Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. ... " Informal Relationships in the W orkplace: Associations with Job Satisfaction, OrganIsational Commitment and Turnover Intentions / A thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy In Psychology at Massey University, Albany, New Zealand. Rache1 Kathleen Lopes Morrison / . . -- . ---_ . . -- _ . . . . -- " 'iiliil l ii� 1 il iil ill /111 iiil� 1061756820 2005 1 Abstract Abstract The aim of this thesis was to develop and test a theoretical model of friendships in the workplace. Friendships within organisations may have it profound effect on an employee's experience of work, pofentially either hindering or facilitating organisational functioning, yet friendships have seldom been studied in an organisational context. The association between friendships at work and organisational outcomes such as job satisfaction, organisational commitment and turnover intentions were investigated in three studies, assessing support for a theoretical model. In the first study, employees of a large Auckland hospital I (n = 124) were surveyed using a written questioIUlaire. Results indicated that cohesiveness and opportunities for friendship were related to increased job satisfaction, leading to increased organisational commitment and decreased turnover intentions. The actual prevalence of friendships was primarily related to decreased turnover intentions. Overall there was good support for the proposed model. The need for further research to ascertain the generalis ability of the findings was highlighted. A second study was conducted using an Internet based questionnaire, accessed both from within New Zealand and worldwide. A diverse sample of employees responded (n = 412). The analysis (structural equation modelling) indicated further support for most aspects of the model, suggesting that the fmdings are generalisable and the model is robust. The model of workplace relationships was cross-validated in a third study, confmning linkages between friendships at work and organisational outcomes. The model was then tested for inter­ group invariance. The model was invariant across groups reporting differing needs for affiliation, autonomy and achievement, but non-invariant across groups occupying relatively less or more interdependent jobs. Results suggest that the interdependence of individuals' jobs affects the salience of work friendships more than respondents' subjective needs. Overall, the research suggests that the presence of workplace friendships has a significant effect on several workplace variables, with the effect of friendships being more salient for individuals in interdependent work roles. The implications of the research findings are potentially far reaching. Not only do workplace friendships improve employees' experiences of work, but they also have the potential to affect the fmancial "bottom line" through factors such as enhanced organisational commitment, job satisfaction and reduced intentions to leave. ) Waitemata District Health Board, Auckland (NZ) 11 Acknowledgements Acknowledgements I would like to thank my supervisors, Stuart Carr and Richard Fletcher for their help, support and invaluable feedback during this process. I am also grateful to Massey University for awarding the doctoral scholarship that made this project possible. In addition, the Massey University psychology department provided a grant to assist in travel to London in 2004, where parts of this project were presented at the Fourth Annual Conference on Emotions in the Workplace (PhD forum). I would like to acknowledge Waitemata DHB, for providing access to staff, many of whom took the time to fill out the initial survey, which began this entire project. In addition, the thesis could not have continued without those who generously gave their time to respond to the Internet based survey. Finally I would like to thank my family and friends, particularly my husband Justin Lopes, for listening, supporting and understanding. Thanks also to Karren Towgood and Rhiannon Herrick for proofreading. Note: Parts of this thesis have been published in the New Zealand Journal of Psychology: Morrison, R. (2004). Informal relationships in the workplace: Associations with job satisfaction, organisational commitment and turnover intentions. New Zealand Journal of Psychology, 33(3), 1 1 4-1 28. Parts of this thesis have been published in The Bulletin: Morrison, R. (2004). Gender comparisons in workplace friendships. The Bulletin(102), 3 5-40. 111 Table 01' Contents Table of Contents Chapter 1: Introduction .....•................. . .........•....................•...............•.......................... 