THE ROLE OF PRIMARY/SECONDARY CONTROL IN POSITIVE PSYCHOLOGICAL ADJUSTMENT Dr. Luke J. Heeps B.A./B.Sc. (Australian National University) B.Sc. Honours (University of Wollongong) D.Psyc (Clin. Psyc.) (Deakin University) Submitted in fulfillment of the requirements for the degree of Doctor of Psychology (Clinical) Deakin University October 2000 THE ROLE OF PRIMARY/SECONDARY CONTROL IN POSITIVE PSYCHOLOGICAL ADJUSTMENT L. J. Heeps Abstract This thesis concerns the relationship between control and positive psychological adjustment. The two-process model of Primary and Secondary Control is a more recent conceptualisation of control that proposes two main strategies by which people may develop a sense of control. Primary control involves the person manipulating
263
Embed
The Role of Primary-secondary Control in Positive Psychological
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
THE ROLE OF PRIMARY/SECONDARY CONTROL IN POSITIVE
PSYCHOLOGICAL ADJUSTMENT
Dr. Luke J. Heeps
B.A./B.Sc. (Australian National University)
B.Sc. Honours (University of Wollongong)
D.Psyc (Clin. Psyc.) (Deakin University)
Submitted in fulfillment of the requirements for the degree of
Doctor of Psychology (Clinical)
Deakin University
October 2000
THE ROLE OF PRIMARY/SECONDARY CONTROL IN POSITIVE PSYCHOLOGICAL
ADJUSTMENT
L. J. Heeps
Abstract
This thesis concerns the relationship between control and positive psychological adjustment. The two-
process model of Primary and Secondary Control is a more recent conceptualisation of control that
proposes two main strategies by which people may develop a sense of control. Primary control involves
the person manipulating environmental circumstances in order to suit their needs and wants, whereas
secondary control strategies involve the person manipulating their internal cognitive/affective states in
order to reduce the psychological impact of events.
Primacy theory proposes that primary control is functionally more adaptive than secondary control and
that secondary control mostly functions to compensate for low primary control. However, some
empirical evidence suggests that exercising too much control over the environment may be associated
with negative physical and psychological outcomes. Furthermore, cross-cultural evidence suggests that
Chapter 1 179
some cultures (particularly Asian cultures) reinforce the benefits associated with secondary control over
primary control. In addition, the original authors of the primary/secondary control constructs
theoretically suggested the importance of a balance in the ratio of primary to secondary control levels for
optimal adaptive functioning, rather than the adaptive importance of one control process over the other.
Previous empirical tests of primacy theory have generally produced mixed or inconclusive results. In
addition, previous studies testing primacy theory operationalised secondary control in vague, non-
descriptive terms that may have equated secondary control with helpless acceptance. Furthermore,
studies have only narrowly defined adaptive functioning in terms of negative psychological outcomes
(e.g. depression and anxiety), while neglecting the relationship between control and positive expressions
of psychological health (e.g. Subjective Quality of Life).
Hence, a new measure of primary and secondary control was developed for this study, the Primary and
Secondary Control Scale (PSCS). The PSCS operationalises both primary and secondary control in
terms of specific cognitive and behavioural strategies aimed at either control of environmental
circumstances (primary control) or control of internal states in order to minimise psychological impacts
(secondary control). Factor analyses and reliability analyses of the PSCS items reported in Studies 1 and
2, demonstrate the PSCS as a valid and reliable measure of the primary/secondary control construct.
Studies 1 and 2 also tested hypotheses regarding the primacy of primary control. In particular, it was
hypothesised that rather than primary control being more adaptive than secondary control, it is a balance
in the levels of primary and secondary control that is important for optimal psychological adjustment.
Furthermore, it was predicted that secondary control’s role in adaptive functioning is greater than simply
acting to compensate for low primary control. In addition, the relations between primary/secondary
control and more comprehensive and positive measures of psychological health were investigated, as
opposed to the relationships with negative measures of psychological health as reported in previous
studies.
Results observed in Study 1, then confirmed in Study 2, did not support the primacy of primary control.
That is, people that were “control-imbalanced” (high levels on one control process, but low on the other)
reported lower levels of positive psychological adjustment compared to people who reported a balance of
average or above-average levels of both primary and secondary control (“control-balanced”).
Furthermore, secondary control was found to predict variance in positive adjustment for both high and
low groups of primary control, suggesting that secondary control’s role in adaptive functioning is more
than simply acting as a back-up strategy for low primary control.
2
2
Chapter 1 179
In addition Study 2 aimed to explore the capacity for the personality dimensions of Extraversion and
Neuroticism to explain the relationship between primary/secondary control and positive psychological
adjustment. Results showed that shared variance between primary/secondary control and positive
adjustment was only partially explained by the personality dimensions of Extraversion and Neuroticism.
Further analyses conducted showed that personality factors explained the primary control-adjustment
link to a greater extent than the secondary control-adjustment link.
The findings are then discussed with reference to the implications they have for understanding adaptive
psychological functioning. Hence, it is argued that adaptive psychological functioning involves
maintaining higher levels of both primary and secondary control to allow the person to be flexible in
their use of approach-avoidant strategies in order to deal with situations with varying levels of objective
and subjective controllability associated with them. In contrast, emphasising one control process over
the other, it is argued, is associated with vulnerability to the inherent mix of controllable and
uncontrollable circumstances that we face in life. Furthermore, secondary control’s role in adaptive
functioning, it is argued, is more than simply a back-up strategy for low primary control; it is critical to
the effective functioning of primary control behaviour.
Overall, the present analysis does not support the primacy of primary control in adaptive psychological
functioning, but rather the strong functional inter-dependence of primary and secondary control.
3
3
Chapter 1 179
Acknowledgments
I would like to give greatest thanks to my mother and father. Firstly, to my father, Robin Heeps, whose
financial support made the last few years possible. Without his help and belief in what I was trying to
achieve for my life I would not have developed the quality of research, clinical and personal skills that
will stand me in good stead in later working life.
To my mother, Diane Heeps, whose very spirit of wisdom, tolerance and compassion for others inspired
and continue to inspire the theory and practice that my thesis and clinical work aspire to.
I would like to dedicate this work to my older brother, Benjamin James Alexander Cottam Heeps, who
was killed suddenly and under tragic circumstances last year. His quality of character and respect for
those in all-walks of life was truly remarkable. His influence on my development as a person both as a
young child and more recently in Melbourne is even more remarkable and defies words.
To my brother Simon Heeps, I acknowledge a greater strength and resolve in both of us to have a rich
and fulfilling life, and although we may be distant much of the time, our bond of shared struggle in
family and wider life I hope will continue into the future.
I also acknowledge two other very important men in my life that also died during the completion of this
work. Firstly, Paul Stuckey, whose spirit for adventure and love of natural environments lead us into so
many memorable days on the rocks, in the mountains or exploring whatever there might be to explore. I
miss you, yet believe me, I am continuing the journey.
To Richard Walker-Powell who courageously died last year in a selfless effort, during a United Nations
food drop over Kosovo. Rich reminds me about the relationship between belief in oneself, effort and
defying the odds. Then to fail and get back into life as quickly as possible. I acknowledge this role Rich
has had in my life and will continue to. Thank you for reminding me what to do in times of self-doubt.
A special thanks to my supervisor, Professor Bob Cummins, for his professional and respectful attitude
toward students. This allowed me the autonomy and respect to pursue a line of research that this work
embodies. His willingness to provide direct hands-on effort at finding relevant articles, helping with
ethics clearance and achieving a sample allowed me to focus more clearly on the development of new
theory and a new measure. In addition, it was due to Bob’s encouragement for much reading and
engaging in lengthy discussions of often confusing theory that assisted greatly in the development of the
ideas that feature in this work.
4
4
Chapter 1 179
To my peers and colleagues at Deakin for their emotional and professional support, Roger Brink,
Margaret De Judicibus, Monique Garwood, Eoin Killackey, Litza Kiropoulos, Sophie Kurts, Karen
Marriage and Fausta Petitio. Also to the administrative staff including, Julie, Betina Gardener, Anne
Moirea, and especially Trudy Wallace for her professional and respectful manner in dealing with
students needs and the good management of her department functions (e.g. the test libarary).
To my friends, both in climbing circles and ‘over the river’ in Fitzroy whose support was akin to the
family away from home. Particular thanks to Angus Boyle, Bridget Leo and Geoff Hooker. A heartfelt
thank you to Anastasia Konstantelos for her loving friendship, support and cherished visits. Special
thanks to Todd Zemeck for his genuine compassion and willingness for lengthy discussions to do with
the management of my own health over the last twelve months, thanks Todd. Special thanks to the staff,
patrons and friends of the Napier Hotel, especially the publicans Guy and Arty, Peter Hickey and Jody,
Glen (G.P.) and Vicky, Pia Emery, Yvette, Katie, Roger, Arnie, old Eric, Mark Collins and Mandy for all
extending a normalising, compassionate hand during an unusual time.
5
5
Chapter 1 179
Lastly, to the mountain playgrounds of the Grampians (Gariwerd), Mount Arapiles (Djurite), Victorian
Alps and the New Zealand Southern Alps for the wonder, inspiration, enjoyment and understanding that
these challenging landscapes instill in me. It is these lands that give my particular journey meaning.
Luke Heeps
October, 2000.
