pubhlth.pdf [September 19, 2001]. Web-based Education Commission. (2000). The power of the Internet for learning: Moving from promise to practice Available: http://interact.hpcnet.org/webcommission/index.htm [September 23, 2001]. Wenger, S. B., Holloway, K. C., & Garton, E. M. (1999). The effects of Internet-based instruction on student learning. Journal of Asynchronous Learning [Online] 3(2), 1-9. Available: http://www.aln.org/alnweb/jour nal/Vol3_issue2/Wegner.htm [September 19, 2001]. White, J.A., Carey, L.M., & Dailey, K.A. (2001). Web-based instrumentation in educational survey research. WebNet Journal: Internet Technologies, Applications and Issues 3 (1), 46-50. White, J., Nicholson, T., & Duncan, D. (2000). Drugnet 2000 [Online]. Available: http://www.accessky.n et/illicit-drugs/ [January 25, 2001]. Willis, B. (1995). Distance education at a glance guides 1-13, and glossary [Online]. Engineering Outreach, College of Engineer ing, University of Idaho. Available: http://www.uidaho.edu/evo/distglan.html [December 11, 2000].

PAGE 162

154 Young, J. R. (2000). Scholar concludes that distance ed is as effective as traditional instruction [Online]. The Chronicle of Higher Education Available: http://chronicle.com/free/2000/02/2000021001u.htm [September 23, 2001].

PAGE 163

155 Appendices

PAGE 164

156 Appendix A. A Structural Model of the Theory of Planned Behavior. Figure 2. A Structural Model of the Theory of Planned Behavior Subjective Norms for Behavior Attitude Toward Behavior Perceived Behavioral Control for Behavior Intention Usage/Behavior

PAGE 165

157 Appendix B. A Structural Model of the Theory of Reasoned Action. Figure 3. A Structural Model of the Theory of Reasoned Action. Subjective Norms for Behavior Attitude Toward Behavior Intention Usage/Behavior

PAGE 166

158 Appendix C. Listing of Hea lth Education Competencies Part A: Entry-Level Competen cies for Health Educators 1 1. Assessing individual and commun ity needs for health education Obtain health-related data about social and cultural environments, growth and development factors, needs, and interests Distinguish between behaviors that fost er and those that hinder well-being Infer needs for health educati on on the basis of obtained data Determine factors that influence le arning and development (graduate level) 2. Planning effective health education programs Recruit community organizat ions, resource people, and potential participants for support and assistance in program planning Develop a logical scope a nd sequence plan for a h ealth education program Formulate appropriate and m easurable program objectives Design education program consistent with specified program objectives Develop health education programs using social marketing principles (graduate level) 3. Implementing health education programs Exhibit competency in carrying out planned programs Infer enabling objectives as needed to implement instructional program in specified settings Select methods and media best suited to implement program plans for specific learners Monitor educational programs and adjust objectives and activities as necessary 4. Evaluating effectiveness of health education programs Develop plans to assess achievement of program objectives Carry out evaluation plans Interpret results of program evaluation Infer implications from findings for future program planning

PAGE 167

159 Appendix C (Continued) 5. Coordinating provision of health education services Develop a plan for coordinating health education services Facilitate cooperation between and among levels of program personnel Formulate practical modes of collabo ration among health agencies and organizations Organize in-service training for teacher s, volunteers, and other interested personnel 6. Acting as a resource person in health education Use computerized health information retrieval system effectively Establish effective consultative relationshi ps with those requesting assistance in solving health-related problems Interpret and respond to reque sts for health information Select effective educational re source materials for dissemination 7. Communicating health and health edu cation needs, concerns, and resources Interpret concepts, purposes, a nd theories of health education Predict the impact of societal value systems on health education programs Select a variety of communication met hods and techniques in providing health information Foster communication between health care providers and consumers

PAGE 168

160 Appendix C (Continued) Part B: Graduate-Level Competen cies for Health Educators 2 8. Applying appropriate research principl es and methods in health education Conduct thorough reviews of the literature Use appropriate qualitative a nd quantitative research methods Apply research to health education practice 9. Administering health education programs Develop and manage fiscal resources Develop and manage human resources Exercise organizational leadership Obtain acceptance and support for programs 10. Advancing the professi on of health education Provide a critical analysis of current and future needs in health education Assume responsibility for advancing the profession Apply ethical principles as they relate to the practice of health education

PAGE 169

161 Appendix C (Continued) Part C: Continuing Education Competencies for the Currently Employed Public Health Education Workforce 3 1. Advocacy Communication skills to work with polit ical officials at various levels of government Integrating multidisciplinary understandings Knowledge of legal boundaries and ramifications Leadership in the legislative process Political analysis and acuity and organizational politics Public policy development and environmental change Strategies to influe nce key decision makers 2. Business management and finance Budgeting Fiscal management Grant writing Resource development 3. Communication Media advocacy Media relations Social marketing 4. Community health planning and developm ent, coalition build ing, and leadership Capacity-building skills Community-change strategies and coalition building Community organizing Consultation Ecologic approaches and multiple strategies at multiple levels Organizing natural helpers and community-based lay extenders Skills to support local health planning bodies, facilitation, and decision making

PAGE 170

162 Appendix C (Continued) 5. Computing and technology Computing literacy Distance learning Electronic communications and access to the World Wide Web 6. Cultural competency Adapting public health education prac tice to the needs of diverse populations Developing bilingual capacity Understanding the implications for public health of growing racial, ethnic, and linguistic diversity and the need for inclusivity 7. Evaluation Assessing and using evidence-based data and other information in designing programs Using quantitative and qualitative methods Defining success and outcomes of health education practice Developing methods that evaluate complex so cial factors that i ndicate shifts in health status 8. Strategic planning Community health assessment Environmental forecasting and assessing community readiness and trends Incorporating social change and social justice into the public health system Intersectoral skills Systems analysis Team-building skills Translating theory into practice