1 Chapter 2: Literature Review ............................••... � ...........•.......................................... 6 2.1 Where does this Project Sit Within Psychology? ........................................... 6 2 . 1 . 1 Social Psychology: A History of Friendship Research .. . . . . ... . . . . ....... . .. .. . . ... 6 2 . 1 .2 Organisational Psychology: A History of Human Factors Research .... . . . . . . 9 2 . 1 .3 The Current Research . . . . . . . . . . . ... . . . . . . . .. . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . .. . 1 2 2.2 What is it to be a Friend? ............................................................................... 14 2 .2 . 1 Friends Compared to Other Relationships .. . . ... ... . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . ... . . . . . 1 5 2.2.2 The Nature of Friendship . . . ... . . . . . . . . . . . ..... . . . . . . . . . ... .. . . . . . . . . . . .. . . . . . . . . . . . .. . . . .. . . . . . . . . 1 7 2.2.3 Gender Differences in Friendship . . . . . . . . . ... . . . . . . . . . ... ... . . . ...... .. ...... . . .... . . . . . . .. . . . 1 9 2.2.4 Why Do People Engage in Friendships? ... . . .. . . . . . . . . . . . . . .. . . . . . . . ... . . . . . . . . . . . . . . . . . . . 2 1 2.3 Relationships in a Workplace Context .......................................................... 24 2.3 . 1 Types of Organisational Peer Relationships . . . ..... . . . .. . . . ....... .... . . . . . .. . . . . . . . . .. . 25 2.3.2 The Informal Structure of Organisations .... . . . . . . . . . ... . . . . ... . . . . . .. . . . . . . . . . . . .. . . .. . . . . 29 2.3 .3 The Impact of Friendships in the Workplace . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . . . 3 1 2.3 .4 The Impact of the Organisation on Friendships ... . . . . . . . . . . . . ..... . . .. . . . . . . . . . . . . . ... 50 2.3 .5 Friendship Prevalence ........ . . . . . . . . . . . . . . . . . . ....... . . ..... . ... ...... ....... ... . . ....... . . . . . . . ... 57 2.3 .6 Gender Differences in Organisational Relationships .. .. . . ... . . . . .. . . . . ... . . . . . . . . .. 6 1 2.3 .7 Negative Relationships . . . . . . . . ... . . ...... . . . . . . . . . . . .. . . . ..... . . . . . . . .. . .. . . . . . . . ... . . ... . . . . . ...... 63 2 .3 .8 Blended Relationships: Friends as Work Associates ... . ......... . . . . .... . . . . . . . . . . . 66 Chapter �: Testing the Model in a New Zealand Hospital ......................•.......•......•.. 70 3.1 Introduction ..................................................................................................... 70 3.1 . 1 A New Zealand Hospital as a Setting for Organisational Research . . . . . . . . . . 70 3 . 1 .2 Research Questions .... . ................ . .... . ....... . . .-. . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . ......... 72 .3 . 1 .3 Research Methodology . . . ..... . . .. . . . . . . ........ . . . . .. . . ... . ..... . . . ... . . ..... . ...... . .... . . ... .. . .. 73 3.2 Method .............................................................................................................. 74 3 .2.1 Participants .... . . . . . . . . . ...... ..... . . .. . . .. . ...... . .. . . . .. . . . .. .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . .. . . . 74 3.2.2 Materials . . . . ...... .. . . ... .. . . . .. . . . .... � . . . . . ......... . . . . . ..... ... .. . ..... . ... . . . . . . ... . ... . .. . . . . . . . . . . . . . 75 3.2.3 Procedure ... . . . . . . . . . . .. . . ...... . ... . . . . ... . .. . . . ... ... ... . . . . ..... . . . . . . . . . . .. . .. . . . ... . . ... . ... . . . . . . . .. . . 82 iv Table of Contents 3.3 Analysis and Results ................................................................................•....... 83 3.3 . 1 Imputing Missing Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . ..... . . . 83 3.3.2 3.3.3 3.3.4 Factor Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . ... . . . . . . . . . . . . . 83 Path Analysis . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . ... ... . . .... . . .. . . . . . . . . 95 Peer Types . . . . . . . . . . . . . . . . . . . . . . . . ... . ........... . ...... . ...... . ... . . . . . . . ... . . . ... . . . . . . . . .... . . . .. . . . . . 1 03 3.4 3.4. 1 Discussion ... . .................................................................................................... 1 04 3.4.2 3.4.3 3.4.4 3.4.5 3.5 3.6 . Research Question 1 . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . .. ... . . 1 04 Research Question 2 . . . . . . . . . . . . . . . . . . . . . . . . : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... . . . 