Adapting to the work life interface the influence of individual differences, work and family on well being, mental health and work engagement

Adapting to the work-life interface: The influence of individual differences, work and family on well-being, mental health and work engagement.. Keywords Bronfenbrenner, dispositional op

Adapting to the work-life interface: The influence of individual differences, work

and family on well-being, mental health and work engagement

By

Prudence M R Millear

B Sc Ag (Hons), Grad Dip Psych, B Psych (Hons)

A thesis submitted in fulfilment of the requirements of the degree of

Doctor of Philosophy

School of Psychology and Counselling

Faculty of Health Queensland University of Technology

February 2010

Keywords Bronfenbrenner, dispositional optimism, coping self-efficacy, affective commitment, skill discretion, job autonomy, life satisfaction, psychological well-being, mental health, work engagement, burnout, longitudinal modelling, gain spirals, loss spirals, Conservation of Resources, resource caravans, working adults

Statement of Original Authorship

The work contained in this thesis has not been previously submitted to meet

requirements for an award at this or any other higher education institution To the best of my knowledge and belief, this thesis contains no material previously

published or written by another person except where due reference is made

Signature ………

Date ………

Publications and presentations arising from the PhD research

Journal articles

1 Millear, P.M & Liossis, P.L., Gain spirals and resource caravans: An integrated

longitudinal model of well-being, mental health and work engagement among Australian workers, under review, Journal of Occupational and

Organizational Psychology

Book Chapters

1 Millear, P.M & Liossis, P.L (2010) Longitudinal modelling of individual

differences and the workplace: well-being and work engagement Chapter 18

in Hicks, R.E (Ed.) Personality and Individual Differences: Current

Directions Brisbane, Australia: Australian Academic Press

2 Millear, P.M & Liossis, P.L Doing it for yourself: The choices and strategies of

managing the work-life challenge, accepted for publication, Wayfinding through life‟s challenges: Coping and survival, Nova Science Publishers, NY,

K Gow & M Celinski (Eds)

Conference Presentations

2008 European Academy of Occupational Health Psychology conference at the

University of Valencia, November 2008, 2 presentations: 1 Longitudinal

modelling of well-being and mental health in Australian workers; 2 Exploring

burnout and work engagement in diverse occupations: A continuum or two separate factors?

2008 Australian Conference for Personality and Individual Differences (ACPID),

Bond University, (November, 2008) “Longitudinal modelling of the influence

of individual differences and the workplace on well-being and work

engagement”

2009 8th Industrial and Organizational Psychology conference, Sydney (June

2009), Paper Presentation, “An integrated longitudinal model of well-being, mental health and work engagement among Australian workers”

Cameron particularly for deciphering the process of longitudinal modelling I would like to thank my postgraduate friends for their unstinting support, coffee and

sympathy and thank my family and friends for helping where they could

My greatest thanks are to my husband and children for bearing with me and understanding the work involved in completing my thesis, in the middle of family life, rugby and the house renovations We have done this together

Abstract Bronfenbrenner‟s Bioecological Model, expressed as the developmental equation, D

f PPCT, is the theoretical framework for two studies that bring together diverse strands of psychology to study the work-life interface of working adults

Occupational and organizational psychology is focused on the demands and

resources of work and family, without emphasising the individual in detail Health and personality psychology examine the individual but without emphasis on the individual‟s work and family roles The current research used Bronfenbrenner‟s theoretical framework to combine individual differences, work and family to

understand how these factors influence the working adult‟s psychological

functioning Competent development has been defined as high well-being (measured

as life satisfaction and psychological well-being) and high work engagement (as work vigour, work dedication and absorption in work) and as the absence of mental illness (as depression, anxiety and stress) and the absence of burnout (as emotional exhaustion, cynicism and professional efficacy)

Study 1 and 2 were linked, with Study 1 as a cross-sectional survey and Study

2, a prospective panel study that followed on from the data used in Study1

Participants were recruited from a university and from a large public hospital to take part in a 3-wave, online study where they completed identical surveys at 3-4 month intervals (N = 470 at Time 1 and N = 198 at Time 3) In Study 1, hierarchical

multiple regressions were used to assess the effects of individual differences (Block

1, e.g dispositional optimism, coping self-efficacy, perceived control of time,

humour), work and family variables (Block 2, e.g affective commitment, skill

discretion, work hours, children, marital status, family demands) and the work-life interface (Block 3, e.g direction and quality of spillover between roles, work-life

balance) on the outcomes There were a mosaic of predictors of the outcomes with a group of seven that were the most frequent significant predictors and which

represented the individual (dispositional optimism and coping self-efficacy), the workplace (skill discretion, affective commitment and job autonomy) and the work-life interface (negative work-to-family spillover and negative family-to-work

spillover) Interestingly, gender and working hours were not important predictors The effects of job social support, generally and for work-life issues, perceived

control of time and egalitarian gender roles on the outcomes were mediated by

negative work-to-family spillover, particularly for emotional exhaustion Further, the effect of negative spillover on depression, anxiety and work engagement was

moderated by the individual‟s personal and workplace resources

Study 2 modelled the longitudinal relationships between the group of the seven most frequent predictors and the outcomes Using a set of non-nested models, the relative influences of concurrent functioning, stability and change over time were assessed The modelling began with models at Time 1, which formed the basis for confirmatory factor analysis (CFA) to establish the underlying relationships between the variables and calculate the composite variables for the longitudinal models The CFAs were well fitting with few modifications to ensure good fit However, using burnout and work engagement together required additional analyses to resolve poor fit, with one factor (representing a continuum from burnout to work engagement) being the only acceptable solution Five different longitudinal models were