PAGE 171

163 Appendix C (Continued) Part D: Core Competencies for th e Current Public Health Workforce 4 1. Public health values and acculturation: pr ovide a basic understanding of public health, its history, its heroes, its value, and its methods 2. Epidemiology, quality assurance, and econom ics: provide basic sk ills in evaluative science and concepts, and their application to public health 3. Informatics: provides a basic understanding of how to use technology to communicate information effectively 4. Communication: provides a basic understa nding of the princi ples of effective communication and the importance of co mmunication in educating, marketing, and multidisciplinary collaboration n ecessary in public health practice 5. Cultural competency: provides a basic unde rstanding of the importance of cultural competency in public health practice 6. Team building and organization effectiv eness: provide a basic understanding of teamwork, the principles associated w ith effective organization, and the value these have in public health practice 7. Strategic thinking and planni ng/visioning: provide a basic understanding of the tools and value of strategic thinking and vi sioning to the practice of public health 8. Advocacy, politics, and policy developmen t: provide a basic understanding of how public health policy is developed an d changed, including understanding who makes policy, how it is made, what it is based on, and how it is implemented 9. External coalition building and mobilizati on: provide the skills needed for developing and sustaining needed community relationships Source: Allegrante, J.P., Moon R., Auld, M.E., Gebbie, K.M. ( 2001). Continuing-education needs of the currently employed public health education workforce. American Journal of Public Health, 91 (8), 1230-1233. 1 National Commission for Health Education Credentialling. (1996). A Competency Framework for Professional Development of Certified Health Education Specialists New York, NY: Author. 2 National Commission for Health Education Credentialling. (1999). A Competency-Based Framework for Graduate-Level Health Educators Allentown, PA: Author. 3 Allegrante, J.P., Moon R., Auld, M.E., Gebbie, K.M. (1998). Preparing Currently Employed Public Health Educators for Change in the Health System. New York, NY: Colunbia University School of Nursing. 4 Gebbie, K.M., Hwang, I. (1998). Preparing Currently Employed Public Health Professionals for Change in the Health System New York, NY: Colunbia University School of Nursing.

PAGE 172

164 Appendix D. Letters of Approval from the Institutional Review Boards.

PAGE 173

165 Appendix D (Continued)

PAGE 174

166 Appendix D (Continued)

PAGE 175

167 Appendix D (Continued)

PAGE 176

168 Appendix D (Continued)

PAGE 177

169 Appendix E. Pre-notice Electronic Message From: Jane Ellery To: jellery@hsc.usf.edu Sent: Monday, October 15, 2001 8:27 PM Subject: Dissertation Survey: Computer-Med iated Instruction in Continuing Education Dear Colleague, My name is Jane Ellery, and I am a graduate st udent at the University of South Florida. I am interested in learning more about the c ontinuing education practi ces and interests of health educators and other he alth promotion professionals. Over the past few years, computers and the Internet have been identifi ed as important continuing education tools. Unfortunately, we have little information about the use of compute r-mediated instruction within our profession. The goal of my projec t is to collect information that may be helpful in identifying and developing strategies to make this innovative practice an effective form of professional development. At the start of next week, you will receive a message from me directing you to a survey web site. Your input is important, and I w ould greatly appreciate it if you would take a few moments to complete the questionnaire. In exchange for your time, I will be happy to send you a listing of the online learning opportunities shared by survey respondents and a summary of the results of my study. Ho wever, if you prefer not to participate, simply reply to this message with the word "remove" in the body, and you will not receive any further notifications. Thank you in advance for time and contribution to my project. If you have questions or comments regarding this study, please cont act me at 813-991-6270, or by e-mail at jellery@hsc.usf.edu. Sincerely, Jane Ellery Department of Community and Family Health College of Public Health, University of South Florida

PAGE 178

170 Appendix F. Electronic Mess age Requesting Participation From: Jane Ellery To: jellery@hsc.usf.edu Sent: Monday, October 15, 2001 9:22 PM Subject: Dissertation Survey: Computer-Med iated Instruction in Continuing Education Dear Colleague, I would like to take this opportunity to invite yo u to participate in a survey of health educators and other health promotion professionals. The pur pose of this survey is to gain a better understanding of your perceptions and practices rela ted to computer-mediated instruction. Several reports and organizations encourage the use of computers and the Internet for the delivery of educational programs for workforce developmen t. As such, collecting information from the individuals who will ultimately be using th ese services is imperative to the successful development of meaningful continuing education opportunities. Please understand that your participation in this su rvey is completely voluntary, and your consent to participate is implied by your decision to link to the survey website. Although I will monitor your personal identification number (PIN) and Password to remove your name from future notifications once you have completed the survey, these fields will be removed from the file before the data is analyzed. This project has been approved by the Univers ity of South Florida Institutional Review Board (IRB #100277/IRB telephone 813-974-5638), and all info rmation will remain confidential. The survey should take you less than 15 minutes to complete. The opening screen is password protected to limit individuals who have not been invited to participate from submitting information. To access the survey, you will need to use the PIN and password listed below. PIN: <<&key5>> PASSWORD: <<&firstname>> Survey Website: http://pe.usf.edu/~eller y/jane/survey/Intro.htm Your input is very important to this project. Thank you in advance for completing the survey. Because the study is being c onducted over the Internet, it is possible, however unlikely, unauthorized individuals could gain access to your response. If you have any questions, please feel free to contact me at 813-991-6270, or by e-mail at jellery@hsc.usf.edu. Sincerely, Jane Ellery Graduate Student, University of South Florida