6
6
Chapter 1 179
Table of Contents
Abstract……………………………………………………………..iv
Acknowledgments…………………………………………….……vii
Table of Contents…………………………………………………...ix
List of Tables…………………………………………………….xiv
List of Appendices……………………………………………….xvi
CHAPTER 1 – INTRODUCTION TO STUDIES OF PRIMARY AND SECONDARY CONTROL
1.1 General Introduction: An Overview…………………………...1
1.2 Control Concepts and Definitions: A Brief Review…………...2
1.3 Primary and Secondary Control: The Two-Process
Model of Control………………………………………………3
1.3.1 Specific Secondary Control Strategies…………………..5
1.3.2 Alternative Conceptions of Primary and
Secondary Control………………………………………6
1.3.3 The Expanding Classification of
Control Strategies……………………………………….7
1.3.4 Primary Control Strategies……………………………...8
1.4 The Relationship Between Primary/Secondary Control
and Psychological Adjustment: The Adaptive Value of
Primary Control……………………………………………10
1.4.1 The Adaptive Value of Secondary Control……………..12
1.5 The “Primacy” of Primary Control……………………………14
1.5.1 Empirical Tests of Secondary Control as a
Back-Up Strategy For Low Primary Control……………15
1.5.2 Cultural Differences in Preferences for Primary
Versus Secondary Control: Is Secondary Control
Simply Used as a Back-Up Strategy?……………………17
1.5.3 Empirical Tests of the Greater Adaptive Value
of Primary Control…………………………………….19
1.5.4 The Adaptive Limits to Primary Control……………...21
1.6 The Adaptive Limits to Secondary Control and
7
7
Chapter 1 179
Primary/Secondary Control Imbalances…………………25
1.7 The Adaptive Value of Maintaining a Balance Between
the Levels of Primary and Secondary Control…………...28
1.8 Summary…………………………………………………30
CHAPTER 2 – OPERATIONALLY DEFINING PSYCHOLOGICAL ADJUSTMENT IN CONTROL
RESEARCH
2.1 The Negative Affect Bias in Control-Adjustment
Research…………………………………………………….32
2.2 Towards a More Positive and More Comprehensive
Measure of Psychological Adjustment……………………...34
2.3 Conclusions and Hypotheses for Study 1…………………...36
CHAPTER 3 – STUDY 1: METHOD
3.1 Participants………………………………………………….38
3.2 Materials…………………………………………………….38
3.3 Procedure……………………………………………………41
CHAPTER 4 – RESULTS: STUDY 1
4.1 General Introduction…………………………………………43
4.2 The Factor Structure of the Primary and Secondary Control
Scale (PSCS): Initial Analyses………………………………44
4.2.1 Final Solution and Reliability Analyses…………………49
4.2.2 Possible Interpretations of the Negative Loadings
Observed for Factor 2………………………………………50
4.3 Analyses of the Hypotheses to Study 1…………………….54
CHAPTER 5 – DISCUSSION OF STUDY 1
5.1 The “Primacy” of Primary Control…………………………60
8
8
Chapter 1 179
5.1.1 Differences With Past Theory and Research: The
Importance of Operational Definitions of
Primary and Secondary Control in
Control-Adjustment Research…………………………..64
5.2 Limitations to Study 1……………………………………...67
CHAPTER 6 – INTRODUCTION TO STUDY 2
6.1 General Introduction……………………………………….69
6.2 A Summary of Previous Theory and Evidence from
Study 1: The Importance of Maintaining a Balance
in the Levels of Primary and Secondary Control…………..69
6.2.1 Secondary Control’s Adaptive Function is More
Than a Compensatory Mechanism For Low primary
Control………………………………………………..70
6.2.2 The Relationship Between Primary/Secondary
Control and Positive Measures of Psychological
Adjustment…………………………………………..70
6.3 Personality, Control and Psychological Adjustment…...71
6.3.1 The Relationship Between Personality and Positive
Psychological Adjustment………………………….72
6.3.2 Conceptual Similarities Between Coping and
Primary/Secondary Control………………………...75
6.3.3 Differences Between Coping Measures and the
Measurement of Primary/Secondary Control………78
6.3.4 The relationship Between Personality
(Extraversion and Neuroticism) and Coping……….80
6.4 Hypotheses for Study 2…………………………………… 81
CHAPTER 7 – STUDY 2: METHOD
7.1 Participants…………………………………………………84
7.2 Materials……………………………………………………84
9
9
Chapter 1 179
7.3 Procedure…………………………………………………...87
CHAPTER 8 – RESULTS: STUDY 2
8.1 Data Screening…………………………………………..….89
8.2 Factor Analyses and Reliability Analyses of the Primary
and Secondary Control Scale (2nd Version)………………...89
8.3 Replication of Initial Hypotheses: Hypothesis 1 – The
Relationship Between Primary/Secondary Control and
Positive Adjustment Variables……………………………..92
8.3.1 Hypothesis 2 – The Role of Maintaining a Balance
Between Control Processes for Positive Psychological
Adjustment…………………………………………….93
8.3.2 Hypothesis 3 – Secondary Control as Simply a
Back-Up Strategy for Low Primary Control………....95
8.4 Exploratory Analysis: The Role of Personality in
Explaining the Relationship Between Control
and Positive Psychological Adjustment…………………....97
CHAPTER 9 – DISCUSSION OF STUDY 2
9.1 The Measurement of Primary/Secondary Control:
The Primary and Secondary Control Scale (PSCS)…………..103
9.2 Primary/Secondary Control and Positive
Psychological Adjustment…………………………………….105
9.3 The Primacy of a Balance in the Levels of Primary and
Secondary Control for Positive Psychological Adjustment…..107
9.3.1 Possible Explanations for the Primacy of the
Primary/Secondary Control Balance in
Positive Psychological Adjustment………………………110
10
10
Chapter 1 179
9.4 Secondary Control’s Functional Role in Adjustment
is Greater Than Simply Compensating
For Low Primary Control……………………………………..115
9.4.1 The Function of Secondary Control in
Positive Psychological Adjustment………………………117
9.5 The Role of Personality in Explaining the Relationship
Between Primary/Secondary Control and
Positive Psychological Adjustment…………………………..120
9.6 Further Limitations and New Directions for Research……….123
CHAPTER 10 – CONCLUDING OVERVIEW
10.1 Summary and Conclusion…………………………………..125
10.2 Clinical Implications………………………………………..130
REFERENCES……………………………………………………134
List of Tables
Table 4.1 Abbreviated Items and Factor Loadings
for the PSCS (initial extraction)……………………………………….46
Table 4.2 Primary and Secondary Control Factors
and Item Loadings for the PSCS………………………………………50
Table 4.3 Means and Standard Deviations for
Factor 1 (Primary Control) and Factor 2 (Secondary Control) for
The Total Sample (N = 192), Participants < 45 Years of Age
and Participants > 45 years…………………………………………….54
Table 4.4 Summary Table for Multivariate Analysis of Variance
Examining Group Differences in Subjective Quality Of Life,
11
11
Chapter 1 179
Positive Affect and Positive Thinking for the Four Combinations
of Primary & Secondary Control……………………………………...56
Table 4.5 Summary Table for Hierarchical Regression Analyses
Investigating The Relationship Between Primary/Secondary Control
And Subjective Quality of Life for Participants with High Primary
Control and Participants with Low Primary Control…………………..58
Table 4.6 Standard Regression Analyses Investigating the
Relationship Between Primary/Secondary Control and
Measures of Positive Psychological Adjustment………………………59
Table 8.1 Primary and Secondary Control Factors and
Item Loadings for the PSCS (2nd Version)……………………………..91
Table 8.2 Study 2: Standard Regression Analyses and Squared
Semi-Partial Correlations (% of Unique Variance Accounted for)
Investigating The Relationship Between Primary/Secondary
Control and Measures of Positive Psychological Adjustment
(Subjective Quality of Life, Positive Affect
and Positive Thinking)…………………………………………………92
Table 8.3 Study 2: Summary Table for Multivariate Analysis
of Variance Examining Group Differences in Subjective Quality
of Life, Positive Affect and Positive Thinking for the Four
Combinations of Primary & Secondary Control………………………94
Table 8.4 Study 2: Summary Table for Hierarchical Regression
Analyses Investigating the Relationship Between
Primary/Secondary Control and Subjective Quality of Life
for Participants with High ( N = 94) and Low (N = 75) Primary Control………………….
……………………………………………..96
Table 8.5 Study 2: Summary Table for Hierarchical Regression
Analyses Predicting Subjective Quality of Life (SQOL) and
12
12
Chapter 1 179
Positive Thinking with Personality Variables (Neuroticism &
Extraversion) and Control Variables (Primary &
Secondary Control)……………………………………………………98
Table 8.6 Study 2: Summary Table for Hierarchical Regression
Analyses Predicting Positive Affect and Negative Affect with
Personality Variables (Neuroticism & Extraversion) and
Control Variables (Primary & Secondary Control)
………………………………………………………………..99
Appendices
Appendix A1: Information Sheet and Covering Letter Sent to
Participants for Study 1
Appendix A2: Information Sheet and Covering Letter Sent to
Participants for Study 2
Appendix B1: Questionnaires Used in Study 1 and Brief Notes
Describing the Origin of Items for the Primary & Secondary
Control Scale (PSCS 1st version)
Appendix B2: The 2nd Version of the Primary & Secondary
Control Scale (PSCS) and the Extraversion and Neuroticism scales from the NEO FFI Personality
Inventory
Appendix C1: Ethics Approval Documents for Study 1
Appendix C2: Ethics Approval Documents for Study 2
13
13
Chapter 1 179
Appendix D1: Examiners’ Reviews
14
14
Chapter 1 179
CHAPTER 1
INTRODUCTION TO STUDIES OF PRIMARY AND SECONDARY CONTROL
1.1 General Introduction: An Overview
Over the past thirty years extensive psychological theory and research has produced a wide variety
of constructs explaining how people may develop a sense of control over the environment and
outcomes in life. One recent conceptualisation of control currently under investigation has
expanded the control concept to include two-processes of control. Primary control refers to
perceptions, beliefs or actions aimed at manipulating the objective conditions of situations.
Secondary control refers to perceptions, beliefs or actions aimed at controlling one’s internal
reactions to events, in order to reduce their psychological impact. Recent theorising proposed that
primary control has greater adaptive value than secondary control, and secondary control is
conceived as a back-up strategy when primary control is low. The following analysis of the
literature suggests that these theories are over-simplified, and that maintaining a balance between
the levels of primary and secondary control is optimally adaptive. Furthermore, it is argued that
secondary control may have intrinsic adaptive value for the individual; rather than functioning only
as a back-up strategy. In addition, it is argued that psychological adjustment has been narrowly
defined in the control literature and that more comprehensive and positive measures of
psychological functioning should be used to explore the relationship
15
15
Chapter 1
between control processes and psychological adjustment. A more
comprehensive and positive method of operationalising adjustment can be
achieved by measuring subjective quality of life as well as a mix of
cognitive and affective processes such as positive affect and positive
thinking.
1.2 Control Concepts and Definitions: A Brief Review
In the late 1950’s and 1960’s researchers began to take an interest in the
psychology of personal control. During the 1950’s, the early behaviourist
movement removed psychological concepts such as self-control, will and
voluntary control of consciousness from psychology’s research agenda, to assist
in the development of a new scientific study of human behaviour (Shapiro,
Schwartz & Astin, 1996).
However, over the following three decades, a substantial body of literature on
control was developed offering a plethora of definitions and conceptions of
personal control and control related constructs. It must be noted that the
following descriptions are a selection of concepts and in no way represent an
inclusive overview of the enormity of psychological constructs that have been
related to control (for an overview see Rodin, Schooler & Schaie, 1990).
As examples which demonstrate the breadth of the available constructs, Averill
(1973) conceived personal control as comprising three main strategies, or types
of control: behavioural control (direct action over the environment), cognitive
(the interpretation of events) and decisional (exercising choice amongst goals).
Rotter (1966) proposed the locus of control construct which viewed control as a
16
Chapter 1179
generalised belief concerned with whether the person perceived outcomes in life
as contingent on their own actions (internal locus of control) or on forces outside
of the self (external locus of control). Baron and Rodin (1978) emphasised the
difference between perceived control and the persons’ actual, objective ability to
control outcomes. Rodin (1990) discussed the underlying theme of control
within Bandura's (1977) theory of self-efficacy and outcome expectancies.
Peterson and Stunkard (1989) defined control as a person’s belief in their ability
to bring about good outcomes and avoid negative outcomes, and Langer, (1975)
proposed that control was sometimes a cognitive illusion.
Whilst numerous control-related constructs have been explored, the dominant
conceptualisation overall has equated personal control with the person’s actual
or perceived ability to influence existing social, physical and behavioural
Rothbaum et al. (1982) believe that it is the adaptive consequences of secondary
control that differentiate it from passive helplessness. Thus, secondary control
should be operationalised in ways that demonstrate consistent positive
relationships with measures of psychological adjustment. Hence, a less
problematic way to conceptualise and operationalise secondary control is to
emphasise the basic process that characterises secondary control, a process that
Rothbaum et al. (1982) argue is commonly associated with positive outcomes.
That is, acceptance of the objective conditions in one's life.
20
20
Chapter 1179
In addition, Thompson, Cheek & Graham (1988) and Weisz (1990) suggested
that primary and secondary control are conceptually related to the concept of
problem-focused and emotion-focused coping, respectively. Coyne, Aldwin &
Lazarus (1981) defined problem-focused coping as referring to “...efforts to deal
with sources of stress, whether by changing one’s own problem maintaining
behaviour or by changing environmental conditions”. In contrast, emotion-
focused coping was defined as “...coping efforts aimed at reducing emotional
distress” (p.440). In comparison with the previous definition of primary and
secondary control offered by Weisz et al. (1984), the problem-focused and
emotion-focused constructs are almost indistinguishable conceptually.
Furthermore, specific secondary control strategies outlined by
primary/secondary control theorists (Rothbaum et al., 1982; Heckhausen &
Schulz, 1995) have the ultimate function of controlling the psychological impact
(particularly the emotional impact) of outcomes in life; as do emotion-focused
coping strategies (see Carver, Scheier & Weintraub, 1989). For example,
predictive secondary control functions to protect the person from disappointment
whilst “positive reinterpretation” (an emotion-focused strategy) serves to reduce
emotional distress associated with life events (Carver et al., 1989 p.269).
1.3.3 The Expanding Classification of Secondary Control Strategies
Whether one conceives of secondary control in terms of the specific processes
outlined by Rothbaum et al. (1982), more generally in terms of acceptance
(Thompson et al., 1994) or in terms of the related construct of emotion-focused
coping (Weisz, 1990), the concept of controlling the psychological impact of
events has led to a growing number of psychological strategies being conceived
in terms of secondary control processes.
21
21
Chapter 1179
For instance, Weisz (1990) described the process of selective attention as a
secondary control strategy in which the person focuses attention away from a
problem so as to reduce the impact of the aversive thoughts and feelings
associated with it.
Heckhausen and Schulz (1995) described the process of socially comparing oneself with people worse off as a secondary control strategy that helps a person accept poor outcomes. In addition, attributional processes involving attributions for failure to external causes and success to internal causes were construed as a secondary control strategy that allows people to protect and control their level of self-esteem.
Furthermore, Cummins (1997a) identified fifteen separate secondary control processes (such as optimistic expectancies and creating worse case scenarios) that allow people to maintain a sense of control over their external and internal environments (self-esteem and subjective well-being).