investigated as the Well-Being, Mental Distress, Well-Being-Mental Health, Work Engagement and Integrated models using differing combinations of the outcomes The best fitting model for each was a reciprocal model that was trimmed of trivial paths The strongest paths were the synchronous correlations and the paths within

variables over time The reciprocal paths were more variable with weak to mild effects There was evidence of gain and loss spirals between the variables over time, with a slight net gain in resources that may provide the mechanism for the

accumulation of psychological advantage over a lifetime The longitudinal models also showed that there are leverage points at which personal, psychological and managerial interventions can be targeted to bolster the individual and provide

supportive workplace conditions that also minimise negative spillover

Bronfenbrenner‟s developmental equation has been a useful framework for the current research, showing the importance of the person as central to the

individual‟s experience of the work-life interface By taking control of their own life, the individual can craft a life path that is most suited to their own needs Competent developmental outcomes were most likely where the person was optimistic and had high self-efficacy, worked in a job that they were attached to and which allowed them to use their talents and without too much negative spillover between their work and family domains In this way, individuals had greater well-being, better mental health and greater work engagement at any one time and across time

Table of Contents

Keywords iii

Statement of Original Authorship v

Publications and presentations arising from the PhD research vii

Acknowledgements ix

Abstract xi

Table of Contents xiv

List of Tables xix

List of Figures xxii

List of Appendices xxiii

Chapter 1: Theories and literature review of Bronfenbrenner‟s developmental equation, applied to individuals and the work-life interface 1

1.1 Bronfenbrenner‟s Bioecological Model of Human Development 2

1.2 Theories for D, the developmental outcomes, defined by well-being, mental health, burnout and work engagement 9

1.2.1 Well-being 9

1.2.1.1 Prevalence 12

1.2.1.2 Stability of well-being 14

1.2.1.3 Australian health and working provisions 14

1.2.2 Mental health, as the absence of mental illnesses 15

1.2.2.1 Costs and prevalence 18

1.2.3 Burnout and work engagement 22

1.2.4 Bringing together well-being, mental health, burnout and engagement 26

1.3 Understanding the person, P, in the developmental equation 29

1.3.1 Generative dispositions and demand characteristics 29

1.3.2 Theories of the generative disposition of P, the person occupying and managing multiple roles 30

1.3.3 Linkages between the generative disposition and positive affect, positive psychology and resilience 34

1.3.4 Gender and the generative disposition of the active participant, P 39

1.3.5 Gender 40

1.3.5.1 Gender and the work environment 41

1.3.5.2 Gender and parenting 46

1.3.5.3 Gender and house work 48

1.3.6 Dispositional optimism 50

1.3.7 Self-efficacy, as coping self-efficacy 56

1.3.8 Perceived control of time 58

1.3.9 Theories of the demand characteristics of P, the person occupying and managing multiple roles 60

1.3.10 Humour 64

1.3.11 Social skills and relationships 69

1.3.12 Conclusion for P, the Person 73

1.4 Understanding C, the Context for multiple roles 75

1.4.1 Theories and models of C, the Context for multiple roles 75

1.4.2 Direction for the literature review of C, the context 80

1.4.3 Working hours and schedules 82

1.4.4 Demands and resources 87

1.4.5 Affective commitment 93

1.4.6 Managerial support of work-life issues 95

1.4.7 Family characteristics 99

1.4.8 Multiple roles and spillover 104

1.4.9 Exploring the interactions between work and non-work domains 106

1.4.9.1 Comparing types of jobs 108

1.4.9.2 Importance of roles 110

1.4.9.3 Individual factors 112

1.4.9.4 Workplace and family factors 113

1.4.10 Exploring work-life balance and work-life fit 120

1.4.11 Conclusions of the Context of the work-life interface 122

1.5 T, the time frame over which multiple roles develop and occur 124

1.5.1 Longitudinal studies from a developmental perspective 130

1.5.2 Longitudinal studies from an organizational perspective 136

1.5.3 Conclusions for Time in the developmental equation 142

1.6 Proposed research program 143

1.6.2 Study 1 143

1.6.3 Study 2 145

Chapter 2, Study 1: Using hierarchical multiple regressions to explore the predictors of well-being, mental illness, burnout and work engagement of working adults 147