1 09 Research Question 3 . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 2 Job Satisfaction and Commitment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . .. . . . . . ... . . . . . . ... .. 1 1 4 Satisfaction/Commitment and Turnover . . . . .... . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . .... . . 1 1 5 Limitations of the Study and Directions for Future Research ........... ....... 115 Conclusions and Implications ...............................................................•.....•. 117 Chapter 4: Internet Study ......................................................................................... 120 4.1 Introduction ................................................................................................... 120 4. 1 . 1 Gender Differences in Friendships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . ... 1 2 1 4.2 Method ....................................................................................................•....... 122 4.2 . 1 Participants . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . .. . . . . . . ........ .. . . . . . .... . .. . . . . . . .. 1 22 4.2.2 Materials . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . .................... ........... . . . .... . . . . . . . . .... . . ... . . . . . . . . ... 1 24 4.2.3 Procedure . . . . . . . . . . . . . . . . . . . .. . . . . . . . .. . . . . . . . . . ... . . . . . . . . . . . . . ... . .. ... . . . . . . . . . . . .. .. . . . . . ... ....... ... 1 24 4.3 . Analysis and Results ...........•...........................................•................•....•.......• 125 4.3.1 Factor Analysis .. . ... . . . .... . . . . . . .... ·. . . . . . . ...... . . . ... ... . .... ... . . . ... . . . .......... .. . . . . .......... 1 25 4.3.2 Building the Causal Model.. .. . . ... . . . . . . . .... . . . . . ... . . . . . ........ .. . . ..... . ... ..... . . ....... . . 1 43 4.3 .3 Statistical Significance of the Model.. . ... ... . . . . . ... . . . ... .... ... ......... . . .... ..... .. . . . 1 45 4.3 .4 Other Research Questions .. ..... ... . . .... . .. .. . . . . . . . . ........ . . . ... ... . . ........... ... . .... . . .. 1 49 4.4 Discussion ....................................................................................................... 152 4.4.1 Resume of Findings . . . . . . ... . . . . ... . . . . . . . . .. . . . ... ... ... . . . . .. . . . . . .... .. . . . . . ...... ... . . ...... . ... 1 52 4.4.2 Links to Existing Research . . . . . . . . . . . . ... . . . . . . . . . . . ...... . .... . ... . . . . . ..... . . .............. ... 1 57 4.5 Limitations of the Study and Directions for Future Research ................•. 162 v Table of Contents 4.6 Conclusions and Implications .............................. � ........................................ l63 Chapter 5: Cross-Validating the Causal Model and Testing for Invariance Between Sub-groups Differing in Need Strength and Interdependence of Job ......................... 165 5.1 5 .1.1 5 . 1 .2 5.2 5 .2.1 Introduction ................................................................................................... 165 Needs for Affilia�ion, Autonomy and Achievement . . . . . .. . . . . .. . . . .... . . . . ... . ... 166 Interdependence of Work Role . . . .. . ..... . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . .... . . . ... . . . 1 71 Method ............................................................................................................ . 174 Participants . . . . ..... ..... ... . . . . . . . . . . ... . . .. . . . . ... .. . . . . ... . . . ..... . . ... . . . . . . . . . . . . . .... . . . . . . .... . . . . 174 5 .2.2 Materials . . . . . . . . . . . ... . . .. . . ... .. . ... . . . . . . . . . .. . . . . . . ... . . ... . .. . . . . . . . .... . . . . . ........ . . . . . . . ... . . . . ... 175 5 .2 .3 Procedure .. . .. . . . . . . . . . . . . . . ... . . ... . . . . � . . ... . . . . .. . . .. .. . . . . .. . . .. . .. . . . .. . . . . . . ... . . . . . . .. . . . ... . . . . . . . 178 5.3 Analysis and Results ...................................................................................... 179 5 .3 . 1 Measurement Models of the Scales . . . . . . ... . . . . . .... .. . . . .. . . ... . . . . . . . . ... . .. . . . . . . . . . . . . . 180 5 . 3 .2 Model ofWorkplace Relationships . . ... . ..... . . . . . . . . . . . . . . .. ..... . . . . . . . . . . . . . . . ...... . . . . 180 5 .3 . 3 Comparing Groups . . .. . . . . . . . . . .. ... . . .. . . . ... . . . . . .. . . . . . . ..... . . ... . .. . . . .... .. . . . . . .... . .. . . . .... 