1.3.4 Primary Control Strategies
According to Rothbaum et al. (1982) and Weisz (1990) the strategies of
predictive, illusory, vicarious and interpretive secondary control also have their
primary control counterparts. For example, predictive primary control may
involve trying to accurately predict, select and execute the strategies that are
most likely to alter the environment to fit one's needs and wishes. Illusory
primary control may involve attempts to influence chance to allow a better fit
with the desires of the person (e.g. spitting on a coin before tossing or “riding a
lucky streak”). Primary vicarious control may result in a person emulating the
behaviour of powerful others in order to influence the environment. Lastly,
primary interpretive control involves attempts to make sense of a problem in
22
22
Chapter 1179
order to be able to master it, influence it and reduce it's likelihood of recurrence
(Weisz, 1990).
The analysis above suggests that both primary control and secondary control can
involve similar processes. For example, believing in another's power to control
(i.e. vicarious control) can be both a primary and secondary control strategy.
According to Weisz (1990), the principle distinction determining whether any
particular strategy is of the primary or secondary type must be inferred from the
person's overall reason or goal for control; either to change the objective
conditions of the environment (primary control), or to accept the objective
circumstances and seek to control the psychological impact of those conditions
(secondary control).
In summary, according to the two-process model of control, there are two paths
a person may pursue to develop a sense of personal control. Primary control is a
strategy involving attempts to influence or change objective conditions in the
environment. In contrast, secondary control involves accepting the existing
environmental conditions as they are, and focusing on controlling internal
processes in order to control the psychological impact of outcomes. Primary and
secondary control can be conceived in terms of specific strategies, as a general
strategy of acceptance, or in terms of problem-focused and emotion focused
coping. These various conceptions have allowed for a wide variety of cognitive
and behavioural strategies to be understood in terms of primary and secondary
control processes.
What then are the benefits or costs of seeking a sense of control, whether it be a
sense of primary control or secondary control?
23
23
Chapter 1179
1.4 The Relationship Between Primary/Secondary Control and
Psychological Adjustment: The Adaptive Value of Primary Control
A substantial body of empirical studies and literature reviews suggest a
significant relationship exists between primary control beliefs and levels of
psychological distress or adjustment to illness. This research suggests the
adaptive value of perceived primary control.
For instance, cancer patients’ levels of anxiety and depression have been found
to be significantly negatively correlated with primary control beliefs related to
control over physical symptoms (r = -.59), family relationships (r = -.29) and
relationships with friends (r = -.41) (Thompson, Sobolew-Shubin, Galbraith,
Schwankovsky & Cruzen, 1993). Taylor, Lichtman and Wood (1984) also
found, amongst a sample of women with breast cancer, that perceptions of
personal control and other's control (e.g. medical staff) over their cancer were
significantly related to better adjustment to their illness.
Thompson et al. (1994) found that primary control was significantly negatively
correlated with depression amongst a sample of HIV-positive men (r = -.40),
even when controlling for the amount of stressful life events associated with
living with HIV. Furthermore, primary control beliefs remained predictive of
less depression within a particularly low-control circumstance (HIV-positive and
incarcerated in prison) (Thompson, Collins, Newcomb & Hunt, 1996).
In addition, Weisz (1990) in a review of his own empirical studies, provided
evidence for the potential adaptive value of primary control beliefs in children.
24
24
Chapter 1179
Control beliefs were operationalised in terms of the child's belief in their
competence, contingency and control in solving problems at school and at home.
The operationalised focus on control as a method for solving external
circumstances lends itself to a primary control classification. Results indicated
that perceived competence beliefs accounted for approximately 14 percent (r =
-.38) of the variance in a measure of childhood depression. Furthermore,
contingency and control beliefs were also found to be significantly correlated
with children’s behavioural improvements in response to a six-month course of
psychotherapy (r = .48 and r = .41, respectively).
Research on the personality construct of hardiness (Kobasa, 1979) also indicates
the adaptive value of perceived primary control. After reviewing research
investigating the relationship between health and the individual components of
the hardiness construct, Wallston (1989) concluded that perceived control
(defined as the ability to influence events) is a major, if not the major component
of hardiness that explains the buffering effects of hardiness from physical and
emotional ill-health.
Similarly, Peterson and Stunkard (1989) in their review of control and health
research cite a large number of studies implicating the link between control
(defined as the ability to bring about good events and avoid bad events) and
physical and psychological health, claiming that the link is now “...well
established” to the point that precise mechanisms should be investigated (p.822).
1.4.1 The Adaptive Value of Secondary Control
25
25
Chapter 1179
The previous discussion of secondary control strategies provides a conceptual
basis for understanding how secondary control may relate to psychological
adjustment. That is, by definition, the strategies aim to reduce or control the
psychological impact of objective conditions. Moreover, empirical evidence
suggests that secondary control strategies do indeed relate to better
psychological adjustment.
Thompson et al. (1993) asked patients with cancer to rate the degree to which
they perceived they had control over the symptoms of the disease and control
over the negative emotions associated with a cancer diagnosis. Whereas
perceptions of control over the objective conditions of the cancer symptoms
reflect primary control, the perception of control over the emotional impact of
cancer reflects secondary control processes. They found that patients who
perceived control over their emotions accounted for 49% of the variance
(r = -.70) in patients emotional adjustment; defined by levels of anxiety and
depression. Secondary control strategies reported by the patients included,
keeping “faith” and a “positive attitude” (p.301).
By operationalising secondary control in terms of acceptance over outcomes
associated with HIV diagnosis, Thompson et al. (1994) also found that
secondary control significantly predicted lower levels of depression.
Carver et al. (1993) used an emotion-focused coping scale that measured
secondary control strategies including acceptance of stressful events, turning to
religion, positive reframing, and use of humour amongst women diagnosed with
early stage breast cancer. Results showed that acceptance, humour and positive
26
26
Chapter 1179
reframing all had significant associations with less distress. Furthermore, after
controlling for the inter-relationships amongst the various strategies, acceptance
and positive reframing remained unique predictors of distress.
Similarly, Carver et al. (1989) found that accepting stressful events was
significantly related to optimistic beliefs, whilst positive reinterpretation was
associated with optimism, feelings of control, self-esteem, hardiness and less
anxiety amongst a sample of undergraduates.
In addition, the tendency to re-interpret a stressful event (impaired fertility) as
leading to a strengthening in one's marriage, and attributing the loss to external
bio-medical causes, was shown to predict lower levels of psychological distress
(Mendola et al., 1990).
Overall, the studies described above would suggest that both primary and
secondary control strategies are associated with good psychological adjustment
to illness and positive mental health. Nevertheless, recent theoretical proposals
and research efforts have made an effort to determine which strategy (primary
control or secondary control) may play the more important role in adapting to
negative outcomes in life.
1.5 The "Primacy" of Primary Control
Heckhausen and Schulz (1995) proposed the functional primacy of primary
control over secondary control. In short, they argue that primary control has
greater functional importance for the individual than does secondary control.
There are two key aspects to this proposition. Firstly, the authors suggest that
27
27
Chapter 1179
primary control has greater adaptive value for the organism because it allows the
person to control their environment to fit the individual's particular needs;
therefore allowing the person to reach their "...full developmental potential"
(p.286). Hence, the theory proposes that an individual who manipulates
situations to get what they need, will actualise greater positive growth than
someone who accommodates to the existing conditions in order to control the
psychological impact of events.
Secondly, primacy theory argues that the function of secondary control is to act
as a compensatory mechanism to protect the person against actual or anticipated
losses in primary control (Heckhausen and Schulz, 1995). According to the
theory, when an individual anticipates failure or actually does fail (i.e.
anticipates or experiences a loss of primary control), secondary control strategies
help the person to re-establish their sense of primary control over the
environment. For example, when an employee feels that they have failed,
attributing their performance to external factors such as over-work and stress
(rather than personal factors such as competence) protects their self-esteem and
assists them in re-developing a sense of personal efficacy or control over the
demands of their job.
1.5.1 Empirical Tests of Secondary Control as Simply a Back-Up Strategy
for Low Primary Control
Thompson et al. (1994) conducted a study amongst people living with HIV to
test whether secondary control acted as a compensatory (back-up) mechanism
for people with low primary control. Hence, they hypothesised that secondary
control would be associated with less depression amongst individuals with low
28
28
Chapter 1179
perceptions of primary control, but for those with higher perceptions of primary
control, secondary control would not be associated with less depression. The
researchers claimed that the hypothesis was supported since there was only a
"weak" relationship between secondary control and depression amongst those
with high primary control (p.543). In contrast, for those with low primary
control, secondary control was associated with less depression.
However, due to the lack of thorough statistical reporting it is difficult to assess
the exact degree to which the hypothesis was supported. That is, no
correlational analysis is reported which shows the relative associations between
secondary control and depression amongst high and low primary control sub-
groups. Instead, the reader is referred to a plot, without scales, plotting
secondary control against depression. The absence of scales on the plot and the
unreported correlations make it impossible for the reader to judge how “weak”
the relationship between secondary control was for those with high perceived
primary control. According to primacy theory, if secondary control acts only as
a back-up strategy then secondary control should be uncorrelated with
depression amongst individuals with adequate levels of primary control.
However, there is no statistical report of the hypothesised lack of relationship.
Thompson et al. (1996) repeated the hypothesis of secondary control as a back-
up strategy amongst a group assumed to be characterised by particularly low
levels of perceived primary control (prison inmates with HIV). The results
showed that secondary control did not predict levels of depression amongst
individuals with low, nor high perceptions of primary control. Thus refuting the
hypothesis that secondary control acts as a back-up strategy for low primary
control. The fact that secondary control did not predict less depression in this
29
29
Chapter 1179
study was inconsistent with numerous other studies that have found secondary
control to be associated with less depression and distress (Taylor et al., 1984;
Carver et al., 1989; Mendola et al., 1990; Carver et al., 1993 & Thompson et al.,
1994). The researchers suggested that the sample in the study may have equated
acceptance with helplessness, and therefore secondary control did not predict
lower levels of depression.
The hypothesis that secondary control acts as a back-up strategy when primary
control is low, has received mixed empirical support. Moreover, some research
suggests that secondary control may have a greater role for adaptive functioning
than simply acting as a back-up strategy. For example, cross-cultural research
suggests that while Western culture emphasises the importance of primary
control, Eastern culture emphasises the greater importance of secondary control
over primary control.
1.5.2 Cultural Differences in Preferences for Primary Versus
Secondary Control: Is Secondary Control Used Only as a
Back- Up Strategy ?
Weisz et al. (1984) argue that Western culture encourages and emphasises the
value of primary control, whereas Japanese culture prefers the use of secondary
control. They cite numerous examples of differences in child-rearing practices,
socialisation, work practices, religion, philosophy and psychotherapy which
reflect Japanese culture's greater preference for secondary control over primary
control. For instance, Western schools of psychotherapy (e.g. behaviour therapy
and psychoanalytic therapy) often emphasise the importance of reducing
symptoms and changing behaviour problems. In contrast, the Japanese Morita
30
30
Chapter 1179
and Naikan schools of psychotherapy emphasise the importance of reducing
distress by accepting one's symptoms and developing a sense of meaning to
those symptoms through re-interpretation.
Furthermore, Azuma (1984) pointed out numerous popular Japanese aphorisms
which view the ability to yield and accept (as opposed to asserting oneself over
others and situations) as a sign of self-control, maturity and flexibility. For
example, when Japanese children argue, parents may gesture the phrase "...to
lose is to win"; meaning that it is more highly valued to yield than assert oneself
(p.970).
Although researchers have described Japanese culture as emphasising a greater
preference for secondary control (Weisz et al., 1984; Azuma, 1984; Kojima,
1984), the role of yielding and accepting circumstances is not absent from
Western culture. Rosenberg (1990) cited the Western fable of the fox and the
grapes, in which the fox's attempts at primary control (reaching for the grapes)
fails, leaving the fox to ameliorate his disappointment by declaring the grapes
were sour (secondary control).
In addition, cognitive therapy often teaches clients skills aimed at controlling
one’s internal cognitive/affective reactions to situations, rather than changing the
situation per se. For example, Burns (1980) gives the example of a mother who
berates herself as a “bad mother” (p.73). Rather than developing primary goals
for becoming a ‘good mother’ (e.g. by spending more time with the children),
cognitive therapy is aimed at teaching her not to interpret her self in terms of
dichotomous categories such as good and bad. By learning to perceive her
mothering skills more accurately, and manipulating her attentional focus to her
31
31
Chapter 1179
positive actions (not just negative ones), the client seeks to control their
depressed emotional state rather than the perceived mothering problem.