2.1.1 Hypothesis for Study 1 148

2.2 Methods 149

2.2.1 Participants 149

2.2.1.1 Recruitment 149

2.2.2 Internet survey development 150

2.2.3 Internet survey methodology 151

2.2.4 Measures 155

2.2.4.1 Demographics 155

2.2.4.2 Schedules, education, job conditions and income 156

2.2.4.3 Work-life fit, work-life balance, feeling busy and personal problems 157

2.2.5 Reliabilities and details of the measures 157

2.2.6 P, the Person: Generative disposition variables 158

2.2.6.1 Dispositional optimism 158

2.2.6.2 Coping self-efficacy 158

2.2.6.3 Control 159

2.2.7 P, the Person: Demand characteristic variables 161

2.2.8 C, the Context: Workplace conditions 161

2.2.8.4 Managerial support for work-life issues 162

2.2.8.5 Affective commitment 162

2.2.9 C, the Context: The work-life interface 163

2.2.10 Well-being, mental illness, burnout and work engagement 164

2.2.10.3 Satisfaction with life domains 165

2.2.10.5 Burnout 166

2.2.10.6 Work Engagement 166

2.2.11 Procedure 167

2.2.12 Analytical strategy for the hierarchical multiple regression (HMR) analyses 169

2.3 Results 174

2.3.1 Data cleaning and screening 174

2.3.2 Demographics 176

2.3.3 Scale construction and sample size 179

2.3.4 Means, standard deviations and correlations between the variables 180

2.3.5 Presentation of the results of the HMR 189

2.3.6 Life satisfaction 191

2.3.7 Psychological well-being 193

2.3.8 Satisfaction with work 194

2.3.9 Work vigour 198

2.3.10 Work dedication 200

2.3.11 Work absorption 201

2.3.12 Depression 205

2.3.13 Anxiety 207

2.3.14 Stress 210

2.3.15 Emotional exhaustion 212

2.3.16 Cynicism 215

2.3.17 Professional efficacy 218

2.3.18 Summary of the significant predictors of the hierarchical multiple regressions 220

2.3.19 Post-hoc analysis: Examining moderation between the most common predictors for the outcomes 224

2.3.20 Post-hoc analysis: What happened to humour? 230

2.3.21 Post-hoc analysis: An examination of gender 232

2.3.22 Post-hoc analysis: What predicts negative spillover? 234

2.3.23 Post-hoc analysis: Understanding positive spillover 236

2.4 Discussion 238

2.4.1 Limitations and strengths of Study 1 255

Chapter 3, Study 2: Longitudinal modelling 257

3.1.1 Hypothesis for Study 2 259

3.2 Methods 259

3.2.1 Participants 259

3.2.2 Recruitment of participants, survey methods and materials 260

3.2.3 General process for longitudinal modelling 260

3.2.4 Introduction to SEM and associated terminology 261

3.2.5 Assessing model fit 263

3.2.5.1 Normed Chi-Squared statistic 264

3.2.5.2 Root Mean Square Error of Approximation (RMSEA) 265

3.2.5.3 Akaike Information Criteria (AIC) 267

3.2.5.4 Comparative Fit Index (CFI) 267

3.2.5.5 Expected Cross-Validation Index (ECVI) 268

3.2.6 Early SEM models 269

3.2.7 Confirmatory Factor Analysis (CFA) 270

3.2.8 Models to be considered in the CFAs and for longitudinal modeling 271

3.2.9 Constructing composite variables for the longitudinal models 273

3.2.10 Naming the composite variables 275

3.2.11 Calculations of the composite variables 276

3.2.12 Analytical strategy for longitudinal modelling 276

3.2.12.2 Model trimming 280

3.2.13 Summary of methods used for the longitudinal modeling 281

3.3 Results of the Longitudinal Modeling 282

3.3.1 Sample size and characteristics 282

3.4 Time 1 SEMs as a basis for longitudinal models 286

3.5 Confirmatory factor analyses (CFAs) 288

3.5.1 Confirmatory factor analysis of Well-Being model 289

3.5.2 Factor Score Weights for Well-Being model 290

3.5.3 Confirmatory factor analysis of the Mental Distress model 292

3.5.4 Factor score weights for the Mental Distress model 294

3.5.5 Confirmatory factor analysis for the Well-Being-Mental Health model 295 3.5.6 Factor score weights for the Well-Being – Mental Health model 297