PAGE 179

171 Appendix G. Follow-up Notification 1 From: Jane Ellery To: jellery@hsc.usf.edu Sent: Monday, October 15, 2001 9:37 PM Subject: Dissertation Survey: Computer-Med iated Instruction in Continuing Education Dear Colleague, Earlier this week, I sent you an e-mail message ask ing you to visit a website to take a survey related to your experiences and pe rceptions about continuing edu cation and computer delivery of educational programs. I realize you are busy with many other projects and may even be out of the office this week. I wanted to send you this quick reminder because I value your input. I contacted you and other health promotion prof essionals because you are the only individuals who can provide this insight. To make it easy for you to link to the website, I past ed a copy of your original invitation below. This message contains the web address, your PIN, and your Password. Thank you in advance for your participation. Sincerely, Jane Ellery Graduate Student, University of South Florida ****************************************************************************** Note: When you complete the survey, if you provide your PIN and Password on the survey form your name should automatically be removed from fo llow up notifications. If you are not able to complete the survey at this time and would like to have your name removed from receiving additional follow up messages, please reply to this message with the word "remove" in the body. Original message: Dear Colleague, I would like to take this opportunity to invite yo u to participate in a survey of health educators and other health promotion professionals. The pur pose of this survey is to gain a better understanding of your perceptions and practices rela ted to computer-mediated instruction. Several reports and organizations encourage the use of computers and the Internet for the delivery of educational programs for workforce developmen t. As such, collecting information from the individuals who will ultimately be using th ese services is imperative to the successful development of meaningful continuing education opportunities.

PAGE 180

172 Appendix G (Continued) Please understand that your participation in this su rvey is completely voluntary, and your consent to participate is implied by your decision to link to the survey website. Although I will monitor your personal identification number (PIN) and Password to remove your name from future notifications once you have completed the survey, these fields will be removed from the file before the data is analyzed. This project has been approved by the Univers ity of South Florida Institutional Review Board (IRB #100277/IRB telephone 813-974-5638), and all info rmation will remain confidential. The survey should take you less than 15 minutes to complete. The opening screen is password protected to limit individuals who have not been invited to participate from submitting information. To access the survey, you will need to use the PIN and password listed below. PIN: <<&key5>> PASSWORD: <<&firstname>> Survey Website: http://pe.usf.edu/~ellery/jane/survey/Intro.htm Your input is very important to this project. Thank you in advance for completing the survey. Because the study is being c onducted over the Internet, it is possible, however unlikely, unauthorized individuals could gain access to your response. If you have any questions, please feel free to contact me at 813-991-6270, or by e-mail at jellery@hsc.usf.edu.

PAGE 181

173 Appendix H. Follow-up Notification 2 From: Jane Ellery To: jellery@hsc.usf.edu Sent: Monday, October 15, 2001 9:37 PM Subject: Dissertation Survey: Computer-Med iated Instruction in Continuing Education LAST CHANCE... This will be the last reminder I send encouragin g you to reply to my survey. I'm sorry to continue to prompt you, but I think everyone's opinion is im portant, and I don't want to analyze the data without including your input! The survey website will be available through June 8th. Your answers are confidential and will be combined with others before disseminating results. Please understand that your participa tion in this survey is completely voluntary, and your consent to participate is implied by your decision to link to the survey website. This project has been approved by the Univer sity of South Florid a Institutional Review Board (IRB #100277/IRB telephone 813-974-5638) and the survey should take you about 15 minutes to complete. The openi ng screen is password protected to limit individuals who have not been invited to participate from submitting information. To access the survey, you will need to use the PIN and password listed below. PIN: <<&key5>> PASSWORD: <<&firstname>> Survey Website: http://pe.usf.edu/~ellery/jane/survey/Intro.htm Your input is very important to this proj ect. Thank you in advance for completing the survey. Because the study is being conducted over the In ternet, it is possible, however unlikely, unauthorized individuals could gain access to your response. If you have any questions, please feel free to contact me at 813991-6270, or by e-mail at jellery@hsc.usf.edu. Sincerely, Jane Ellery Graduate Student, University of South Florida

PAGE 182

174 Appendix I. Review Panel Members Table 19. Review Panel Members I nvited to Review Survey Tool Name Professional Affiliation Areas of Experience HE CE CI IS Ginger Phillips, M.A. University of South Florida X M. Elaine Auld, M.A. SOPHE X X Nancy Gaston, M.A, R.D. United States Department of Agriculture X X Jay Bernhardt, Ph.D. Emory University X X X X Mark Kittleson, Ph.D. Southern Illinois University X X X X Ellen Daley, Ph.D. University of South Florida X Lisa Pealer, Ph.D. Centers for Disease Control and Prevention X X James White, Ph.D. University of South Florida X Alyson Taub, Ph.D. New York University X X X Cathy Hutsell, M.A. Centers for Disease Control and Prevention X John Allegrante, Ph.D. Columbia University X X Nancy Atkinson, Ph.D. University of Maryland X X X Note. Individuals represented in italics responded to request. HE=Health Education, CE=Continuing Education, CI=Computer-mediated Instruction, IS=Internet Survey Delivery

PAGE 183

175 Appendix J. Operationalizati on of Independent Variables Table 20. Study Variables and the Statements and Format for their Operationalization Variable Statements Operationalization Attitudes related to self-directed computermediated instruction for continuing education If I were able to use self-paced continuing education programs delivered using computers it would improve my ability to participate in professional development programs. Using computers to deliver self-paced continuing education programs is an effective option for professional development. I would be a good candidate for self-paced continuing education classes delivered using computers. Self-paced continuing education delivered using computers is an effective way for me to learn. Delivering continuing educati on programs using self-paced computer-mediated instruction is a good idea. I am willing to take self-paced continuing education courses delivered using computers. I would use self-paced educational programs delivered using computers if they were available to me. Composite index Subjective norms related to self-directed computermediated instruction for continuing education Individuals who are important to me think using computers for self-paced continuing education is a good idea. Professional organizations important in my field encourage the use of computers to deliver self-paced continuing education programs. Individuals who I respect encourage the use of self-paced continuing education programs delivered using computers. My managers and supervisors encourage the use of selfpaced continuing education programs delivered using computer-mediated instruction. My colleagues encourage participation in self-paced continuing education programs delivered using computers. Individuals who influence my behavior think using selfpaced computer-mediated instruction for continuing education is a good idea. Composite index Continued on the next page