Furthermore, as discussed previously, Western patients with cancer and HIV
have been found to use secondary control strategies such as humour, acceptance,
re-interpretation and optimistic beliefs in order to cope (Mendola et al., 1990;
Carver et al., 1993; Thompson et al., 1994).
In summary, it has been suggested that certain cultures may actually show a
preference for secondary control and that the potential benefits of secondary
control are also evident in Western culture. Combined with the previous
analysis that indicated mixed empirical support for the back-up hypothesis of
secondary control, it is possible that secondary control strategies have intrinsic
value beyond their potential to compensate for low levels of primary control.
That is, although the compensatory role of secondary control as proposed by
Heckhausen and Schulz (1995) is intuitively logical and supported by some
evidence, the current analysis suggests that secondary control may have
functional value in its own right.
In short, conceptualising secondary control as merely a back-up strategy for
primary control potentially under-rates the adaptive value of secondary control
strategies as proposed by the original authors of the primary/secondary control
construct (Rothbaum et al., & Weisz et al., 1984).
1.5.3 Empirical Tests of the Greater Adaptive Value of
Primary Control
32
32
Chapter 1179
Thompson et al. (1996) hypothesised that primary control has greater adaptive
value than secondary control (acceptance). That is, primary control would be
more strongly associated with less psychological distress than secondary control.
The researchers claim that this hypothesis was supported because primary
control predicted levels of distress when controlling for secondary control.
Whereas secondary control did not account for any significant variance in levels
of distress when primary control was controlled for.
However, the finding that perceptions of secondary control added no prediction
of distress above that offered by primary control is not surprising considering
that secondary control was found to be significantly associated with greater
psychological distress. Although this association was statistically significant for
only one of the four data sets (measured at different times), the direction of the
correlation between distress and secondary control was always positive. Thus, it
remains to be seen whether the “...greater adaptive value” of primary control
(Thompson et al., 1996 p.1307) is maintained in a sample that shows secondary
control to be helpful in the first place. As the evidence stands now, it can only
safely be said that primary control is more adaptive than secondary control when
the sample perceives secondary control as unhelpful.
In summary, only two empirical studies have hypothesised and explicitly tested
the primacy of primary control theory (Thompson et al., 1994; and Thompson et
al., 1996), resulting in mixed support for the potentially superior benefits and
greater functional importance of primary control versus secondary control. In
addition, difficulties associated with the statistical reporting and samples used in
these studies throw some doubt onto the precise degree of support for primacy
theory. Regardless of the above analysis, Thompson et al. (1996) concluded that
33
33
Chapter 1179
“...there may be no limits to the benefits of primary control” (p. 1316).
However, other control research is not so supportive of the “limitless benefits”
of primary control.
1.5.4 The Adaptive Limits of Primary Control
As discussed previously, theoretical as well as empirical evidence demonstrate significant associations between primary control and positive psychological adjustment, suggesting the adaptive value of primary control. However, it has also been suggested that primary control may be associated with reduced psychological and physical health.
In terms of psychological health, Shapiro and Shapiro (1984) discussed a model
of intimate relationships in which issues of control are seen as central to the
adaptive or non-adaptive functioning of the relationship. According to their
model, “assertive control” over situations and people (i.e. primary control) may
be positive in that it initially brings the person (and the couple) a sense of
effectiveness, achievement and competence (p.97). However, if one partner
develops a reliance on primary control over time, this may adversely impact on
the quality of the relationship. A partner who relies on primary control may be
perceived as over-controlling, domineering and aggressive. Without the
knowledge and ability to yield some primary control and to accept another’s
views on some issues (secondary control), conflict may result.
Thompson et al. (1988) reviewed a number of control studies that suggested
negative effects of perceived control over situations. Their review suggested
that there may be many factors that determine whether perceived control will be
34
34
Chapter 1179
helpful or harmful to the individual. The relevant factors included: the amount
of effort and attention needed to exercise control over a stimulus, whether
attempts to exercise control were successful, the preferred coping style of the
person and the actual amount of perceived control.
Considering this last factor, research suggests that higher opportunities for
primary control can be associated with increased levels of distress. Mills and
Krantz (1979) showed that the more opportunities to control the situation offered
to patients giving blood, then the greater the level of patient distress. Likewise,
Corah and Boffa (1970) found that both a behavioural and decisional mode of
control independently reduce levels of distress in subjects exposed to
experimental threat. When given the opportunity to exercise both modes of
control, subjects reported as much discomfort as they did when given no
behavioural or decisional control over the threat. It was not clear why more
opportunity for primary control was associated with increased stress in these
studies. As Averill (1973) commented, whether having control is helpful or
harmful in reducing stress depends on the context in which the control response
is made. Hence, Mills and Krantz (1979) suggested that people in a blood-
doning procedure may have preferred a nurse to take full control. According to
the authors, the greater opportunity for control over a stressor may heighten
stress in situations in which the person does not want much control.
Other evidence indicates that having a high desire for control or perception of
control over one’s environment may be maladaptive when the environment is
objectively difficult to control, or does not require such high levels of personal
control. Evans et al. (1993) reviewed control research to demonstrate that
persons with a high desire for control, high competencies for control and high
35
35
Chapter 1179
levels of perceived control, are more vulnerable to stress and depression within
low-control environments. According to their review, the increased
vulnerability may occur because high-controlling people may be unable to
accommodate or accept situations that afford a low degree of control. For
example, Collins, Baum and Singer (1983) found that residents involved in the
Three-Mile Island power plant disaster showed lower degrees of stress (self-
reported, psychophysiological and task performance) when they coped by
accommodating to the reality of the situation, compared to those who attempted
to instrumentally change the situation through political action and complaints.
Similarly, when exposed to uncontrollable stimuli, individuals with a high desire
for control were found to be significantly more prone to learned helplessness
effects than those with a lower desire for control (Burger and Arkin, 1980).
Furthermore, perceiving that others have a degree of control over the threat of
disease (primary vicarious control) was only helpful when others were actually
able to exert control (Helgeson, 1992). When the vicarious control of medical
staff was not evident, primary vicarious control predicted poor psycho-social
adjustment to cardiovascular illness.
In addition, Strube and Werner (1985) found that Type A persons had such a
high need for control over their environment that they had difficulty in
relinquishing control of a task to a superior performing partner. By failing to
relinquish control to others, the researchers concluded that Type A persons may
create high levels of stress by increasing their workload and job involvement,
eroding relationships and feeling role-conflict in managerial jobs that require
delegation of tasks. Indeed, partial support for these conclusions were provided
by Kirmeyer and Biggers (1988) who found that Type A individuals
36
36
Chapter 1179
systematically create a more demanding and more stressful work environment
for themselves.
Some evidence also suggests that a high desire and perception of primary
control may differentiate clinical and non-clinical populations. For instance,
Shapiro, Blinder, Hagman and Pituck (1993) found that perceiving too much
control over situations and people (over-control) is a factor that discriminated
between women with eating disorders and those with healthy eating habits.
Research also indicates that high levels of perceived primary control may not be
advantageous in terms of physical health. Using a structured Type A interview,
researchers have found that the behavioural cluster of being verbally
competitive, alert and quick to respond was prospectively associated with
12. Prim -.5448. Prim -.4939. Sec .528 -42614. Sec .526 .497
Finally, reliability analysis of the PSCS items indicated the items have good
internal consistency as indicated by Cronbach’s Alpha = .90.
4.2.2 Possible Interpretations of the Negative Loadings Observed for
Factor 2
The negatively loaded items that comprise the primary control factor (Factor 2)
suggests that the items should read as the polar opposite concept to what the item
states. Hence, the polar opposite concept of the primary control items of Factor 2
reflect an absence of primary control. That is, “not doing” whatever the strategy
states. For example, item 18 states that “when something gets in the way of a
goal, I work out how to remove it”. However, since none of the items were
50
50
Chapter 4
designed to be reversed scored, the negative loading suggests the item should read
“…not working hard to achieve goals”.
An explanation for the negative loadings of Factor 2 may be provided by
Rothbaum et al. (1982). They believe many learned-helplessness researchers
confuse passive behaviours (e.g. externally attributing outcomes to limited ability
or chance factors, behavioural disengagement etc.) with primary control
relinquishment, but may actually be a form of secondary control that may increase
levels of perceived control.
In their view, “inward behaviours” (being passive, withdrawing from active
problem solving, being submissive to environmental events and making causal
attributions to limited ability, chance or more powerful others; p. 5), rather than
being evidence for the relinquishment of control, may represent a different type of
inward-control that allows for the control of emotional and cognitive reactions to
problems (secondary control).
Perhaps then, the apparent ‘giving-up’, passive behaviour that the negative
loadings suggest (i.e. “not working hard to achieve goals”, “not learning new
skills”, “not removing blocks” etc.) are the reflection of the intrinsic passivity of
secondary control toward changing the environment. That is, behavioural
disengagement. According to Rothbaum et al. (1982) and Heckhausen and Schulz
(1995), the adaptive value of behavioural disengagement is that it avoids what
they argue is one of the most deleterious psychological experiences for a human
being, disappointment. Disappointment is particularly deleterious in that it
represents a failure of both control process options, primary and secondary
control failure. That is, disappointment arises from failing to bring the
environment in line with subjective wishes (primary control failure) as well as
failing to successfully accommodate emotionally to the situation (secondary
control failure). In addition, the adaptive value of behavioural disengagement is
that it is the necessary first step toward coping with primary control failure. If
51
51
Chapter 4
one cannot control the environment to achieve their desired goal, continuing to
try change the situation may incur costs of an escalating perceived sense of loss
of control. This may include all the associated cognitive and emotional reactions
that may stem from such an experience (e.g. loss, frustration, anxious uncertainty,
negative self-talk etc.). Hence, by behaviourally disengaging from the goal, the
person can begin to manage these cognitive-affective reactions. In this way,
behavioural disengagement may be a gateway to secondary control. That is, by
giving-up on trying to change the environment the person can then allocate
greater effort at maintaining control of their self.
Another possible explanation for the negative loadings being investigated is that
they reflect the control processes of a relatively aged sample (mean = 45.77
years). Heckhausen & Schulz (1995) theorised that the prevalence of primary
control reduces with increasing age across the life-span. Hence, perhaps items
loading onto the primary control factor are negatively loaded because the older
sample used in the analysis responded in a way that reflected a relative absence of
primary control efforts. To test this hypothesis, the sample was stratified by age
(median split) into two groups: > 45 years (N = 93) and < 45 years (N = 93).
Separate principal-components analyses were then conducted on both the younger
and older portions of the sample.
Results showed a number of significant changes in the factors and their loadings
following stratification by age. For the younger portion of the sample (< 45
years) the primary control factor had 3 primary control items (items 12, 16 and
21) and one secondary item loading (item 2). These primary control items
previously loaded negatively onto the general secondary control factor extracted
in the first analysis. However, when analysing only the younger portion of the
sample, the same primary control items (items 12, 16 and 21) now positively load;
indicating the presence, not the absence of these primary control strategies in the
younger participants responses. In fact, all the negatively loading primary control
items from the first analysis now loaded positively on factors derived from the
52
52
Chapter 4
older portion of the sample. In behavioural terms, instead of the primary control
items reflecting the polar opposite concepts to those stated (e.g. “… do not learn
the necessary skills…” or “…I do not invest as much time as I can…” etc.) the
positive loadings suggest the item should now read as stated (i.e. “…I learn the
necessary skills…” or “I invest as much time as I can…”).
Interestingly, in the analysis of the older portion of the sample (> 45 years), a
well-defined primary control factor did not clearly emerge at all. Primary control
items tended to load across Factors 3, 4, and 5 along with numerous secondary
control items amongst these factors. This complex structure for Factors 3, 4 and 5
shows that no definite primary control factor emerged amongst the older portion
of the sample. In addition, the analysis of the older portion of the sample
extracted a greater representation of secondary control factors (two secondary
control factors) compared to the younger portion of the sample where only one
secondary control factor emerged.
These findings demonstrate that age can effect the nature of primary control item
loadings. That is, by analysing only the younger portion of the sample the
direction of primary control item loadings reversed direction from negative to
positive. Additionally, the pattern of loadings was also affected by the age of the
sample, from a clear primary factor emerging in the younger sample, to no
definite primary control factor emerging in the older sample. Furthermore, the
analysis of the older group yielded more secondary control factors than either the
analysis of the younger group or the sample as a whole. Taken together, these
findings may suggest that more elderly participants’ responses reflected a lesser
engagement of primary control efforts, but a greater engagement of secondary
control strategies when coping with problems. However, although the age of the
sample appears to effect the pattern and direction of loadings, other evidence
suggests that the older portion of the sample perceived they used as much (on
average) primary and secondary control as the younger portion of the sample.