3.5.7 Confirmatory factor analysis for the Work Engagement model, based on the scales of burnout and work engagement 297

3.5.8 CFA for Burnout and Engagement alone 299

3.5.8.1 One-factor CFA 299

3.5.8.2 Two factor CFA 300

3.5.9 Confirmatory Factor Analysis for the Work Engagement model 301

3.5.10 Factor score weights for the Work Engagement model 302

3.5.11 Confirmatory factor analysis of the Integrated model 303

3.5.12 Factor score weights for the Integrated model 308

3.6 Comparing the longitudinal models 310

3.6.1 Competing sets of longitudinal models 310

3.6.2 The longitudinal Well-Being Model 313

3.6.3 The longitudinal Mental Distress model 316

3.6.4 The longitudinal Well-Being – Mental Health model 318

3.6.5 The longitudinal Work Engagement model 319

3.6.6 The longitudinal Integrated model 323

3.6.7 Synchronous correlations, standardized regression weights and significance of paths in the longitudinal models 325

3.4.8 Individual Factors in the longitudinal models 330

3.6.9 Positive Workplace Factors in the longitudinal models 330

3.6.10 Negative Spillover in the longitudinal models 331

3.6.11 Overall Well-Being in the longitudinal models 331

3.6.12 Mental Illness in the longitudinal models 334

3.6.13 Work Engagement in the longitudinal models 335

3.6.14 Gain and loss spirals 336

3.6.15 Squared multiple correlations from the models 340

3.6.16 Summary of the results of the longitudinal models 340

3.7 Discussion of the longitudinal models 343

3.7.1 Discussion of the Time 1 SEMs 344

3.7.2 Confirmatory factor analyses 346

3.7.3 Factor score weight from the CFAs 352

3.7.4 How factor score weights explain the relationships of the Integrated model 355

3.7.5 The longitudinal models 357

3.7.6 Stability and change in the longitudinal models 359

3.7.6.1 Stability in the longitudinal models 359

3.7.6.2 Change in the longitudinal models 362

3.7.7 Limitations and strengths of Study 2 370

3.7.8 Conclusions 371

Chapter 4: Discussion of research findings and conclusions 375

4.1 The developmental equation, D f PPCT 376

4.1.1 P, the person: The generative disposition 376

4.1.2 P, the person: Their demand characteristics 376

4.1.3 C, the context 377

4.1.4 T, Time 377

4.1.5 Summary of D f PPCT 378

4.2 Major findings 378

4.3 Interesting non-findings 384

4.4 Applications of the research 385

4.5 Future research 388

4.6 A final word 393

References 395

Appendices 442

Appendix A: Call for volunteers from the university alumni 442

Appendix B: Call for volunteers from the public hospital 443

Appendix C Time 2 Call to action 444

Appendix D: Time 3 Call to action 445

Appendix E: Second and third reminder calls to action 446

Appendix F: Measures used in Study 1 and 2 447

Appendix G: Simple slopes of the moderated regression analyses 454

Appendix H: Results of the Time 1structural equation modelling 457

Appendix I: Confirmatory Factor Analyses for the longitudinal models 469

Appendix J: Results of the longitudinal models 483

Appendix K: Terms and glossary for Study 2, Longitudinal modelling 510

List of Tables Table 2.1 Variables in each block as blocks are entered into hierarchical multiple

Table 2.6 Results for the three steps for the hierarchical multiple regression for

Table 2.7 Results for the three steps for the hierarchical multiple regression for work

Table 2.13 Results for the three steps for the hierarchical multiple regression for

Table 2.17 Results at Step 2, showing the significant interactions in the moderated

Table 2.18 Simple slopes for the predictor variable (X1) and the criterion variable (Y)

at Low and High levels of the second moderating variable (X2).………….228 Table 2.19 Z scores for the indirect effects between humour and the outcomes,

through dispositional optimism and coping self-efficacy as the

Table 3.1 Latent and observed variables used in confirmatory factor analyses

……… 272 Table 3.2 Factor Score weights for composite variables for the Well-Being model

Table 3.6 Factor score weights for the composite variables for the Integrated

Table 3.7 Improvement in the fit of models in the Well-Being model by including the

auto-lagged pathways from Time 1 to Time 3……….311 Table 3.8 Results of longitudinal model testing for Well-Being Model………… 315 Table 3.9 Results of longitudinal model testing for Mental Distress Model …… 317 Table 3.10 Results of longitudinal model testing for Well-Being-Mental Health

Table 3.11 Results of longitudinal model testing for the Work Engagement model

……….……… 322 Table 3.12 Results of longitudinal model testing for the Integrated model……….324 Table 3.13 Effect sizes of the standardized regression weights for the auto-lagged

and cross-lagged paths for all the longitudinal models ………….….332-333 Table 3.14 Squared Multiple Correlations for all models for Time 2 and Time 3

List of Figures Figure 3.1 Simplified representation of the components used in SEM ………… 262 Figure 3.2 Simplified representation of the confirmatory factor analyses……… 271 Figure 3.3 Representation of the basic relationships to be tested in the longitudinal

and burnout……… ….391

List of Appendices Appendix A: Call for volunteers from the university alumni ……….….441 Appendix B: Call for volunteers from the public hospital ……….… 442 Appendix C Time 2 Call to action ……… ……….……… 443 Appendix D: Time 3 Call to action ……… 444 Appendix E: Second and third reminder calls to action……….………… 445 Appendix F: Measures for Study 1 and 2 ……… ………… 446 Appendix G: Simple slopes of moderated regression analyses… … ………….453 Appendix H: Results of the early structural equation modelling ……….456 Appendix I: Confirmatory Factor Analyses for the longitudinal models.…………468 Appendix J: Results of the longitudinal models……….…… 482 Appendix K: Terms and glossary for Study 2, Longitudinal modelling …….…….510

Chapter 1: Theories and literature review of Bronfenbrenner‟s developmental

equation, applied to individuals and the work-life interface

The fullness of life is not just our working selves but also our non-work or family selves It is difficult to explain an individual‟s well-being and mental health

by only exploring the workplace factors that influence well-being and mental health, without considering the individual‟s out-of-work responsibilities and activities, whether any or all of these factors can have positive or negative influences, and without considering the person engaged in all these roles Yet much of the

organizational psychology literature has only recently included the positive effects of work and without any particular focus on the „person‟ who is doing the work (Frone, 2003) Working adults are treated as a homogenous group of individuals, upon whom workplace factors, such as working hours, have similar results Where individual differences are introduced, these are often limited to age, gender and negative affect (for example, de Jonge et al., 2001) Whilst men and women are dissimilar in

obvious ways, such as the ability to bear children, gender is not the major difference

as this demarcation suggests (Barnett & Rivers, 2004) Similarly, using age reflects chronological differences, but does not account for the current life stage of an

individual For example, later childbearing in women could mean that comparing 40 year old women may not account for one having a teenage child, another a two year old child, and another who has not had children In the positive psychology literature, whilst the characteristics and strengths of the individual are explored, individuals are not studied in their usual context as working adults, or parents, or adult children caring for aging parents

In order to fully account for all influences on the working adult,

Bronfenbrenner‟s Bioecological Model of human development will form the

framework of this thesis This theoretical perspective will allow person-environment interactions (proximal and distal) to be implicitly explored, across the lifespan and at different lifestages, and with the multiple roles of work and family Whilst

Bronfenbrenner‟s model was originally formulated to account for child development (Bronfenbrenner, 1979; Bronfenbrenner & Morris, 1998) and applied to successfully understand the processes of child development (Steinberg, Darling, Fletcher, Brown,

& Dornbusch, 1995), it has also been applied to adult development The ecological framework adds to the explanatory power for the work-life interface, by including more breadth to the factors to be considered as important to the work-life interface (Barnett, 1998; Grzywacz & Marks, 2000b) There is a similarity of