PAGE 184

176 Appendix J (Continued) Table 20 (Continued) Perceived behavioral control related to self-directed computermediated instruction for continuing education I have the skills that are need ed to participate in self-paced continuing education programs delivered using computers. Programs or classes using computers to deliver self-paced continuing education are available to me. It would be easy for me to take a self-paced continuing education program delivered using a computer. There are opportunities available for me to continue my education using self-paced programs delivered using computers. I have the time that is needed to participate in self-paced continuing education programs delivered using computers. I have the financial resources that are needed to participate in self-paced continuing educati on programs delivered using computers. I have the equipment that I would need to participate in selfpaced continuing education programs delivered using computers. If offered the chance to take a self-paced course that was of interest to me and that was delivered using a computer, I am confident that I would be able to complete it. Composite index Continued on the next page

PAGE 185

177 Appendix J (Continued) Table 20 (Continued) Behavioral intentions related to selfdirected computermediated instruction for continuing education I intend to use self-paced computer-mediated instruction as a continuing education tool. During the next 12 months, I will take a self-paced educational program that is delivered using a computer. When considering the following courses or training programs delivered using computers, in which of the following would you be willing to participate? A program delivered from a CD-ROM/DVD. A program offered over the Internet. A class located on a computer (such as seen with a computer kiosk). A program using multiple computer-mediated delivery methods, such as a CD-ROM and the Internet. Other Have you ever tried to locate continuing education programs being delivered using a computer? Researcher developed score based on combined responses Continued on the next page

PAGE 186

178 Appendix J (Continued) Table 20 (Continued) Behavior/ Experience related to self-directed computermediated instruction for continuing education In which of the following type(s) of computer-mediated instruction courses or training programs have you previously participated? A program delivered from a CD-ROM/DVD. A program offered over the Internet. A class located on a computer (such as seen with a computer kiosk). A program using multiple computer-mediated delivery methods, such as a CD-ROM and the Internet. Other How many classes or training programs have you taken that were delivered using computer-mediated instruction? Have you ever taken classes or training programs using computer-mediated instruction, from which completion resulted (for you or any of the participants) in continuing education credit, university credit, or some type of certification? Researcher developed score based on combined responses Continued on the next page

PAGE 187

179 Appendix J. (Continued) Table 20 (Continued) Barriers related to self-directed computermediated instruction for continuing education Current level of comfort with using computers Current computer skills Current level of comfort with using the Internet Current Internet skills Computer access Internet access/connection Lack of motivation to continue education Lack of discipline to complete self-directed programs Not a good way for me to learn Lack of technical support for programs Lack of opportunity to apply training Lack of incentives for continuing education Lack of work time release for continuing education Lack of access to information about programs Expense associated with taking continuing education programs Lack of programs Lack of immediate feedback during programs Lack of relevant topics for programs Difficulty level of currently available courses Lack of interest in using computer-mediated learning Lack of professional rewards for continuing education Lack of importance placed on continuing education in my current position Lack of importance placed on continuing education in the health education/health promotion field Lack of professional networking during programs Lack of social interaction Lack of time Lack of interaction with faculty/instructor Other Individually

PAGE 188

180 Appendix K. Study Questionnaire

PAGE 189

181 Appendix K (Continued)

PAGE 190

182 Appendix K (Continued)

PAGE 191

183 Appendix K (Continued)

PAGE 192

184 Appendix K (Continued)

PAGE 193

185 Appendix K (Continued)

PAGE 194

186 Appendix K (Continued)

PAGE 195

187 Appendix K (Continued)

PAGE 196

188 Appendix K (Continued)

PAGE 197

189 Appendix K (Continued)

PAGE 198

190 Appendix L. In-depth Interview Guide Interview Guide Self-paced computer-mediated instruction as a delivery method for continuing education in health education Im going to ask you some questions about using computers and the Internet for continuing education. I want to get a better understanding of your attitudes and opinions about computer-mediated instruction. Here are some definitions I have pu t together to related to the study. What type of programs/classes would you consider to fall into this category? If you saw this definition at the start of a su rvey, do you think it is enough of a primer to get you thinking about CMI, or should the definition be expanded? What is your overall opinion about the us e of computer-media ted instruction for continuing education? What kind of an attitude do you have rela ted to computer-mediated instruction? (effective, useless, isolating, lik e the idea, limits networking) Who most influences your prof essional decisions, such as deciding whether to participate in computer-mediated instruction for continuing education? What issues and barriers come to mind when I ask you if you thought you would be able to take a computer-mediated instruc tion program for continuing education?

PAGE 199

191 Appendix L (Continued) What would you consider the advantages of computer-mediated instruction? Disadvantages? Have you ever participated in computer-media ted instruction? For continuing education? Do you need continuing education for licensure/certification? If you received a request to take this survey in an e-mail, what incentives would increase the likelihood that you would complete it? Notes from survey items: Do you think the experience questio n should carry a time qualifier? If yes, how long? 1 year? 2 years? Can you think of other questions I should be asking? Can you think of other definitions it would be important to include?