Table 4.3 shows that by calculating the means for Factor 1 (the general secondary
53
53
Chapter 4
control factor extracted in the first analysis) and Factor 2 (the negatively loaded
primary control factor) we find that the younger sample’s mean levels of primary
and secondary control differ only marginally from the older sample means.
Table 4.3 Means and standard deviations (sd) for Factor 1 (primary control) and Factor 2 (secondary control) for the total sample (N = 192), participants < 45 years of age, and participants > 45 years
Total sample
(N = 192)
< 45 years
(N = 93)
> 45 years
(N = 93)
Variable mean sd mean sd mean sd
Factor 1 (Primary)
Factor2 (Secondary)
40.95
42.63
10.72
8.86
39.89
41.58
10.88
9.25
43.56
42.32
8.44
10.61
Therefore, although some evidence suggested that the older group used less
primary control than the younger group, the actual mean levels of primary and
secondary control are nearly equal. Hence, it appears the negative loadings
derived from the initial analysis of the whole sample are better explained by an
intrinsic characteristic of secondary control involving a passivity toward changing
the environment; rather than the effect of an age bias that is theoretically linked to
reduced levels of primary control with increasing age.
4.3 Analyses of the Hypotheses for Study 1
One of the aims of the study was to investigate the importance of the individual
maintaining a balance between their primary and secondary control processes.
Specifically, it was hypothesised that control-imbalanced groups (i.e. above-
average on one control process, but below average on the other) would self-report
less Subjective Quality of Life (SQOL), Positive Affect and Positive Thinking
compared to individuals who demonstrate a relative balance between their control
processes (i.e. average or above-average levels of both primary and secondary
control).
54
54
Chapter 4
To test this hypothesis an independent variable (Group) was specified by splitting
the sample into four groups:
Group 1 (above-average primary control combined with below-average secondary
control).
Group 2 (below-average primary control combined with above-average secondary
control).
Group 3 (average and above-average levels of both primary and secondary
control).
Group 4 (below average levels of both primary and secondary control).
To test the hypothesis a one-way Multivariate Analysis of Variance was
performed using SPSS GLM-Multivariate on SPSS for Windows (Version 8.0) on
the three dependent variables (SQOL, Positive Affect and Positive Thinking).
The number of cases relative to the number of dependent variables was 7:1 in the
cell with the smallest N (N=21), indicating adequate cell-sizes. Normality of the
dependent measures was assessed by group via skewness statistics and inspection
of histogram plots using SPSS Explore. With the exception of the distribution of
SQOL scores for group 1, all other groups for each of the dependent variables
were negatively skewed. Skewness statistics for the groups ranged from -.131 to -
1.35 indicating that participants scores on measures of positive adjustment were
typically toward the upper limits of the distribution of scores. The decision was
made not to transform any of the data since Subjective Quality of Life is known to
be naturally negatively skewed (Cummins, 1995). Since Positive Affect and
Positive Thinking are conceptually and empirically related dimensions to
Subjective Quality of Life these distributions are assumed to also naturally skew.
Hence, none of the dependent variables were transformed. Furthermore, with
large samples (>100 participants), variables displaying moderate levels of
skewness can be tolerated (Tabachnick & Fidel, 1996). Univariate and
multivariate outliers were also retained in the analysis because they represent true
55
55
Chapter 4
observations of participants’ self-reported degree of adaptive functioning. Box’s
M was not significant, indicating multivariate homogeneity of variance between
groups (18, 27554) = 31.93, p > .01.
Wilk’s lambda F(9, 430) = 15.59, p < .001, ß = .20, revealed a significant global
difference between the four groups. Groups differed on all the dependent
measures of SQOL, Positive Affect and Positive Thinking as shown in Table 4.4.
Table 4.4 Summary table for Multivariate Analysis of Variance examining group differences in Subjective Quality of Life, Positive Affect and Positive Thinking for the four combinations of Primary & Secondary Control
Groups
Variable 1
(Hi prim-Lo Sec)
2
(Lo Prim-Hi Sec)
3
(Hi Prim-Hi Sec)
4
(Lo Prim-Lo Sec)F p ²
Subjective Quality of Life
X 52.50SD 9.93
X 51.24SD 6.64
X 58.08SD 6.97
X 43.69SD 9.97
31.10 .001 .343
Positive Affect
X 74.17SD 11.36
X 70.10SD 10.30
X 78.44SD 8.98
X 59.08SD 13.30
34.55 .001 .367
Positive Thinking
X 192.60SD 48.07
X 188.14SD 35.87
X 233.38SD 39.35
X 155.93SD 52.16
32.69 .001 .354
Post-hoc Tukey’s analyses revealed that both control-imbalanced groups 1
(HiPrim-LoSec) and group 2 (HiSec-LoPrim) reported significantly less SQOL
compared to the group that maintained average or above average levels of both
primary and secondary control processes (i.e. the control-balanced group). In
terms of Positive Affect, the HiSec-LoPrim group reported significantly less
positive emotions compared to the HiSec-HiPrim group. However, the HiPrim-
LoSec group were not significantly different on positive emotions compared to
the control-balanced group. In short, low levels of secondary control skills did not
appear to effect Positive Affect. In contrast, persons who reported low levels of
primary control (despite having high levels of secondary control) did have less
Positive Affect than persons with high levels of both primary and secondary
56
56
Chapter 4
control. Hence, primary control appeared to affect Positive Affect, whereas
secondary control did not.
For Positive Thinking, both control-imbalanced groups 1 (HiPrim-LoSec) and 2
(HiSec-LoPrim) reported significantly less positive thoughts than the control-
balanced group. Interestingly, the HiSec-LoPrim group reported very low levels
of positive thoughts that were not significantly different from the control
relinquishment group (LoPrim-LoSec). The control relinquishment group was
found to have significantly less Subjective Quality of Life, Positive Affect and
Positive Thinking compared to all other groups (control-balanced or imbalanced
groups).
Regression analyses were also conducted in order to test whether secondary
control appeared to act only as a back-up strategy for low primary control. The
cases to variables ratio was far in excess of 20:1. Outliers were retained as
before, and the assumption of linearity was met. The normal probability plot of
standardised residuals demonstrated normal distribution of the residuals.
To test the role of secondary control as simply a back-up strategy for low primary
control, a median-split of primary control was taken, creating a low (N = 83) and
high (N = 92) primary control group. Table 4.5 shows that, primary control was
entered first into a hierarchical regression predicting of SQOL, followed by
secondary control for both the low and high primary control groups.
Results showed that within the low primary control group, primary control
significantly predicted SQOL accounting for 6.1% of the variance. However,
when secondary control was also entered into the prediction of SQOL, primary
control ceased to make a significant prediction, with secondary control accounting
for 7.3% of the variance in SQOL.
57
57
Chapter 4
For the high primary control group, primary control significantly predicted
SQOL, accounting for a total 20.4% of the variance. When secondary control was
also entered into the regression, it also significantly predicted SQOL, accounting
for an additional 3.1% of the variance. The squared semi-partial correlation
showed that secondary control accounted for a unique 5.0% of the variance in
SQOL, compared to 16.8% variance uniquely accounted for by primary control.
Table 4.5 Summary table for hierarchical regression analyses investigating the
relationship between Primary/Secondary control and Subjective Quality
of Life for participants with high and low levels of primary control
Low primary control (N= 83) High primary control (N = 92)
Variable Entry
ß t Adj. R²(R² Change)
Semi-partial r( r²)
ß t Adj. R²(R² Change)
Semi-partial r( r²)
1.Primary Control
.270 2.52* .061(6.1%)
.270(7.3%)
.461 4.93** .204(20.4%)
.461(21.3%)
2.Primary Control
.112 .954 .106(1.1%)
.404 4.24** .410(16.8%)
Secondary Control
.327 2.786** .134(7.3%)
.297(8.8%)
.207 2.17* .235(3.1%)
.224(5.0%)
**p <.01* p <.05
The study also aimed to conduct an initial exploration of the relationship between
primary/secondary control and positive measures of psychological health (e.g.
Positive Affect, Subjective Quality of Life) rather than measures of negative-
distress (e.g. depression, anxiety etc.).
Standard multiple regressions were used to predict SQOL, Positive Affect and
Positive Thinking from primary and secondary control. Results are shown in
Table 4.6.
58
58
Chapter 4
Table 4.6 Standard regression analyses investigating the relationship between Primary/Secondary control and measures of positive psychological adjustment
Primary Control Secondary ControlVariable ß t Semi-
Partial rß t Semi-Partial r
SubjectiveQuality of Life
.398 5.60** .378.
.275 3.87** .272
Positive Affect .474 7.28** .390 .330 4.60** .322
Positive Thinking
.337 5.10** .350 .425 6.44** .427
**p<.001
Primary and secondary control significantly predicted Subjective Quality of Life,
accounting for a total of 35.2% of the variance. Squared semi-partial correlations
showed that primary control accounted for 14.3% of the variance and secondary
control 7.4%.
Primary and secondary control also significantly predicted Positive Affect,
together accounting for a total of 47.1% of the variance. Primary control
accounted for 15.2% of unique variance and secondary control 10.3% unique
variance.
Similarly, Positive Thinking was significantly predicted by primary and
secondary control, accounting for a total 45.4% of the variance. Primary control
uniquely accounted for 12.3% of variance, compared to 18.2% of the variance
explained by secondary control.
59
59
CHAPTER 5
DISCUSSION OF STUDY 1
5.1 The “Primacy” of Primary Control
Previous theory and research investigating the relationship between
primary/secondary control and psychological adjustment suggested that primary
control has functional primacy over secondary control (Heckhausen & Schulz,
1995; Thompson et al., 1994; Thompson et al., 1996; Thompson & co-workers,
1998). However, the present data suggests a more prominent role for secondary
control, in accommodating to problems for psychological health.
Results demonstrated that rather than high levels of primary control being
optimally adaptive, it is important to consider the relative balance or ratio of
primary to secondary control levels. As hypothesised, results showed that
individuals who reported an imbalance between their levels of primary and
secondary control reported less SQOL, less Positive Affect and less Positive
Thinking than individuals who reported average (or above average) levels of both
control processes.
One explanation for the greater reported psychological health of the control-
balanced group compared to unbalanced groups may lie within the comprehensive
nature of the coping skills available to them by maintaining adequate levels of
both primary and secondary control. Without a comprehensive base of coping
skills involving both control processes, the person may be vulnerable to situations
that do not match the preferred control process.
60
60
Hence, those with high levels of primary control combined with low levels of
secondary control are vulnerable to situations that may not be easily controllable
by direct action (e.g. someone resents you). In contrast, those with high levels of
secondary control skills but with low primary control skills are vulnerable to
missing opportunities for directly manipulating situations to fit their needs. Evans
et al. (1993) cited numerous empirical studies that demonstrate the dysfunctional
nature of exercising high levels of control behaviour, motivation or cognitions in
the context of low environmental opportunities for control. However, individuals
who utilize both control process possibilities, have a more comprehensive
repertoire of coping skills allowing them to adapt their coping response to the
objective controllability of the situation.
The present data suggest that simply having average to high levels of primary
control skills is not optimally adaptive, one must also have average to high levels
of secondary control. One possible exception however, is for individuals whose
control imbalance is expressed in terms of high primary control combined with
low secondary control. This group reported significantly less SQOL and positive
thoughts compared to the control-balanced group, but not less Positive Affect.
The explanation for this difference may be less sensitivity of Positive Affect, than
the other positive adjustment variables to variations in secondary control.
Theoretically speaking, Positive Affect may be less sensitive to variations in
secondary control due to the relative stability of the Positive Affect dimension
compared to SQOL and Positive Thinking.
Watson and Clarke (1984) operationalised the PANAS scales to conceptualise
affect as a mood-like personality variable in which individuals were predisposed
toward perceiving events either with a negative bias (Negative Affect) or a
positive bias (Positive Affect). Recent empirical evidence supports the
dispositional quality of Positive and Negative Affect showing the variables are
69
69
closely related to the personality constructs of introversion-extraversion and
neuroticism, and that the personality variables ceased to predict occupational
strain after the PANAS variables were included in the model (Fogarty, Machin,
Sutherland, Lalor & Revitt, 1999).