Bronfenbrenner‟s model to Bandura‟s social learning and behaviour (Bandura,

1986), although Bronfenbrenner has broader outcomes about lifelong development, rather than being focused on learning outcomes

This chapter will start with an outline of Bronfenbrenner‟s bioecological model, and his conceptualisation of the model‟s components The components of the person, their context and time frame will then be examined, examining the relevant theoretical bases and research literature for the component that will be used to

understand how competent development would occur over time For example, when considering the person, self-regulation can explain how individual differences lead to higher well-being and mental health Similarly, when considering the context in which the person is active, role theory can explain how spillover, gender role

attitudes, and role salience influence an individual‟s enactment of their life‟s roles

1.1 Bronfenbrenner’s Bioecological Model of Human Development

Bronfenbrenner‟s bioecological systems theory states that competent

development is the outcome of the bidirectional interactions between an active

individual and a dynamic environment As it began with understanding child

development, it was framed in terms appropriate for the settings in which a child develops (Bronfenbrenner, 1979), the principles and hypotheses that Bronfenbrenner gave can be equally applied to adult development in the work and family settings of adult life, as shown in research by Barnett (1998) and Grzywacz and colleagues (Grzywacz & Bass, 2003; Grzywacz & Marks, 2000b) The formulations of

organizational stress (Kahn, Wolfe, Quinn, Snoek, & Rosenthal, 1964) and social cognitive theories (Bandura, 1986) have similar bases to Bronfenbrenner Both theories include the characteristics of the individual and their environment when considering how the relevant outcomes are achieved, as the response to stress or learning outcomes, respectively The bioecological model accounts for each setting within which the individual acts and the dynamic relationships between the settings (Bronfenbrenner & Morris, 1998, 2006)

For the purposes of this thesis, the effective and competent development in Bronfenbrenner‟s model is viewed as the outcomes of maximising gains and

minimising the loses that occur throughout the life span and the resilience of

maintaining functioning in the face of challenge (P.B Baltes, Lindberger, &

Staudinger, 1998; Bronfenbrenner, 1979; Masten, 2001) The best outcomes of competent development and highest levels of psychological functioning over time lead to mature adults as healthy and fit, with an alert and vital mind, maintaining meaningful roles, either in a continuing vocation or in new activities, maintaining relationships with family and friends and involvement with the community, and finally, to be effective and wise problem-solvers (Csikszentmihalyi & Rathunde, 1998) Given the diversity of these outcomes, effective and competent development will be measured in this thesis in a number of ways that reflect the long term balance

of gains and losses and maintenance of positive functioning, being measured as being, mental health (or the absence of mental illness), burnout, and work

well-engagement Well-being and work engagement fit together as the positive markers of competent development, whilst mental illness and burnout fit together as the negative markers Well-being will be measured as both subjective well-being (Diener, Lucas,

& Oishi, 2002; Diener, Suh, Lucas, & Smith, 1999) and psychological well-being (Ryff, 1989; Ryff & Keyes, 1995) Mental health will be taken as the absence of depression, anxiety, and stress (Beck, 2002; P F Lovibond & S H Lovibond,

1995) Burnout will be measured as exhaustion, cynicism and the loss of professional efficacy (Maslach, Schaufeli, & Leiter, 2001) and work engagement will be

measured with the dimensions of vigour, dedication and absorption (Gonzalez-Roma, Schaufeli, Bakker, & Lloret, 2006; Schaufeli, Salanova, Gonzalez-Roma, & Bakker, 2002) By considering a broad range of outcomes, a better understanding of

competent outcomes can be achieved It should be noted that the genetic component

of development is implied in this model and is understood to be expressed only as a function of the individual‟s environment (Bronfenbrenner & Ceci, 1994) As such the genetic component is beyond the scope of the current research project

The first description of the ecological system focused on the context of

development in great depth, which allowed researchers to pinpoint the factors, both proximal and distal, that are of importance to an individual‟s development This ecological environment was conceived as a set of concentric spheres, nested within each other, similar to a set of Russian dolls (Bronfenbrenner, 1979) The individual sits at the centre of their life domains, or spheres of influence which grow larger and more distant from the individual The closest is the microsystem, the immediate settings in which the individual operates Next is the mesosystem, where two

microsystems interact, then the exosystem of indirect influences, for instance of a partner‟s job or government policy, followed by the macrosystem, as the influence of society or culture, and finally, the chronosystem, which defines the particular point in history (Bronfenbrenner, 1979; Bronfenbrenner & Morris, 1998) Each domain in which the individual operates contain the specific roles, activities and relationships For example, the individual can operate as spouse or partner and/or parent in their household setting (one microsystem), as an employee (a second microsystem),

balancing commitments to home and work (the mesosystem), as a citizen of

Australia (the macrosystem) in the early years of the 21st century (the chronosystem) Whilst other individuals may have the same macrosystem (Australia) and same chronosystem (current time frame), differing personal circumstances, such as divorce

or self-employment will change the experiences within the microsystems and how those microsystems interact These elements provide the experiences that the

individual has in the domain or microsystem and provide a way in which to capture the development influences around the individual By specifying the nature of each element, a richer understanding of the dynamic processes between individual and environment can be gained Similarly, appreciating the changing nature of roles, activities and relationships can reflect how different lifestages influence

developmental outcomes over time, as the individual masters new skills and

situations across the lifespan (Bronfenbrenner, 1979)

Most of the research on work-life issues focuses on the mesosystem of the individual‟s work and home domains and the nature of the boundary and balance between the demands and needs of the two microsystems (Barnett, 1998; Voydanoff, 2002).Whether the time and commitment required for participation in a role leads to strain and conflict with other roles in other settings (Goode, 1960; Greenhaus &