182 5.4 Discussion ....................................................................................................... 189 5.5 Conclusion ...................................................................................................... 190 Chapter 6: Synthesis and Implications ..................................................................... 192 6.1 Summary of Major Findings ........................................................................ 192 6. 1 . 1 Hospital Study ... . . . .... ... ..... . . . . . . .. .. . . . . . . . . . . . . . . . . . . . .. . . . ... . . . . ... . . . ... . ... . . .. . . . . .. . . . . . . . 192 6. 1 .2 Internet Study . .. . ... .. . .... . . . . ... . ... ... . . . ... . .... . .. . .... ... . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . .... . . ... 194 6.1 .3 Invariance Testing Study . . ... . . . . . .. . .. ..... . . . . . . ... . . . . . .... . . . . . ... . . . . . . . . . . ... .. .... . . ..... . 195 6.2 Synthesis ......................................................................................................... 196 6.2 . 1 Friendship Opportunities versus Friendship Prevalence . . . . .. . . . . . . . . . . .... . .... 196 6.2.2 Interdependence of Work Role versus Individual Differences in Needs 198 .6.2.3 Gender Differences . . . . . . . . . . . . . .. . . . . . .... . . . . . . . . ... .... . ... .. . . ... ... . . . .... ........ . . . . . ...... ... 199 6.3 Future Directions ........................................................................................... 202 6.3 .1 Performance ... ...... . . . .. . .......... . .. . . . . . . . . . . . ..... . . . . .... ... . . ... . . . . . . . . ... . . .... . . . . . . . . . .. .. ... 202 6.3 .2 Negative Relationships . . . ... ... . . . . . . . . .. . . . .. .... . . . . .... . . . ... . . . .. . .. . . . . : . . . . . . . . . . .. . . . ... . . . 202 6.3.3 Organisational Identity and Attributions of Responsibility .. . ... . . . . . . . ... .... 203 VI Table of Contents 6.4 Implications .......................................................... � ......................................... 207 6.4. 1 Enhancing the Quality of Work Life . . . .. . . . . ... . . . . ..... . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . 207 6.4.2 Situations Where the Findings Might have More Relevance . . . . . . . . . . . . . . . . . . 209 6.4.3 Possible Negative Implications . . . . . ... . . . . . . . .... . . . . . . . . ... . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 2 1 0 6.5 Conclusion ......................................................................... . . . . . . . . . .................... 2 12 References ........................................................................................................................ 213 Appendices ...................................................................................................................... 232 Appendices for the Literature Review (Chapter 2) ................. . . . . . . . . . . . . . . .. .. . ............ 232 Appendices for the Hospital Study (Chapter 3) . ...... .. . . . ............ . . . . . . . . . . ................... 236 Appendices for the Internet Study (Chapter 4) . .. .... . . . . . . ............. . . . . . . . . .. .. . . . . . . . . . . ...... 266 Appendices for the Invariance Testing Study (Chapter 5) .................................... 294 vu List of Tables List of Tables Table 1 : Demographics of Respondents of the Hospital Study ...................................... 75 Table 2: First Pattern Matrix of the Factor Analysis of the Friendship, Cohesion and Dual Role Tension Measures (items to be removed are shaded in grey) ................ 86 Table 3: Final Pattern Matrix of the Factor Analysis of the Friendship, Cohesion and Dual Role Tension Measures .................................................................................. 88 Table 4: First Pattern Matrix of the Factor Analysis of the Satisfaction, Commitment and Leaving Intention Measures (items to be removed are shaded in grey) ........... 9 1 Table 5: Final Pattern Matrix of the Factor Analysis of the Satisfaction, Commitment and Leaving Intention Measures .................................. ........................................... 93 Table 6: Correlations between Measured Variables: Hospital Study ............................. 94 Table 7: Showing Critical Ratio (C.R.) Values (parameter estimate divided by standard error) of the Regression Paths in Model 1 ............................................................... 96 Table 8: Showing Critical Ratio Values (parameter estimate divided by standard error) of the Regression Paths in Model 1 When Specified with Non-Significant Regression Paths Removed ..................................................................................... 