PAGE 200

192 Appendix M. Compiled Field Note s from In-Depth Interviews Interview Guide Self-paced computer-mediated instruction as a delivery method for continuing education in health education Im going to ask you some questions about using computers and the Internet for continuing education. I want to get a better understanding of your attitudes and opinions about computer-mediated instruction. Here are some definitions I have pu t together to related to the study. What type of programs would you cons ider to fall into this category? Programs on the computer, unive rsity classes, CE programs. Web-enhanced, web-based, tr aining, college course, CE -business, training for new products. College courses Discussion groups Networking Tutorial University course work, Professional/CE classes, In-service workshops offered over the Internet If you saw this definition at the start of a su rvey, do you think it is enough of a primer to get you thinking about CMI, or should the definition be expanded? Yes. But add something about for this study Yes. General enough. Good It is enough Okay as is What about adding the term self-paced to the definition? How does it change the type of programs you consider to be included? Eliminates things like teleconferencing.

PAGE 201

193 Appendix M (Continued) What is your overall opinion about the us e of computer-mediated instruction for continuing education? Good for people with limited time, inability to travel. Allows avenue to succeed. Its the way of the future Effective and effecient Limits professional networking. In general a good thing. Not for everyone. Better in some areas than others Groups with which delivered depends on effectiveness May be too easy. Good Self-paced/on your time What kind of an attitude do you have rela ted to computer-mediated instruction? (effective, useless, isolating, like the idea, limits networking) Depends on student. Depends on time. Isolating, cold Intimidating Limits networking/collaboration. Self-paced is good Visual elements effective Notes are good. Ability to go back and review is good. Strong positive attitude Who most influences your professional decisi ons, such as deciding whether to participate in computer-mediated instruct ion for continuing education? Self, family. Administrators/boss Certification requirements Self-directed Professors Colleagues Respected others Spouse Administrator/Boss Self

PAGE 202

194 Appendix M (Continued) What issues and barriers come to mind when I ask you if you thought you would be able to take a computer-mediated instruction program for continuing education? (From Swerissen Programs not relevant, No f unding, Training too expensive, No support, Lack of time, No relevant training, Lack of acce ss to information, cou rses not sufficiently practical, travel restrictions, No work tim e release, other training more important, no opportunity to apply training, lack of profes sional credit Also see incentives) Comfort with computer. Self discipline. Self-motivation. Need set deadlines. Dont enjoy. Internet connection/quality Availability Cost of course Cost Degree of Difficulty Course design Internet exchange/telecom ability Content/workload Time Family responsibilities Poor design of currently available offerings Usability of packages What would you consider the advantages of computer-mediated instruction? (from literature Swerissen, convenience, redu ced cost, continually updating information, reduced time away from workplace, a format th at allows learning to occur at individuals pace; Mamary, too difficult to use, too expensive, too time consuming, prefer in-person training, not interested, dont know what it is, dont know how to use it) Self-paced. Learn at own time and schedule. Flexibility Time-own time. Convenient Avoid boring lecturers Pace according to relevance Own time Ability to communicate with professor Time saved on travel Own pace/own time

PAGE 203

195 Appendix M (Continued) Not being bound to time specifics Convenience Disadvantages? Lack of communication with others Time in interaction with others Lag in getting feedback Takes longer to type than to talk Inability for immediate clarification Must be self-motivated Availability of equipment Technology issues Availability of courses Frustrating if there is no contact person Lack of/poor group discussions and networking Have you ever participated in computer -mediated instruction? For continuing education? No/No Yes/Yes Yes/No Yes/No No/No Yes/No No/No No/No Do you need continuing educati on for licensure/certification? No No Yes No No No No No

PAGE 204

196 Appendix M (Continued) If you received a request to ta ke this survey in an e-mail, wh at incentives would increase the likelihood that you would complete it? Nothing Recognized sender Ease Conference registration mi ght, but probably not. Support of an organization Conference fee (maybe) Decrease on next membership fee. Professional organization in wh ich I had a vested interest Offer of summary of results Maybe, conference registration fee Notes from discussion of the survey items: Do you think the experience questio n should carry a time qualifier? If yes, how long? 1 year? 2 years? No Not needed No Okay as is Can you think of other questions I should be asking? Networking Can you think of other definitions it would be important to include? ********************************************************************** Are all of your centers N/A May have license that is not current/being used Define resources ************************************************************************ Based on Mamary article, add years of practice question Mamary consider adding rank orde r (1, 2, 3) of CE opportunities.

PAGE 205

197 Appendix N. Review Panel Packet for Dissertation

PAGE 206

198 Appendix N (Continued)

PAGE 207

199 Appendix N (Continued)

PAGE 208

200 Appendix N (Continued)

PAGE 209

201 Appendix N (Continued)

PAGE 210

202 Appendix N (Continued)

PAGE 211

203 Appendix N (Continued)

PAGE 212

204 Appendix N (Continued)

PAGE 213

205 Appendix N (Continued)

PAGE 214

206 Appendix N (Continued)

PAGE 215

207 Appendix N (Continued)

PAGE 216

208 Appendix N (Continued)

PAGE 217

209 Appendix N (Continued)

PAGE 218

210 Appendix N (Continued)

PAGE 219

211 Appendix N (Continued)

PAGE 220

212 Appendix N (Continued)

PAGE 221

213 Appendix N (Continued)

PAGE 222

214 Appendix O. Results from Review Panel Structured Panel Review Form Results Construct Reviewer # 1 2 3 4 5 6 7 8 9 Tot Ave Perceived Behavioral Control I have the skills that are needed for participation in continuing education programs delivered using selfdirected computer-mediated instruction. 0 3 2 2 3 2 2 2 16 2.00 I could participate in a selfdirected continuing education program delivered using computer-mediated instruction if I wanted to. 0 2 3 2 3 1 1 2 14 1.75 I have access to self-directed computer-mediated educational opportunities for continuing education. 3 2 3 2 3 2 2 3 20 2.50 There are opportunities available for me to take selfdirected continuing education programs using computermediated instruction. 3 2 2 2 3 2 2 3 19 2.38 It is possible for me to take a self-directed continuing education program delivered using computer-mediated instruction. 3 2 3 2 0 1 2 3 16 2.00 I have control over my ability to use self-directed computermediated instruction for continuing education. 0 2 2 2 3 1 1 1 12 1.50 I have the equipment that I would need to participate in continuing education programs delivered using selfdirected computer-mediated instruction. 0 2 3 2 3 2 2 14 2.00