Beutell, 1985) or enhancement and facilitation with the other roles (Frone, 2003; Greenhaus & Powell, 2006; Marks, 1977; Marks & MacDermid, 1996) depends on the dynamics of the individual interacting with the components of each setting

When the expanded formulation of the bioecological model was published in1998, Bronfenbrenner noted that this emphasis on context obscured a necessary and essential component of the process, that of the individual (Bronfenbrenner & Morris, 1998) The extension of the original model, now called the bioecological model has the individual as an active participant engaged in bidirectional

relationships with a dynamic environment The equation (1),

conceptualises these relationships where the developmental outcome, D, is a function

ƒ of the interactions or proximal processes, P, between the person‟s characteristics,

P, and their context or environment, C, that occur with time, T The characteristics of the person are based on their disposition, gender and resources and their demand characteristics

The proximal processes that occur between the person and their environment are considered to be the drivers of development and involve activities which occur regularly and with increasing complexity These activities are reciprocal exchanges and interactions with people and symbols and can be moderated by the individual‟s developmental capacity and the influence by significant others Examples of

activities that increase in complexity over time are parenting, developing a career, learning skills, problem solving and managing multiple roles These processes are those that an active, competent person would use in managing and adapting to the maturation of roles and responsibilities are fundamental to competent development (Bronfenbrenner & Morris, 1998) These bidirectional influences form a systems

approach to development where each level, whether it is genetic, neural, behavioural,

or environmental, is interconnected (Gottlieb, Wahlsten, & Lickliter, 1998) Each part of the equation can operate at the micro-, meso- or macrosystem level in such a way that the potential for development of the individual can be accounted for For example, time can be regarded in the microsystem by whether or not proximal

processes are continuous in the mesosystem by the periodicity of the proximal

processes and in the macrosystem by changes in the broader community

(Bronfenbrenner & Morris, 1998)

Specifically, proximal processes between the active individual and their dynamic environment involve activities which occur regularly and with increasing complexity over time The processes that are involved in managing and adapting to the maturation of roles and responsibilities are fundamental to competent

development and can be moderated by the individual‟s developmental state and the influence by significant others (Bronfenbrenner, 1979; Bronfenbrenner & Morris,

1998, 2006) Some examples of the activities that increase in complexity over time are parenting, developing a career, learning recreational or sporting skills, problem solving, and managing multiple roles Similarly, implicit in career development is that the individual‟s job description becomes more complicated and involved over time, such that early career jobs involve less responsibility and input than senior positions that oversee many employees and require detailed knowledge of the many facets of the relevant business situation or management goals The active individual therefore uses available skills and their personal and environmental resources to successfully make the transitions from lower levels to higher levels of complexity and ability in their lives Proximal processes are therefore reliant on the individual‟s characteristics and their particular situation in life and can be many and varied,

making the processes difficult to define As such, the focus in the thesis will be directly on the individual and their context Rather than attempting to specify directly which processes are involved or how exactly an interaction may occur between individual characteristics and contextual factors, it is taken that these processes can

be implied from the influential personal and contextual factors As such, proximal process will be implied from the results of the research, rather than explicitly stated

Understanding the individual, their family and their work and spillover

between roles will identify the factors that are most important to well-being at the work-life interface In the current research, explicitly using the ecological equation will highlight the individual as an active participant of the system and in the

interactions between important life spheres Positive person-environment interactions contribute to competent and resourceful outcomes In the face of adversity,

competence and the adaptive use of personal and environmental resources foster the development of resilient children (Kumpfer, 1999; Yates & Masten, 2004)

Resilience can also be considered the actions of a competent person when facing risky or adverse situations (Masten & Reed, 2002) Likewise, in adults the ability to use personal and environmental resources facilitates well-being and role balance (Barnett, 1998; Voydanoff, 2005b) and in families, it allows adjustment and

adaptation to strains caused by demands on the family unit (J M Patterson, 2002) Negative person-environment interactions hinder well-being by increasing

dysfunctional behaviours, for example, when problem drinking is exacerbated by increasing pressure from home and work (Grzywacz, 2000; Grzywacz & Marks, 2000a)

The purpose of the current research is to combine knowledge from diverse strands of psychology, such as from the health and organizational domains and from

positive psychology, in a form that weighs the person and context components of the work-life puzzle By basing the research on Bronfenbrenner‟s bioecological model, neither part can be overlooked The research to be conducted in the current thesis will use a variety of analyses to understand the competent individual The first study will involve regression analyses and the second study will involve longitudinal modelling to explore and model the predictors of competent behaviour This research program will explore the working individual from different viewpoints: how

individuals understand themselves and how outcomes can be understood in a large sample at one time and across time

1.2 Theories for D, the developmental outcomes, defined by well-being,

mental health, burnout and work engagement

A range of outcomes will be used to describe competent development and to reflect the diversity of outcomes used in the literature The markers of competent and successful individual development and management of multiple roles will be defined

as first, well-being, as life satisfaction and psychological well-being, second, mental health (as the absence of depression, anxiety and stress) and third, as the affective work state of burnout and it‟s recently quantified opposite engagement

Bronfenbrenner describes competent outcomes as the result of the individual‟s

actions in each of the domains, or microsystems, in which they have roles, activities and relationships (Bronfenbrenner, 1979)