97 Table 9: Demographic Data for the Internet Study ...................................................... 1 22 Table 1 0: Industry of Respondents ........................................................................ ....... 1 23 Table 11: Initial Pattern Matrix for the Workplace Friendship Scale (items to be removed are shaded in grey) ................................................................................. 1 28 Table 12 : Final Pattern Matrix for the Workplace Friendship Scale ........................... 1 28 Table 1 3 : Initial Pattern Matrix for the Job Satisfaction Scale (items to be removed are shaded in grey) ...................................................................................................... 1 3 1 Table 14: Final Pattern Matrix for the Job Satisfaction Scale ...................................... 1 3 1 Table 1 5 : Pattern Matrix. for the Cohesion Scale ......................................................... 1 3 7 Table 16: Initial Pattern Matrix for the Needs Assessment Questionnaire (items to be removed are shaded in grey) ................................................................................. 1 40 Table 1 7: Final Pattern Matrix for the Needs Assessment Questionnaire ................... 1 40 Table 1 8 : Fit Indices for the Measurement Models: Internet Study ............................. 1 42 Table 1 9: Correlat�ons between Measured Variables: Internet Study .......................... 1 43 Table 20: Alpha Levels of the Sub-Scales ................................................................... 1 43 viii List of Tables Table 2 1 : Showing Critical Ratio Values (parameter estimate divided by standard error) ofthe Regression Paths in Model 2 When Specified with Non-Significant Regression Paths Removed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 Table 22: Fit Indices for the Causal Model: Internet Study ........... .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . 148 Table 23 : Correlations between Subscales, Showing those where Measured Variables are Differentially Correlated for Men and Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . .. ... . .. .. .. . 149 Table 24: Independent Samples T-test, Comparing Respondents With and Without Negative Relationships at Work . . . . . . . . .. . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 1 Table 25 : Hypothesised Relationship between Occupation, Affiliation Opportunities and Opportunities to Acquire Career Capital. . . . . . .. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 7 1 Table 26: Breakdown o f Demographic Data for Respondents to Cross-validation / Invariance study .................................................................................................... 175 Table 27: Needs Assessment Questionnaire Items Used in the Current Study . . . . . . . . . . . . 1 77 Table 28 : Fit Indices for the Measurement Models (n = 445) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 80 Tcibie 29: Fit Indices for the Causal Model : Based on the Full Sample (n = 445) . . . . . . 1 8 1 Table 30: Chi-Square Statistics for Tests of Invariance across Sub-groups of the Sample .......................... � ... ................... ...... ..................... . .......................... ... ... .......... ........ 1 82 Table 3 1 : Goodness of Fit Statistics for Tests of Invariance Across Those in Relatively Less and More Interdependent Jobs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 84 Table 32: Goodness of Fit Statistics for Tests of Measurement Model Invariance Across Those in Relatively Less and More Interdependent Jobs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 86 Table 33: Showing Correlation Coefficients and Critical Ratio Values (parameter estimate divided by standard error) of the Correlation Between the Cohesion (Social Support) and Friendship Opportunity Variables for High and Low Interdependence Groups . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 88 Table 34: Goodness of Fit Statistics for the High and Low Interdependence Groups. 1 89 ix List of Figures List of Figures Figure 1: Showing the hypothesised relationship between friendship opportunities and job satisfaction . . . : . .. . . . . . . ........ . . . . .. . . .. . . . .. . . . . . .. . ..... .. ... ... . . .. . .. . ..... . ....... ................ ........... 3 8 Figure 2: Showing the hypothesised relationship between job satisfaction and organisational commitment . . . . . . .. . . . . . . . . . .. . . ... . . . . .. . . . . . . .. . . . . . . . . . . .. . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. .. 