PAGE 223

215 Appendix O (Continued) I have the time that is needed to participate in continuing education programs delivered using self-directed computermediated instruction. 3 2 1 2 3 2 2 3 18 2.25 I have the financial resources that are needed to participate in continuing education programs delivered using selfdirected computer-mediated instruction. 3 2 3 2 3 2 1 3 19 2.38 Appropriate representation 2 3 4 5 3 1 4 5 27 3.38 Attitude Using continuing education programs delivered through self-directed computermediated instruction would improve my ability to participate in professional development programs. 3 1 3 2 3 1 2 3 18 2.25 Self-directed continuing education delivered using computer-mediated instruction is an effective way for me to learn. 3 2 1 2 3 3 2 3 19 2.38 Using self-directed computermediated instruction to deliver continuing education programs is an effective option for professional development. 3 1 2 2 3 1 1 3 16 2.00 The delivery of continuing education programs to health educators using self-directed computer-mediated instruction is a good idea. 3 2 3 2 3 1 2 2 18 2.25 I would be a good candidate for a self-directed continuing education class delivered using computer-mediated instruction. 3 1 3 2 3 2 1 3 18 2.25 I am willing to take a selfdirected continuing education course that is delivered using computer-mediated instruction. 3 1 3 2 3 1 2 2 17 2.13

PAGE 224

216 Appendix O (Continued) It would be easy for me to take a self-directed continuing education program using computer-mediated instruction. 3 1 1 2 3 0 2 2 14 1.75 Using self-directed computermediated instruction to deliver continuing education programs is an effective option for professional development. 3 2 2 2 3 2 1 3 18 2.25 Appropriate representation 4 3 4 5 2.5 3 3 4 28.5 3.56 Subjective Norm I am a member of professional organizations that encourage me to use self-directed computer-mediated instruction for continuing education. 0 2 1 2 3 1 2 3 14 1.75 Individuals who are important to me think using self-directed computer-mediated instruction for continuing education is a good idea. 3 1 1 2 3 2 1 3 16 2.00 My managers and supervisors think I should use selfdirected computer-mediated instruction for continuing education. 3 0 2 2 3 3 1 3 17 2.13 Individuals who I respect think I should use selfdirected continuing education programs delivered using computer-mediated instruction. 3 0 1 2 3 1 2 3 15 1.88 My colleagues encourage participation in self-directed continuing education programs delivered using computer-mediated instruction. 3 2 1 2 3 2 2 3 18 2.25 Individuals who influence my behavior think I should use self-directed computermediated instruction for continuing education. 3 0 1 2 3 1 0 3 13 1.63 Appropriate representation 4 5 4 4 3 2 3 5 30 3.75

PAGE 225

217 Appendix O (Continued) Intention I would use self-directed continuing education programs being delivered using computer-mediated instruction if I had access to them. 3 1 3 2 3 3 2 3 20 2.50 It is likely that I will participate in a self-directed computer-mediated instructional program for continuing education. 3 1 3 2 3 2 2 3 19 2.38 I intend to use self-directed computer-mediated instruction as a continuing education tool. 3 1 3 2 3 3 2 3 20 2.50 During the next 12 months, I will take a class that is delivered using self-directed computer-mediated instruction. 3 2 3 2 3 2 1 3 19 2.38 My intention is to take a selfdirected continuing education program delivered using computer-mediated instruction. 3 3 2 3 1 1 1 14 2.00 Appropriate representation 4 5 4 4 2 3 1 5 28 3.50 Behavior/Experience Appropriate representation experience 4 4 4 5 3 3 3 5 31 3.88 Appropriate representation continuing-education 2 5 4 5 2 4 4 5 31 3.88 Survey Tool Use with health educators 3 2 4 5 3 4 3 5 29 3.63

PAGE 226

218 Appendix O (Continued) Comments PBC: First, for all inappropriate responses, it was listed as such because I'm not sure if people will know what it will take. For ex ample, will the people know what skills it takes to do a computer-mediated program. I don't know if they can answer it if they don't know what it will take. PBC: Either you can and do, or you don't and won't use computer assisted instruction for continuing education. Those who have the equipment, probably also have the know-how and motivation to pursue compute r-assisted instruction and are comfortable using the technology for instru ction; particularly after having used and experienced computer-mediated instruction the first time specifically, once you have taken an online course the intimidation factor goes away and the next time(s) is is familiar and easier to accomplish. For the question on "I have the time needed to participate" perhaps the question could be worded to include a distinction between work and home time. Work gives me the time (if I can fit it into my workday my s upervisor suports my doing that, but taking a computer -mediated cpomuter course when th ere is other work to be done during the work day is sometimes difficult. Then, unfortunately, since computer-mediated continuing ed. has a 24/7 availablility, one ends up doing it at home instead of work.) PBC: #5 statement unlcear as to "It is possible what does that mean? PBC: Many of these items seem to assess factors related to actual" control (e.g., equipment, resources) or access as opposed to beliefs or perceptions. You may want to ask these questions based on the assumption that such courses are available, because they likely are to be in the futurel. Co nsider adding more items about perceived confidence to complete such courses, assuming they are available. PBC: Your instructions for this section are a little unclear -you might want to be careful about using terms like 't heir ability.' Who are they? ATT: I am strugglingly with availability of such pr ograms/population understanding what this type of instruction might mean (even when provided a definition)(e.g., need a practical application questi on since you are sending to SOPHE members, would it be possible to provide context in which this type of instruction might be usedThis comment is probably more general and is repeated for me throughout, but also reflects/impacts attitude