1.2.1 Well-being

Well-being will be measured by subjective well-being, as life satisfaction (Diener, Emmons, Larsen, & Griffin, 1985) and psychological well-being, as the components described by Ryff (1989) Whilst subjective well-being has been factor analysed into the three components of life satisfaction, positive affect and negative

affect (Arthaud-Day, Rode, Mooney, & Near, 2005), only life satisfaction will be considered in this thesis as there is issue of whether affect should be treated as a state

or trait of the individual, which blurs the construct that is being measured and studied (Wainwright & Calnan, 2002)

Psychological well-being is amongst the loosest and poorly defined terms used for outcome measures in the research literature on personality, health and work-life issues It is used as an umbrella term rather than a specific construct and can be taken as any positive outcome or the absence of negative outcomes For example, in

a review of 25 studies on the benefits of optimism, Scheier, Carver and Bridges (2002) reported „psychological well-being‟ had been measured as lower depression, less anger, less loneliness, anxiety and distress, fewer perceived hassles, less negative mood, and lower stress levels, in addition to higher life satisfaction, higher job

satisfaction, and higher self-esteem Given that psychological well-being is central to the conception of competent development in this thesis, it shall be defined only as described by Ryff, (1989) as the six components that measure challenged thriving and are based on ethical and philosophical traditions These components are

autonomy, environmental mastery, personal growth, positive relations with others, purpose in life, and self-acceptance (Ryff, 1989; Ryff & Keyes, 1995) and represent the conception of the good life, full of meaning and worthwhile activities and

relationships

By including both life satisfaction and psychological being in being, this thesis brings together recent research that shows that these constructs are separate and together add valuable information about the individual‟s mental state (Ryan & Deci, 2001) Life satisfaction can be viewed as hedonia, as happiness and pleasure in life, the „happy life‟, whilst psychological well-being, eudaimonia,

well-defines the purpose and markers of challenged thriving, the „good life‟ (Keyes, Shmotkin, & Ryff, 2002).The combination brings together measurement of a happy and meaningful life

The theoretical explanations of well-being considers the influences of

objective conditions or the circumstances around the person (bottom-up), such as age, gender, income, life domains and culture, and of subjective conditions, such as the person‟s disposition/personality (top-down) to be the „cause‟ of the person‟s well-being, along with adaptation and goals (Diener & Lucas, 2000; Diener et al., 1999) Using an ecological framework, however, allows all of these theoretical inputs to be acknowledged and accounted for, however limited the input may be Whilst the situation factors have limited influence on well-being, i.e demographics account for only 15% of variance (Argyle, 1999, in Diener & Lucas 2002), Easterlin (2006) calculated that life satisfaction across the life span was the sum of satisfaction with various life domains, such as work, family and financial, rather than due to personality factors However, given the high correlation between overall life

satisfaction and domain satisfactions, these calculations may reflect how these

relationships ebb and flow over time, rather than how domain satisfaction „causes‟ life satisfaction over time

The subjective or top-down influences on well-being are more varied, with the theories about personality explaining how well-being is achieved Personality influences are explained by temperament (hereditability of happiness) (Lykken & Tellegen, 1996), traits (DeNeve & Cooper, 1998) and dispositions, such as optimism (Armor & Taylor, 1998) and self-efficacy (Schwarzer & Renner, 2000) These

personality theories will be explored as part of discussions later in this chapter on understanding the person who is occupying and managing multiple roles‟

Adaptation to change and goals are also considered as precursors to being Adaptation to changing circumstances, such as widowhood or winning the lottery, are evidence that there is a set point for well-being, that despite good or bad fortune, an individual will return to previous levels of happiness after a period of time (Fujita & Diener, 2005; Shmotkin, 2005) This hedonic treadmill can explain the stability of well-being over time, although the expectation that an individual inevitably returns to their original functioning, revisions take into account that whilst individuals are mostly happy, they also have multiple set points for different domains and there are individual differences in the way people react to similar situations (Diener, Lucas, & Scollon, 2006) Goals can be seen as the way by which individuals conceive their future, defining the direction and focus of all that individual‟s actions (Emmons, 2003) Expectations about the outcomes of goals are also important in how likely the individual is to persist toward their goals and whether the individual will disengage from insoluble problems (Aspinwall, 2001; Carver & Scheier, 1998) The adaptive value of persistence and disengagement will be explored further

well-through the effects of dispositional optimism on competent development

1.2.1.1 Prevalence Given the historical focus of psychology on pathology (Seligman & Csikszentmihalyi, 2000) it may be surprising to consider that across many studies in large populations, most people are happy most of the time Although many different scales can be used to measure well-being, the results are remarkably consistent By bringing together the results of almost 916 studies, with over 1 million participants in total in 45 countries, Myers and Diener (1996) calculated that the mean „happiness‟ rating, on a scale of 0 to 10, was 6.75, with most surveys reporting ratings between 5.5 and 7.75, and very few surveys where people rated their

happiness at less than 5

In Australia, the Australian Quality of Life Index has been calculated in a large representative sample several times each year from 2001 onwards The 18th edition shows that within that time frame, the Personal Well-being Index (PWI) scale has been remarkably stable in that time The mean has ranged only between 73.4 and 76.4, on a scale of 0 to 100 that rates satisfaction with life as a whole The authors believe that this level of stability represents normal well-being (or homeostasis) where people manage their lives successfully and are optimistic about the future (Cummins, Woerner et al., 2007) The greatest PWI (i.e PWI > 79) was associated with high levels of income and the presence of a partner, whilst those with the lowest PWI (i.e PWI < 70), and most at risk of homeostatic failure are unemployed, have a low income (under $15,000 pa) and live without a partner