In contrast to the enduring, trait-like quality of Positive and Negative Affect,
subjective evaluations of quality of life (SQOL) and the frequency of positive
thoughts over the previous week (Positive Thinking) are variables that are less
dispositional in quality as they are unstable constructs that would fluctuate with
the regular changes in circumstances and events. Perhaps Positive Affect was not
effected by secondary control in the MANOVA, whereas SQOL and Positive
Thinking were, because secondary control strategies function at the level of
effecting individuals daily appraisals and evaluations of their life (subjective
states), not at the level of dispositional perceptual biases (subjective traits).
For example, secondary control strategies such as finding meaning in an event or
downward social comparison may result in the reduction of negative cognitive
and affective states, but they might be less effective at changing the person’s
enduring, characteristic way of perceiving the world. Hence, secondary control’s
function in coping may be more to control positive and negative cognitive-
affective states than it is to control enduring trait-like perceptual biases.
In contrast, results showed that persons with low levels of primary control (the
HiSec-LoPrim group) did have less Positive Affect than the control balanced
group, suggesting that variations in primary control did effect the trait Positive
Affect. This result may be evidence of the role of primary control in effecting the
more enduring, less flexible aspects of psychological functioning (i.e.
personality), whereas secondary control may not function at the level of traits.
Nevertheless, the generality of the finding of Positive Affect being less sensitive
to variations in secondary control (compared to the other two positive adjustment
70
70
variables) was tested again in Study 2. In addition, the robustness of the finding
that suggested the importance of a control-balance for optimal psychological
adjustment was also re-tested in Study 2.
Previous theory and research on primary/secondary control also suggests that the
functional role for secondary control was to simply act as a back-up strategy for
individuals or circumstances where primary control was low (Heckhausen &
Schulz, 1995; Thompson et al., 1994). However, the present data show that
secondary control maintains a significant relationship with SQOL, even when the
person perceives primary control opportunities as high.
This suggests that secondary control has intrinsic adaptive value for the individual
regardless of the level of primary control. The intrinsic value of secondary
control for subjective quality of life may simply be that by engaging in secondary
control cognitions or behaviours (e.g. doing something enjoyable to relax, or
telling oneself that one is better off than others), the person perceives a better
quality of life, regardless of the perceived controllability over objective
environmental conditions.
In order to test the robustness of the finding of secondary control’s relationship
with psychological adjustment amongst both people with high and low levels of
primary control, the observed relations were re-tested in Study 2.
Another aim of Study 1 was to explore the relationship between
primary/secondary control and positive adjustment variables, rather than
investigating the relationships with negative-distress variables that have been
explored in previous studies. The data showed that secondary control strategies
also predict Positive Thinking, Positive Affect and Subjective Quality of Life.
Hence, secondary control strategies may not only function to decrease negative
emotionality but may also increase positive cognitions, and subjective perceptions
of quality of life. Although Positive Affect was predicted by secondary control in
71
71
the regression analyses, the results of the MANOVA analysis suggests that
secondary control may have less effect on the trait-like constructs such as Positive
Affect compared to the effect on subjective states such as Subjective Quality of
Life and frequency of Positive Thinking over the past week.
5.1.1 Differences With Past Theory and Research: The Importance of
Operational Definitions of Primary and Secondary Control in Coping
Research
The relative contributions of primary versus secondary control in explaining
unique variance in positive adjustment variables, indicates that the relative
valence of secondary control’s effect on positive psychological adjustment may
be considered less than the effect of primary control.
Hence, previous research that investigated the relationship between the two-
process model and distress variables (Thompson et al., 1994; Thompson et al.,
1996; Thompson & co-workers, 1998) proposed that the greater portion of
variance in distress variables explained by primary versus secondary control
demonstrates the greater functional importance of primary control compared to
secondary.
However, the smaller unique contribution to positive adjustment variables by
secondary control compared to primary control, disguises the importance of
secondary control to psychological health and the adaptive limits to primary
control. That is, primary control may account for more unique variance in
Subjective Quality of Life, but not if the person perceives low levels of primary
control. For individuals with low levels of primary control, primary control did
not predict Subjective Quality of Life at all, only secondary control significantly
predicted quality of life. In other words, secondary control has functional
primacy over primary control for individuals that perceive low levels of primary
control. Furthermore, even for participants with high levels of primary control,
72
72
secondary control was shown to provide an additional prediction of Subjective
Quality of Life beyond the effect of primary control, indicating secondary control
does not only act as a back-up strategy for low primary control. Still further,
participants with high levels of primary control, but correspondingly low levels of
secondary control reported a lower Subjective Quality of Life and significantly
less Positive Thinking compared to participants with high primary control and
high secondary control. Hence, maintaining high levels of primary control is only
functional if the person balances their control repertoire with average to high
levels of secondary control as well.
Whereas Thompson et al., (1996) commented “…there may not be limits to the
benefits of primary control” the current data suggests that the adaptive value of
primary control is indeed limited by the extent to which the person has also
developed their secondary control skills. Likewise, the adaptive value of
secondary control is limited by the extent to which the person has developed their
primary control skills. Excessive levels of primary control with corresponding
low levels of secondary control may be somewhat analogous to the Type A
behaviour pattern involving excessive levels of active control, with a sense of
time pressure and who are predisposed to become aggressive and hostile when
frustrated (Glass, 1977). In contrast, individuals who maintain high levels of
secondary control but correspondingly low levels of primary control may have
limits to their adaptive potential through missed opportunities to manipulate the
environment to fit their needs.
The more prominent functional role found for secondary control in this study
compared to previous research may partly be explained by differences in the
operational definitions of control and psychological adjustment. Thompson et al.
(1994; 1996) operationalised secondary control as the general degree of
“acceptance” of twenty-four different outcomes related to living with HIV
disease. To conceptualise secondary control in vague, non-specific terms as
general acceptance underestimates the variety and comprehensive nature of
73
73
cognitive and behavioural strategies that people use to reduce negative reactions
in anticipating or experiencing failure. By simply asking respondents how much
they accept negative outcomes in their life does not separate acceptance due to
reasons such as control relinquishment versus acceptance due to finding meaning
in the negative event. Furthermore, the non-specific description of secondary
control does not capture any of the positive processes that underlie secondary
control cognitions and behaviours (e.g. finding meaning, pampering oneself,
preparing for the worst etc.). Indeed, Thompson et al. (1996) operationalised
secondary control as general acceptance and found to their surprise that secondary
control related to increased levels of distress; not decreased levels of stress which
has consistently been suggested in other studies (Carver et al., 1989; 1993;
Thompson et al., 1994; 1998). In contrast, by operationalising secondary control
in terms of specific strategies that people use in order to positively accommodate
to problems, a greater functional role for secondary control for psychological
health was demonstrated.
The classification of particular control beliefs or behaviours into either primary or
secondary control has also been an issue of some contention in previous research
(e.g. Heckhausen & Schulz, 1995). Secondary control strategies typically alter
the internal world of the person (their beliefs about their problems, and various
self-talk strategies that aim to alter some negative cognitive/affective state). In
contrast primary control strategies typically aim to alter the external world (other
people, events etc.). In real terms however, all primary control strategies may
result in internal changes, just as secondary control strategies are believed to
result in renewed motivation to engage the environment in order to tackle
problems (Heckhausen & Schulz, 1995). Hence, primary and secondary control
are best discriminated by the most salient goal of the control strategy. Some
control/coping strategies are more clearly attempts to alter internal cognitions or
behavioural attempts at reducing negative affects (e.g. through exercise or doing
something pleasant for oneself), without engaging in direct manipulation of the
74
74
external world; whereas others are more clearly attempts to alter external
environmental processes.
5.2 Limitations to Study 1
One important limitation of the study is that the investigation of optimally
adaptive combinations of primary and secondary control does not refer to what is
optimally adaptive for a specific problem or at a specific time within the temporal
sequence of coping with a problem. Intuitively speaking, the optimally adaptive
responses for some problems may require different ratios of primary versus
secondary control behaviour, at different times. For instance, being physically
attacked may at first require a high degree of primary control response to ensure
safety, with little or no secondary control response. Later in time however,
secondary control responses may be important to assist the person to cope with
their limited primary control over the levels of poverty and violence in their
neighbourhood. Similarly, the diagnosis of a congenital disease with little known
treatment may require a high level of secondary control to limit negative
emotional reactivity, with little or no primary control response. Later however,
primary control responses may be important for managing the physical effects of
the illness. The present study conceptualised the balance between control
processes in terms of a balance in the individuals ‘coping repertoire’, not what is
adaptive for specific situations. Hence, the present data suggests that an
individual’s repertoire of coping skills would best reflect a relative balance of
average or above-average levels of both primary and secondary control.
A second limitation of the study is intrinsic to the non-experimental design used
for the study. The importance of maintaining a balance in levels of both control
processes assumes that the positive correlation between control and health
variables is underlied by causal processes. Whilst the cross-sectional nature of
the design does not infer underlying causality, most of the previous research on
the adaptive value of control did use experimental manipulations, suggesting that
75
75
control perceptions and behaviour causally effected various positive and negative
psychological outcomes (see Averill, 1973 for a review; Miller, 1979; Mills &
Krantz, 1979; Thompson, 1981). This is not to say that distress and positive
affects would not effect perceptions of control; intuitively speaking, feeling
depressed may lead a person to perceptions of reduced control over their self and
environment. Overall however, it is likely that the relationship between perceived
control and psychological health is bi-directional. Furthermore, from the clinical
perspective, it is less useful for psychological research to determine which
temporally occurs first, emotion or perception. Considering that psychological
interventions rely on manipulation of perceptions, coping research is best aimed at
understanding the role of perceptions in determining various psychological
outcomes. Interventions that aim directly at the manipulation of emotion with
little attention to perceptual processes would infer a more biological/medical
approach than a psychological one.
76
76
CHAPTER 6
INTRODUCTION TO STUDY 2
6.1 General Introduction
The aims of the second study were twofold. First, the hypotheses tested in
the first study were investigated with a second sample in order to see if the
findings could be replicated. Second, a new hypothesis was tested to
determine whether the relationship between control and adjustment could
be explained by the personality dimensions of extraversion and
neuroticism.
6.2 A Summary of Previous Theory and Evidence from Study 1: The
Importance of Maintaining a Balance in the Levels of Primary
and Secondary Control
Previous theory and research into the adaptive potential of primary and
secondary control suggested that primary control has functional primacy
over secondary control (Heckhausen & Schulz, 1993; Heckhausen &
Schulz, 1995; Thompson et al., 1994; Thompson et al., 1996). That is,
primary control is proposed to be intrinsically more adaptive, to be the
control strategy for which people have an underlying preference, and that
secondary control functions in a subsidiary capacity to primary control by
compensating for anticipated or real losses in primary control (failure).
However, data from Study 1 challenges the notion that primary control is
more adaptive than secondary control, suggesting that secondary control
may play a crucial role in maintaining optimal levels of positive
77
77
psychological adjustment. It was found that participants who reported a
control-imbalance between their levels of primary and secondary control
(i.e. average or above-average levels on one
78
78
control process but not the other) reported significantly lower positive
adjustment (Subjective Quality of Life and less Positive Thinking)
compared to people who reported a control balance (average or above-
average levels of both primary and secondary control). These results
suggest that individuals’ ratio of primary versus secondary control skills is
important to adaptive psychological functioning, rather than simply high
levels of primary control. Study 2 re-tested this hypothesis in order to
determine the robustness of the findings from Study 1.
6.2.1 Secondary Control’s Adaptive Function is More than a Compensatory
Mechanism for Low Primary Control
Results from Study 1 also found that the role of secondary control was
more than simply acting as a compensatory mechanism for low primary
control. For participants with low primary control, primary control failed to
predict psychological adjustment, whereas secondary control did. This
suggests secondary control may function as a compensatory control strategy
for people with low levels of primary control. However, for participants
with high levels of primary control, secondary control also explained
significant variance in psychological adjustment. This suggests that
secondary control’s functional role is more than simply compensatory. The
robustness of the finding that secondary control explained variance in
psychological adjustment for both groups of high and low primary control
was re-tested in Study 2.
6.2.2 The Relationship Between Primary/Secondary Control and Positive
Measures of Psychological Adjustment
Study 1 also provided data on the relationship between primary and
secondary control and measures of positive psychological adjustment, as
opposed to measures of negative adjustment or distress. Study 2 aims to re-
84
84
test the robustness of the observed relations between primary/secondary
control and measures of positive psychological adjustment.