PAGE 227

219 Appendix O (Continued) ATT: those are fine ATT: My employer happens to support conti nuing education at work if it can be fit in, so it is possible for me to be positive in answering the above questions. That may not be the case for someone who does not have a support of their employer. Then if the continuing ed is important, you find the time to do the continuing ed. from other parts of your life (non-work hours). ATT: some statements are personal attitude nal statements and others appear to deal with attitudes about professional development (#1 and 3 in particular). #8 is a repeat of #3??? ATT: I know you have defined "sef-directed" and "CMI" but it may still be better to use "lay" terms in these questions. The "it w ould be easy" question might fit better in the perceived control section. I prefer the "first person" it ems over the general attitude items. ATT: I'm sure I'm not the first person to tell you that you've repeated an item. SN: Again, I think there is context missing-or it may be the terms used. I am not confident that SOPHE membership will understand what you mean when you state these items--while this is the technical term for what you are proposing, the words are daunting and may be bette r understood by your audience if you choose an easier phrase--something practical versus technical. SN: professional organizations varymay want to break them into various groupsalso some of these organizations ma y not be health education per saeyet provide great assistance. SN: Change "think I should" to "encourage me to" SN: All these questions seem to ask the same thing. But oddly enough the questions are hard to answer, because colleagues and those who influence me do not "push" computer-mediated instruction on me. They just support con tinuing education in general. Computer-mediated instruction is rather easy to do; you don't have to worry about parking, safety of driving to an unfamiliar neighborhood or in an ufamiliar building where conventional training might take place.

PAGE 228

220 Appendix O (Continued) SN: #1 it reads as if evryone is a memb er of than 1 professional organization -should this be changed? #2 and #6 --> what is difference between individuals who are important to me and individuals who influen ce my behavior? If you are trying to get at what others think about use of self-directed computer in struction for continuing ed (good idea #2 -a in general qu estion ) or whether they have others who think they should use (#6 very specifically related to me question), then OK. Just make sure you understand difference between individuals w ho are important to me and individuals who influence my behavior they may or may not be one in the same. SN: Like the attitude construct,it is hard to really grasp how you are operationalizing subjective norms. Is it how good/ba d SDCMI is? Valuable/valueless? Effective/uneffective? Rewarded /not-rewarded? Useful/useless? Importance/unimportance? etc. You might wa nt to think about this and decide the dimension(s) that are appropriate and revise your items accordingly. INT: this is true for allmany of these stat ements read very awkwardly because of the cumbersome terms used (self-directed, computer-mediated. I don't have an option for youit just appears to be one of those th ings that you'll need to be aware of INT: More concrete measures of intention ie I have checked out CMI, I am aware of organizational offering INT: Again, the questions above really ask the same thing, just in a different way. INT: Like the other TRA constructs, some of these items may be better asked as frequencies (never to often) or may need adjectives to weight the statements with agree/disagree answers if you are likely to get answer distributi ons that distribute normally. INT: Is time frame needed for #2 and #5. I think definitely #5 (During the next 12 months, I intend to take..) BEH: how would a person answer if part of a course was the use of cm programs? BEH: Level? Length of time ago? Number of experiences? Is 1 or greate classes experienced? BEH: I'd say"experienced" if someone had participated in 2 or more of the above classes or training programs.

PAGE 229

221 Appendix O (Continued) BEH: I expect that few people will have done any of these because there are not that many available. You may need to specify whether this includes taking a classroom based course with a web-based co mponent (e.g., discussion boards). BEH: I think you should also have "lack of programs" which I define differently than "Lack of availability of programs" a nd "Lack of access to information about programs"--An additional item might be CE not important for health education (some health educators, an older generation, do not participate in such activities) EXCE: See above. GEN: I think this is a good start, but may require some additions and revisions to better reflect the underlying di mension of interest. Additional Barriers to Include:l ack of programs (which I define differently than "Lack of availability of programs" and "Lack of access to information about programs"), CE not important for health educat ion, Lack of professional rewa rds,lack of relevant topics being taught,lack of interest in the medi um (prefer classroom-based courses), Few incentives for professional development.

PAGE 230

222 Appendix P. Scree Plot for Eigenvalues from Factor Analysis Figure 4. Scree Plot of Eigenvalues from Factor Analysis

PAGE 231

223 Appendix Q. Table of Cross Tabulations and Chi-Square Tests for Behavior Table 21. Respondent Characteristics by Behavior Behavior Overall Behavior Absent Behavior Present 2 Age 5.42 29 and Under 13% (N=62) 8% (N=36) 5% (N=26) 30-34 12% (N=59) 8% (N=40) 4% (N=19) 35-39 14% (N=71) 10% (N=49) 5% (N=22) 40-44 12% (N=56) 8% (N=38) 4% (N=18) 45-49 14% (N=67) 8% (N=38) 6% (N=29) 50-54 16% (N=75) 10% (N=49) 5% (N=26) 55-59 12% (N=57) 7% (N=33) 5% (N=24) 60 and over 6% (N=27) 4% (N=19) 2% (N=8) Gender 0.06 Male 19% (N=91) 12% (N=59) 7% (N=32) Female 81% (N=391) 51% (N=248) 30% (N=143) License/ Certification 9.59* Yes 63% (N=304) 37% (N=180) 26% (N=124) No 37% (N=179) 27% (N=131) 10% (N=48) Continued on next page

PAGE 232

224 Appendix Q (Continued) Table (continued) Behavior Overall Behavior Absent Behavior Present 2 Location for taking CME 6.20* Home 28% (N=131) 20% (N=93) 8% (N=38) Work 29% (N=136) 18% (N=84) 11% (N=52) Both Home and work 43% (N=200) 25% (N=115) 18% (N=85) Highest Level of Education 12.49* Bachelors 7% (N=34) 4% (N=23) 2% (N=11) Some graduate school 4% (N=18) 3% (N=12) 1% (N=6) Masters degree 50% (N=241) 28% (N137) 22% (N=104) Doctoral candidate 5% (N= 26) 4% (N=18) 2% (N=8) Doctorate or professional degree 34% (N=163) 25% (N=120) 9% (N=43) Continued on next page