With regards to work status, not earning an income was particularly adverse for the well-being of men between 26 and 55 years of age (Cummins et al., 2003; Cummins, Woerner et al., 2007) Interestingly, whilst many people note that they would like to reduce their working hours, this did not translate into greatly reduced rating of their well-being Rather it was insufficient work, not too much work, which had the stronger, more detrimental effect on individual‟s PWI Underemployment is linked to lower incomes, a lower sense of achievement in life and boredom and among men between 35 and 55 years of age (Cummins, Woerner et al., 2007) In addition, looking for a job, whether employed or unemployed, reduced satisfaction with life‟s achievements whilst being engaged in work gave a sense of purpose to life and was central to overall well-being (Cummins, Woerner et al., 2007) The specific predictors of well-being, measured as life satisfaction and psychological well-being, will be considered in the later sections of this chapter when the effects of the person and the context are considered

1.2.1.2 Stability of well-being Given the focus of research on over-work as a stressor, this finding that lack of work reduced quality of life should make easy acceptance of the „work-life imbalance‟ mantra less likely Has feeling busy been transformed into being pressured? By interpreting emotional distress as a possible loss of efficacy, the individual could be underestimating how well they do actually manage their role demands (Llorens, Schaufeli, Bakker, & Salanova, 2007),

unrealistically comparing themselves to an ideal worker/parent ideal, or not

questioning the media representations of „having it all‟ Despite the widespread media preoccupation with negative outcomes, the stability of well-being is important

as it denotes a substantial level of happiness that buffers individuals without their being particularly aware of that happiness Longitudinal studies show that mature coping mechanisms, good relationships, particularly marriage, having sufficient income to meet one‟s needs and having meaningful work all contribute to well-being across the lifespan (Howard, 1992; Vaillant, 2000, 2002) Well-being is reduced by excessive alcohol use, maladaptive coping mechanisms (Vaillant, 2000, 2002) and by materialism, which is associated with reduced relationship and family satisfaction (Nickerson, Schwartz, Diener, & Kahneman, 2003; Solberg, Diener, & Robinson, 2004) Similarly, a focus on extrinsic goals was associated with more narcissism and depression than the pursuit of intrinsic goals (Kasser & Ryan, 1996)

1.2.1.3 Australian health and working provisions There are two important caveats, particularly in comparison to the USA that must be taken into account when studying the well-being and mental health of Australian workers Socioeconomic status (which includes job status) has been associated with poorer health and well-being outcomes in the USA because of the link between the provision of health care and employment (for example, Adler et al., 1994) However in Australia, every

person, regardless of employment status or income has access to government funded heath care with access to general practioners and public hospital services at little or

no cost Similarly, Australian Federal Government regulations (available at

www.workplace.gov.au) include provision of sick leave and minimum four weeks‟ holidays in employment contracts to provide a safety net for employees These

provisions mean that Australian employees have medical advantages that are not tied

to any employment and employment conditions that may not be available to US employees These Australian conditions could therefore limit the importance of SES

as a predictor of health and well-being outcomes

1.2.2 Mental health, as the absence of mental illnesses

This thesis will focus on the cognitive models of mental health rather than any biological basis Whilst a link between genetic factors and environment has been shown in a longitudinal study of Australian teachers, depression was more strongly linked to multiple adverse life events, rather than susceptible genetic subtype

(Wilhelm et al., 2006) As noted previously, the link between genetics and

environment is beyond the scope of this thesis, although this may provide an

interesting avenue for future research Further, the focus of the current thesis is on the individual difference, work and family variables (to be defined and described later in this chapter) that are risk and protective factors for mental illness, rather than

a wider range of variables that have been examined in previous research

The cognitive model of depression (Beck, 2002) has been remarkably

successful in explaining the underlying processes involved in depression and many other mental disorders, such as anxiety disorders, panic disorders, and personality disorders and their successful treatment (Beck, 1991) The maladaptive schemas that the individual use to process the positive and negative events around lead to

dysfunctional cognitive styles, which in turn lead to a specific vulnerability to

developing depression (Alloy et al., 2000) Alongside the schemas that give rise to cognitive vulnerability, behavioural and verbal interactions with other people can reinforce and intensify depression which can lessen the social support available to individuals These individuals not only believe that they have poorer social skills, but also exhibit behaviours, such as a monotonous tone of voice and avoiding eye

contact, that provoke negative responses in other people (Segrin & Abramson, 1994) Excessively seeking reassurance from family and friends can hamper interpersonal relationships, as does the tendency to seek negative feedback about oneself

Vacillating between seeking reassurance and seeking negative feedback leads to rejection from peers (Joiner & Metalsky, 1995) and excessive reassurance seeking has been specifically linked to depressive symptoms (Burns, Brown, Plant, Sachs-Ericcson, & Joiner, 2006)

The role of cognition in mental health is formalised by Beck‟s cognitive models of depression Seen as the way in which individuals process information about themselves, their world and their future, negativity is shown by selective abstraction, overgeneralization, dichotomous categorisation and personalization of the problems that occur to the individual and resulting in distorted and depressive cognitions (Beck, 2002) This altered thinking changes the way that individual‟s view the world, resulting in sensitivity to negative or ambiguous cues (Wilkinson & Blackburn, 1981), differences in depressed and non-depressed thinking (Alloy et al., 1999) and differences in interpersonal relationships (Joiner & Metalsky, 1995) There can also be changes in cognitions that are rational or irrational and changing such cognitive patterns requires challenging the irrational thoughts to find more logical and reasonable rational replacements (Ellis, 2004) Beck‟s cognitive models