Regression analyses from Study 1 showed that primary and secondary
control explained significant amounts of variance in all three positive
adjustment variables (Subjective Quality of Life, Positive Affect and
Positive Thinking). However, whilst secondary control significantly
predicted Positive Affect, independent of primary control, a split based on
secondary control did not distinguish between levels of Positive Affect,
even though it did distinguish between levels of the other two positive
adjustment variables. Particularly, participants who reported low levels of
secondary control, and high levels of primary control, reported as much
Positive Affect as participants with high levels of both primary and
secondary control. This was not the case for the other positive adjustment
variables (Subjective Quality of Life and Positive Thinking) in which
groups with low levels of secondary control reported lower Subjective
Quality of Life and less Positive Thinking. The explanation for this
difference may be less sensitivity of Positive Affect than the other two
positive adjustment variables, to variations in secondary control. This
result will also be re-tested in the second study.
6.3 Personality, Control and Psychological Adjustment
The major aim of Study 2 was to address the issue of the relationship
between primary/secondary control, personality and positive psychological
adjustment. Previous research has not addressed this question. However,
data from both the subjective well-being literature and prior research in the
conceptually related area of coping suggests that the personality dimensions
of extraversion and neuroticism might have a role in explaining the
relationship between primary/secondary control and positive psychological
adjustment.
85
85
The following discussion highlights the role of personality in positive
psychological adjustment. Conceptual linkages between the constructs of
coping and control are then discussed, whilst pointing out important
differences in the operational definitions between the control measure used
in this study and past measures of coping (the Ways of Coping models).
That is, the similarities between the constructs of coping and
primary/secondary control provide a firm rationale to expect that control
should also relate to the personality dimensions of extraversion and
neuroticism. However, the similarity between the constructs of coping and
control are more conceptual than operational, suggesting that coping
measures and the primary/secondary control scales used in this study are
not interchangeable. Next, evidence regarding the relationship between
coping and personality is discussed, which suggests that the conceptually
related constructs of primary/secondary control may also be related to
personality.
In short, the following discussion highlights the role of personality factors
in positive psychological adjustment. It is suggested that
primary/secondary control may relate to the personality factors of
extraversion and neuroticism (just as coping strategies are known to relate
to extraversion and neuroticism). Hence, Study 2 aims to extend the results
achieved from Study 1 by testing the capacity of the personality dimensions
of extraversion and neuroticism to explain the link between
primary/secondary control and positive psychological adjustment.
6.3.1 The Relationship Between Personality and Positive Psychological
Adjustment
Personality refers to aspects of thinking, feeling and behaving that are
relatively stable over time and throughout various situations. For instance,
a person may describe themselves as “generally quiet and shy”, referring to
86
86
psychological characteristics that are relatively unchanged as the person
moves between different situations and over time (Carlson, 1987). This
should be contrasted with aspects of behaviour that might be more closely
linked to situational cues rather than stable ways of typically thinking,
feeling and behaving.
Typically, personality theorists have studied personality from the trait-
approach, as opposed to early attempts to describe individual differences in
terms of personality types (e.g. see Carlson for an extended discussion on
“type” conceptions of personality). Traits are dimensions of common
psychological characteristics amongst people by which individuals are
believed to differ, not in quality of behaviour but quantity (Carlson, 1987).
A common approach of trait-theorists work over the last half-century (see
Cattell, 1946; John, Angleitner & Ostendorf, 1988) was the exploration of
thousands of various linguistic terms in English and other languages that
appear to describe particular characteristics of feeling, thinking and
behaviour that are common to peoples description of other people. Using
this lexical approach to identify traits, Costa and McCrae developed (see
John, 1990) one of the more recent and most comprehensive descriptions of
the structure of personality and the measurement of such, in their
instrument the NEO Personality Inventory-Revised (Costa & McCrae,
1992). Their “Big Five” conception of personality refers to a set of five
empirically derived, super-ordinate factors, or “Domains”, which are each
comprised of various inter-correlated “Facets” (traits)(p. 14). The five
factors, or domains of behaviour include Neuroticism, Extraversion,
Openness (to experience), Agreeableness and Conscientiousness.
Of particular interest to Study 2 are the domains of Extraversion and
Neuroticism which previous research has demonstrated are consistently
observed to correlate with a range of psychological indicators of positive
the variance in positive adjustment accounted for by secondary control amongst
people with high levels of primary control may be explained by the importance of
managing emotions during primary control striving.
For example, a worker’s ability to do their tasks (primary control) may be
impeded by them ruminating on the unresolved argument they had with their
partner before work. Furthermore, a reduced ability to control emotions in one
context (family) may begin to effect other people from other contexts (e.g.
colleagues at work), thus hindering others performance as well as their own. In
addition, when faced with more difficult challenges, secondary control strategies
may also allow for greater tolerance of aversive environmental conditions,
allowing the person more time and analysis of the problem situation and it’s
potential solutions. In contrast, a person who faces a stressful challenge without
adequate understanding of how to control their cognitive/emotional reactivity to
events may be prone to reducing their reactivity by making a more immediate
primary control response to the environment. This more immediate primary
control response may reduce the person’s level of discomfort, but it may not
provide adequate time to incorporate all the necessary information in order to
fully understand the problem and decide on the most adaptive solution, amongst a
possible range of solutions.
In summary, data from both Studies 1 and 2 suggest that the role of secondary
control in positive psychological adjustment is more than simply acting to
compensate for low levels of primary control. Most importantly, secondary
control predicted significant amounts of variance in positive adjustment for
people with high levels of primary control. Previous theory suggests that the
141
141
nature of this greater functional role for secondary control may be the importance
of managing cognitive/affective reactivity to events in order to allow for the most
adaptive primary control response. Future research designs could test this
proposition more precisely, by investigating the personal/interpersonal outcomes
following people’s primary control responses, amongst sub-samples of people
with both high and low levels of secondary control. Such a research design could
be used to determine whether higher levels of secondary control do, indeed, relate
to better intrapersonal, inter-personal or objective outcomes following a primary
control response.
9.5 The Role of Personality in Explaining the Relationship Between
Primary/Secondary Control and Positive Psychological Adjustment
The exploratory analysis designed for Study 2 proposed that the relationship
between primary/secondary control and positive psychological adjustment might
be explained by the personality dimensions of Extraversion and Neuroticism.
Hence, a series of hierarchical regressions was conducted in which Extraversion
and Neuroticism were entered first into the prediction of each of the positive
adjustment variables (Subjective Quality of Life, Positive Affect and Positive
Thinking). Next, primary/secondary control were entered into the prediction of
positive adjustment, in order to assess primary/secondary control’s capacity to
predict variance in positive adjustment, after variance in positive adjustment that
could be attributed to personality was already accounted for.
It was found that primary/secondary control continued to predict significant
portions of variance in all three positive adjustment variables (Subjective Quality
of Life, Positive Affect and Positive Thinking) beyond the variance accounted for
by the personality factors. The capacity of both primary and secondary control to
continue to predict significant portions of variance in adjustment demonstrates
that significant portions of shared variance between primary/secondary control
and all three measures of positive adjustment cannot be attributed to the effect of
142
142
the personality dispositions of Neuroticism or Extraversion. Therefore, the role
of personality in explaining the relationship between primary/secondary control
and positive psychological adjustment is limited.
While Extraversion and Neuroticism could not entirely account for the adaptive
value of primary/secondary control, results showed the extent to which
personality did explain shared variance between primary control and adjustment,
and secondary control and adjustment. Hence, the smaller reductions in
proportions of variance in adjustment accounted for by secondary compared to
primary control, after the inclusion of personality variables, suggests that
secondary control’s relationship to positive adjustment is less determined by the
personality factors than the relationship between primary control and positive
adjustment. This suggests a greater influence for personality in explaining the
adaptive value of primary control, compared to secondary control.
These results provide initial data to help identify the mechanisms that explain why
primary and secondary control are associated with positive psychological
adjustment. That is, results from the exploratory analysis in Study 2 suggest that
part of the explanation of why control is associated with higher levels of positive
adjustment is due to the influence of the personality factors of Extraversion and
Neuroticism. Furthermore, these personality factors explain more of the primary
control-adjustment link, than they do the secondary control-adjustment link.
Thus, suggesting that whilst approximately 65-75% of the primary control-
adjustment link was explained by personality, leaving approximately 35% of the
variance yet to be explained, much higher portions of variance in the secondary
control-adjustment link remain unexplained.
Hence, future research could be aimed at determining what other factors might
explain additional portions of control-adjustment variance. Other potential
constructs that might explain additional portions of shared variance may include
other personality variables, objective measures of social/economic conditions, or
143
143
any other construct that can be theoretically, or intuitively linked to both
primary/secondary control and to psychological adjustment. Nevertheless, data
from Study 2 suggests that the positive outcomes enjoyed by people with higher
levels of primary/secondary control cannot be fully explained by the influence of
personality characteristics such as Extraversion or Neuroticism.
The results for negative adjustment however, showed that the relationship
between primary/secondary control and Negative Affect was explained by the
effect of the personality dimension of Neuroticism. It was found that that both
primary control and secondary control failed to predict variance in Negative
Affect, after variance attributed to Neuroticism was accounted for. Extraversion
failed to predict Negative Affect at the first step of the regression. Hence, in
psychological terms, the tendency for people with low levels of
primary/secondary control to experience relatively higher levels of Negative
Affect, may be explained by the influence of a personality disposition that
predisposes the person toward frequent negative feelings (Neuroticism).
This result can be contrasted with the results for the positive adjustment variables
(Subjective Quality of Life, Positive Affect and Positive Thinking) which
continued to be significantly related to primary/secondary control, after variance
attributed to Neuroticism and Extraversion was accounted for. That is, results
from Study 2 show different patterns of relationship between primary/secondary
control, personality and positive adjustment, versus the pattern of relations
observed between control, personality and negative adjustment. This different
pattern of relations between variables measuring positive versus negative
adjustment may be a valid indication that the relations between primary/secondary
control and positive adjustment are not reciprocal to measures of negative
psychological functioning. Hence, in order to develop a more comprehensive
understanding of the adaptive value of primary/secondary control, future research
would best select measures that operationalise both positive and negative
expressions of psychological health.
144
144
Further investigation of the pattern of relations amongst personality, control and
both positive and negative measures of adjustment would assist in determining
whether the different pattern of relations observed between positive and negative
measures are robust. Such research could have important clinical implications in
that attempts to clinically train people in primary and secondary control (e.g.
Weisz, Thurber, Sweeney, Proffitt & LeGagnoux, 1997) may lead to differential
benefits between effecting positive outcomes (e.g. improving Subjective Quality
of Life) versus reducing negative outcomes (e.g. reducing level of depression).
9.6 Further Limitations and New Directions for Research
The implications of the present data suggest the adaptive value associated with a
balance in the levels of control processes, as opposed to preferring one control
process over the other. However, a central assumption within Studies 1 and 2 is
that having higher levels of Subjective Quality of Life, Positive Affect and
Positive Thinking is actually more adaptive, or indicative of better psychological
adjustment. An important qualification is to recognise that the highest levels of
these positive cognitive/affective states does not necessarily imply highest levels
of psychological adjustment or psychological health.
For example, the highest levels of the positive feeling states may be seen in
clinical disorders involving manic or hypomanic episodes (see Kaplan & Sadock,
1998 for an extended description of mania/hypomania). Hence, excessively high
levels of positive cognitive/affective states may actually impede the person’s
ability to adapt to their environment.
Hence, another potential direction for future research could be to study the
adaptive value of primary/secondary control and incorporating more objective
measures of social, emotional or occupational functioning and checking them
against more objective measures of functioning (e.g. the Objective Quality of Life
Scale; Cummins 1997b). By incorporating both subjective and more objective
145
145
measures of psychological functioning within the one study, future research could
empirically confirm that the subjective reporting of higher levels of positive
cognitive/affective states actually corresponds with better objective social,
emotional and occupational functioning. Together, the suggested additions to
future research conducted with the primary/secondary control construct may assist
in the development and understanding of psychological processes that may
improve people’s ability to control any subjective condition (e.g. psychological
disorders, happiness or discontent) or objective conditions in their life (e.g.
housing, work environments or aspects of physical health).
In summary, the present data suggest that future attempts to enhance peoples’
preference for primary and secondary control would best achieve healthy
outcomes by encouraging a comprehensive repertoire of control strategies
involving both control mechanisms. That is, the current analysis demonstrated
that developing a preference for one control process over the other is associated
with reduced levels of positive psychological adjustment. Furthermore, the
present data suggest that the function of secondary control may be more than
simply compensating for low primary control. That is, secondary control may be
of critical importance to the effective functioning of primary control.