PAGE 233

225 Appendix Q (Continued) Table (Continued) Behavior Overall Behavior Absent Behavior Present 2 Professional Identity 22.62* Community/public health educator 49% (N=201) 30% (N=123) 19% (N=78) School health educator 4% (N=15) 2% (N=8) 2% (N=7) Patient educator 3% (N=13) 2% (N=7) 1% (N=6) Health education researcher 12% (N=48) 8% (N=34) 3% (N=14) Univ/College teaching faculty 26% (N=106) 22% (N=89) 4% (N=17) Univ/college administrative faculty 6% (N=24) 3% (N=12) 3% (N=12) Continued on next page

PAGE 234

226 Appendix Q (Continued) Table (continued) Behavior Overall Behavior Absent Behavior Present 2 Professional Role 18.52* Administrative 22% (N=95) 13% (N=55) 9% (N=40) Service delivery 13% (N=55) 6% (N=28) 6% (N=27) Teaching 15% (N=66) 10% (N=46) 5% (N=20) Administrative and service delivery 18% (N=80) 11% (N=48) 7% (N=32) Administrative and teaching 16% (N=68) 10% (N=43) 6% (N=25) Service delivery and teaching 4% (N=18) 3% (N=15) 1% (N=3) Research 13% (N=57) 11% (N=47) 2% (N=10) Continued on next page

PAGE 235

Appendix Q (Continued) Table (continued) Behavior Overall Behavior Absent Behavior Present 2 Current Employer 34.85* College of University 47% (N=196) 35% (N=145) 12% (N=51) Hospital 7% (N=31) 2% (N=10) 5% (N=21) Non-hospital health care facility 3% (N=14) 2% (N=8) 1% (N=6) Insurance company/MCO 3% (N=13) 1% (N=6) 2% (N=7) Local health department 11% (N=45) 8% (N=33) 3% (N=12) State health department 9% (N=36) 5% (N=21) 4% (N=15) Federal government agency 10% (N=43) 5% (N=19) 6% (N=24) Non profit health education organization 9% (N=39) 5% (N=21) 4% (N=18) Note. Due to rounding, total percents may not equal 100. Response items with expected counts less than 5 were removed. *Indicates statistical significance at p<.05. 227

PAGE 236

About the Author Jane Ellery received her bachelors degree in Exercise Science from Purdue University in 1986 and an M.A. in Exercise Physiology in 1988. She worked for 10 years in health promotion and disease prevention before returning to school in 1998 to pursue a doctorate in Public Health. While in the Ph.D. program at the University of South Florida, Ms Ellery was an active participant in many projects including KIDS COUNT (Center for The Study of Childrens Futures), University of South Florida Social Marketing in Public Health Conference Planning Committee (Department of Community and Family Health), Thinking Like a Marketer (National Training Collaborative for Social Marketing), The Moffitt Real Time Patient Satisfaction Study (Moffitt Cancer Center), and Friendly Access (Lawton and Rhea Chiles Center for Healthy Mothers and Babies). She has made multiple paper presentations at National and International conferences, and she is currently working on developing manuscripts.


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 001416914
003 fts
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 031010s2003 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000052
035
(OCoLC) 52831445
9
AJJ4766
b SE
SFE0000052
040
FHM
c FHM
090
RA425
1 100
Ellery, Jane.
0 245
Delivering continuing education in health education using self-directed computer-mediated instruction
h [electronic resource] :
moving from intention to action /
by Jane Ellery.
260
[Tampa, Fla.] :
University of South Florida,
2003.
502
Thesis (Ph.D.)--University of South Florida, 2003.
504
Includes bibliographical references.
500
Includes vita.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 236 pages.
520
ABSTRACT: Using advanced technologies can help increase the availability of educational offerings; however, the steps taken in this direction must be appropriate for the target population and the specific content taught. As such, understanding factors that lead to health educators' intentions and behavior related to computer-mediated instruction for continuing education is an important step in developing and marketing appropriate computer-mediated instruction programs (Hoffman & Novak, 1994). Using the theory of planned behavior (Ajzen, 1988) this study explored the relationships between health educators' perceived behavioral control, attitudes, and subjective norms related to computer-mediated continuing education programs and their intentions to use, and previous experience with, computer-mediated education. Employing a cross sectional survey design, data were collected from 504 members of the Society for Public Health Education (SOPHE) (40% response rate) using an online survey instrument. Logistic regression was used to investigate the associations between attitudes, subjective norm, perceived behavioral control, and intention related to using computer-mediated continuing education programs and a proxy measure representing their computer-mediated continuing education behavior. Perceived behavioral control and attitudes were found to have significant associations with computer-mediated continuing education behavior, with intention partially mediating the association with perceived behavioral control and fully mediating the association with attitudes. When studying a subset of the group composed of respondents with a positive intention toward computer-mediated continuing education programs, respondent characteristics and barriers identified as distinguishing between individuals with positive and negative behaviors included perceived behavioral control, presence of a license or certification, a lack of programs, a lack of relevant topics for programs, and a lack of technical support for programs. These results suggest that for health education and health promotion professionals to engage in computer-mediated continuing education programs, more programs, especially ones that address topics relevant to their current functioning, need to be created and made readily available. Also, ensuring that appropriate technical support is available to assist participants, and informing potential participants of the availability of this technical assistance, may encourage more health educators and health promotion professionals to follow through on their intentions to participate in computer-mediated programs.
590
Adviser: McDermott, Robert R.
653
prefessional preparation.
computer-based training.
professional development.
health promotion.
distance learning.
690
Dissertations, Academic
z USF
x Public Health
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.52