42 Figure 3: Showing the hypothesised relationship between friendship opportunities, job satisfaction and organisational commitment .. . . . . . . . . . . . . .. . . .. . . . . . . ... ... . . . .. . .. . . . . . . . . . . . . . . . . . . . 46 Figure 4: Showing the hypothesised relationship between job satisfaction, commitment and intention to leave . .. .. . . . . . .. . .. . . .. . . . . . .. . . . . . . . .. ... .. .. .. ........ . . .... ... . . . . . . . . .. . .. .. ... . .. . .. . .. . . . . . . 48 Figure 5: Showing the hypothesised relationship between friendship opportunities and cohesion ... .. . . . ......... ...... . . . . . . . . . . . . . . . . ..... . .. .. .. . . . . . . . . . . . . . . . . . . .. . . . . . .. . . . . .. . . .. . .. . . .. .. . . . . . . . . . . . .. . . . . . 54 Figure 6: Showing the hypothesised relationship between cohesion and intention to leave . . . . ... . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . .... . . 56 Figure 7: Showing the hypothesised relationship between cohesion and organisational commitment and job satisfaction . ..... ... ... . ... .. . . .. . . . . . . . . . . . . . . . . .. . . .. . . .. . . . ... ...... . . . . . . . . . . . . . . . . . 57 Figure 8: Showing friendship prevalence as the result of increased friendship opportunities and cohesion . ..... .... . . ..... . ... ....... ...................... .................... ..... . .. . .. . .... 58 Figure 9: Showing the hypothesised relationship between friendship prevalence and intention to leave . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . .... . . .. .. . . . . . . . . . . . .. ... . . . . . . . . . . ... . . . . 60 Figure 1 0: Showing negative relationships added to the model. . . . . ... ... . . . . . . . .. . . . . . . . . . . . . . . . . . 66 Figure 1 1: Showing the hypothesised moderating effect of dual role tension (Model 1 ) ................................................................................................................................. 69 Figure 1 2: Path analysis of Model 1 .. .. . ... ...... . .. . . . . . . ... . ..... .. . . . .. .. . . ... .. . . ..... . .... . . .. . . . . . . . . .. . . . . . 98 Figure 1 3: Path from friendship opportunities and cohesion to organisational commitment . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Figure 14: Path from job satisfaction and organisational commitment to intention to leave . ... ... .. . . . . .. .... . . . . . . . . . . . . . . . . .. ... .. . . . .. . .. .. . .. ..... .. .. . . ... . . . . . . . . . . . .. . . ... . ... .. ..... .... ... . . . . . . . . . . . . . . 1 00 Figure 1 5: Path from friendship opportunities and cohesion to friendship prevalence 1 00 Figure 1 6: Path from cohesion to intention to leave .. . . . . . . . . .. . . . .. .. . ... .. .. . . . ... . .. . . . . ... . . . . . . .. . 1 0 1 Figure 1 7: Path from friendship opportunities to intention to leave . . . . . . . . . . .. . . . . .. .. . .... ... 1 02 x List of Figures Figure 18: Model 1 redrawn showing only the significant regression weights (Hospital Study) .................................. .................................................................................. 103 Figure 19: AMOS diagram for friendship scale .......... : .............................................. .. 130 Figure 20: AMOS diagram for the job satisfaction scale ... . . ........... .. .. .................... . .... 133 Figure 21: Scree plot for Organisational Commitment Questionnaire ........... ........... . .. 134 Figure 22: AMOS diagram for the Organisational Commitment Questionnaire ......... 136 Figure 23: Scree plot for the cohesion scale ......................... ....................... ....... .......... 136 Figure 24: AMOS diagram for cohesion scale ............ ................................. ..... ........ . .. 13 8 Figure 25: AMOS diagram for the Needs Assessment Questionnaire ........... .............. 141 Figure 26: The causal model tested for the Internet Study (latent variables only) . ..... 144 Figure 27: Model 2, SEM results of Internet Study .......................................... ...... ..... 146 Figure 28: Results of SEM analysis of the theoretical model. Values represent standardised estimates and correlations based on the full sample (n = 445) . .. . .... 181 Figure 29: The hypothesised mediating effect of identification and attribution, on the relationship between friendship prevalence and job satisfaction . .................. .. . . . .. 205 Figure 30: Alternative hypothesised mediating effect of identification and attribution, on the relationship between friendship prevalence and job satisfaction . ........ . . . . .. 206 .. . '\.! ' , ' Xl