Furthermore, it was found that the relationship between control and positive
psychological adjustment may be partly explained by the personality dimensions
of Extraversion and Neuroticism.
146
146
CHAPTER 10
CONCLUDING OVERVIEW
10.1 Summary and Conclusion
The aim of this thesis was to test a number of theoretical propositions regarding
the adaptive value of control. Previous theory within the primary/secondary
control framework suggested the primacy of primary control. This theory
proposes that primary control is intrinsically more adaptive than secondary
control, and secondary control is conceived as simply acting as a back-up strategy
for when primary control is low.
However, previous theoretical and empirical evidence reviewed suggests that
rather than primary control being inherently more adaptive than secondary
control, high levels of primary control have been shown to be associated with
various negative mental and physical outcomes. Furthermore, some cross-cultural
evidence suggests that Western cultures show a cultural bias, in which the
benefits associated with primary control are emphasised, whereas Asian cultures
emphasise the adaptive value of secondary control over primary control. In
addition, contrasting with primacy theory, is the theoretical proposal by the
original authors of the two-process model, suggesting the importance of a balance
between the two-processes of control for optimal adjustment (Rothbaum et al.,
1982; Weisz et al., 1984).
Empirical research aimed at testing the functional primacy of primary control has
been limited and the results generally mixed or inconclusive. Furthermore,
previous attempts to operationalise primary and secondary control were vague and
147
147
non-descriptive and therefore may have equated secondary control with helpless
acceptance. In addition, previous studies in primary and secondary control had
148
148
measured only a limited set of constructs pertaining to negative psychological outcomes (i.e. depression and anxiety). Hence, relations between primary/secondary control and more positive and comprehensive indicators of psychological functioning (e.g. Subjective Quality of Life and Positive Affect) remained relatively unexplored.
Therefore, Study 1 involved the development of a new measure of primary and
secondary control, the Primary & Secondary Control Scale (PSCS). The PSCS
operationalises primary and secondary control by specifying precise
cognitive/behavioural mechanisms by which people may seek control over the
external environment (primary control) or control over their internal reactions to
events in order to minimise their impact (secondary control). In particular,
secondary control items were selected with reference to the predictive, vicarious,
illusory and interpretive types of secondary control discussed by the original
proponents of the primary/secondary control construct (Rothbaum et al., & Weisz
et al., 1984). In addition, literature from the coping area was scanned for
additional secondary control mechanisms by which people may minimise the
impact of events.
Factor-analyses of the PSCS conducted in Studies 1 and 2 demonstrated the
factorial validity of the scale, showing a two-factor simple structure in which
primary control items load onto one factor and secondary control items onto the
other factor. Furthermore, reliability analyses showed that the PSCS items
demonstrate adequate item-total correlations and good internal consistency to the
primary control and secondary control sub-scales.
Hypotheses were then developed for Study 1 in order to test primacy theory using
the newly developed Primary and Secondary Control Scale. Hence, it was
hypothesised that people who show a preference for one control process over the
other (“control-imbalanced”) would report lower levels of positive psychological
adjustment compared to people who maintained average or above-average levels
of both primary and secondary control (“control-balanced”). Furthermore, it was
hypothesised that secondary control’s role in adaptive functioning is greater than
134
134
simply acting to compensate for low levels of primary control. Thus, secondary
control was predicted to explain significant amounts of variance in positive
adjustment, for people both high and low in terms of primary control. Study 1
also explored the relations between primary/secondary control and positive
measures of psychological adjustment, as opposed to measures of negative
psychological processes as reported in previous studies.
Results from Study 1 showed that as predicted, people with a control-imbalance
(average or above-average levels on one control process, but below average on
the other) reported lower levels of positive psychological adjustment, compared to
people with average or above levels of both primary and secondary control. It
was also found that secondary control was related to positive psychological
adjustment for both sub-groups of high and low primary control, suggesting that
secondary control has a greater role in adaptive functioning than had been
suggested by previous theory. In addition, results from Study 1 provided
evidence of significant relations between primary/secondary control and positive
measures of psychological adjustment.
A major aim of Study 2 was to conduct an exploratory analysis to determine the
capacity of the personality dimensions of Extraversion and Neuroticism to explain
the relationship between control and positive adjustment. In addition, this study
aimed to re-test the hypotheses from Study 1 in order to assess the robustness of
the observed findings.
Results generally supported those observed in Study 1, such that the control-
imbalanced group (high primary control, combined with low secondary control
[HiPrim-LoSec]) was found to have reduced levels of positive psychological
adjustment. The other control-imbalanced group (LoPrim-HiSec) did not show
reduced levels of positive adjustment as observed in Study 1. However, this
result was explained in terms of methodological problems associated with the
135
135
creation of the cluster-group such that the LoPrim-HiSec group actually had
average levels of primary control.
In regards to the hypothesis regarding secondary control as having a greater
functional role than simply compensating for people with low primary control,
results in Study 2 supported those observed in Study 1. Results from both studies
show that secondary control explained variance in positive adjustment amongst
both high and low primary control sub-groups, confirming secondary control’s
role in positive adjustment as more than just acting to compensate for low primary
control.
The consistency of the findings across both studies does not support the primacy
of primary control. That is, optimally adaptive functioning was reported amongst
people who have a repertoire of control strategies involving relatively higher
levels of both primary and secondary control strategies. This makes evolutionary
sense in that the inherent shortfalls associated with each of the control processes
can be minimised by the operation of the other. This allows the person to adapt to
the variety of situations in life, which have varying levels of objective
controllability associated with them, by selecting the optimally adaptive mix of
approach-avoidance strategies. Hence, situations that are less amenable to
primary control (e.g. Multiple Sclerosis) could be met with a range of secondary
control strategies for minimising the psychological impact of the event. Whereas
situations with greater levels of objective controllability (e.g. a broken arm), may
be met with a range of primary control strategies that change the existing problem
situation by engaging the environment (e.g. going to Emergency).
In contrast, the person who shows a preference for one control process over the
other (as primacy theory suggests), appears to experience a reduced Subjective
Quality of Life, less positive thoughts and to a lesser extent, reduced levels of
positive feelings, compared to people who use both control processes (i.e. control-
balanced).
136
136
Hence, it is suggested that people who emphasise their use of primary control
while tending to neglect secondary control, are at risk of experiencing strong
cognitive/affective states that may interfere with cognitive processes (e.g.
attention, concentration, problem solving and decision making) that are important
psychological resources for the optimal operation of primary control strategies.
For example, a worker with few secondary control strategies who performs poorly
at an important meeting may become distracted by their feelings of self-doubt and
disappointment over the perceived failure, leading them to concentrate less on the
day’s tasks at hand. Thus, secondary control strategies may minimise
psychological impacts of primary control failure, thereby reducing distraction
from subjective feelings of discontent (e.g. worry) or positive feelings (e.g.
romantic love). In short, the use of secondary control strategies allows the person
to maintain the necessary level of cognitive resources (attention, concentration
etc.) that permit full focus on primary control strategies and goals.
This function of secondary control is more to do with ensuring that the process of
primary control striving is optimally supported by good levels of cognitive skills,
thereby ensuring that the person develops and decides on the most advantageous
courses of solution and action considering the inherent limitations of the
environment. This “quality-assurance” function of secondary control is
conceptualised as additional to the back-up function of secondary control
proposed by Heckhausen and Schulz (1995), which is more concerned with re-
motivating the person’s primary control striving after failure or losses in primary
control.
On the other hand, people who emphasise their use of secondary control over
primary control run the risk of missing opportunities to achieve developmental
needs and wants. In the extreme case, very low levels of primary control could
result in failure to meet basic requirements for survival (e.g. food, shelter,
income).
137
137
Study 2 also explored the capacity of personality factors of Extraversion and
Neuroticism to explain the control-adjustment relationship. Results showed that
both primary and secondary control continued to predict significant variance in
positive adjustment after variance attributed to Extraversion and Neuroticism had
been removed. This shows that the shared variance between primary/secondary
control and positive adjustment was only partially explained by the personality
dimensions of Extraversion and Neuroticism. Further analyses also demonstrated
that Extraversion and Neuroticism explained the primary control-adjustment link
to a greater extent than the secondary control-adjustment link. It was found that
this result was consistent across all three of the positive adjustment variables,
Subjective Quality of Life, Positive Affect and Positive Thinking.
10.2 Clinical Implications
The clinical implications of the present results are wide reaching across all models
of counselling and psychotherapy. That is, the findings concerning the role of a
balance in levels of both primary control and secondary control, suggest that,
regardless of the model of change that a clinician works with, optimal levels of
adjustment may best be achieved with a focus on the dual goals of improving the
person’s ability to manage their emotions (e.g. whether via insight or newly learnt
strategies for change), as well as encouraging pro-active attempts to achieve the
person’s desired needs and wants by engaging the environment. This is not to say
that every counselling relationship, or every problem would profit from a mix of
primary and secondary control interventions, because some problems necessitate a
greater level of one control process over the other due to differences in the
objective controllability associated with different problems and goals. However,
the results do suggest that clinicians and clients could benefit from a general
analysis of the person’s control repertoire over the course of the counselling
relationship and across all the assessed problem areas, in order to provide both
client and therapist with a road map to possible interventions.
138
138
In summary, the primary/secondary control dichotomy represents a super-ordinate
construct for human change by providing a simple framework to understand the
two main ways by which people may achieve goals and solve problems. The
dichotomy is super-ordinate because all models of counselling and psychotherapy
can be conceptualised as attempting to improve the cognitive, affective and
behavioural functioning of the person by manipulating either the internal
functioning of the person’s beliefs and attitudes (e.g. psychoanalysis: secondary
control focus), the way the person engages their environment (e.g. solution-
focused therapy: primary control focus) or a combination of both (e.g. cognitive-
behavioural therapy: a combination of environmental and internal manipulation).
Hence, regardless of the client and clinician’s chosen framework for change, the
findings on the importance of the primary/secondary control-balance helps both
parties navigate through all the possible courses of intervention, suggesting to
clinicians to be cognisant of the importance of both types of control in people's
lives. Without paying attention to both control processes as potential sources for
intervention the present results suggest that clients may not achieve the optimal
benefit from their counselling because they remain vulnerable to the shortfalls
associated with the underdeveloped control process.
Another important clinical implication of the present results concerns clinicians’
interpretations of commonly experienced client behaviours in counselling
relationships. As Rothbaum et al. (1982) argue, learned helplessness research
(Seligman, 1975) had falsely interpreted behaviours involving a passive attitude
toward the environment, or withdrawing from attempts to control environmental
stressors, as indicative of relinquished control. In their view, certain behaviours
that reflect a passive attitude toward the environment and characterised at times
by psychological or behavioural withdrawal do not reflect a reduced desire or
attempt for control (giving-up), but a secondary attempt to minimise the
psychological impact of potentially debilitating and threatening situations. In
particular, the researchers argue that a lot of secondary control behaviour is aimed
139
139
at minimising a particularly debilitating aspect of failure – disappointment.
Disappointment is particularly debilitating, in their view, because it represents a
loss of both primary control (to get what one wanted) as well as a loss of
secondary control (ability to control the emotional impact of the failure). Hence,
secondary control behaviours function to reduce chances of disappointment by
showing the person that there is some meaning, or benefit to be achieved from
failure experiences, thus preventing the person from disappointment.
The results from Studies 1 and 2 show that secondary control behaviours
characterised by passivity toward the environment or withdrawal (e.g. “I know
something bad will happen”, an example of predictive secondary control) are
linked to positive psychological outcomes such as a higher Subjective Quality of
Life, Positive Affect and Positive Thinking. Hence, evidence suggests that
behaviours such as the predictive, illusory, interpretive and vicarious types of
secondary control described by Rothbaum et al. (1982) and Weisz et al. (1984)
are indeed positive psychological processes concerned with attempting to control
the psychological impact of events rather than being indicative of relinquished
control.
For clinicians in practice, behaviours that reflect an inherent passivity toward the
environment are commonly interpreted as further signs of clients’ depressed
motivation and thus a symptom to be discouraged either by direct verbal
challenge or via behavioural experiment. However, considering the potential
benefits associated with such secondary control behaviours, the present results
suggest that clinicians, at the very least, should interpret apparent signs of the
client “giving-up”, or choosing less pro-active mechanisms for control as possible
attempts by the person to control the psychological impact of the failure. Less
pro-active secondary control behaviours such as those described by
Rothbaum et al. (1982) may well be unacceptable to those clinicians focused on
achieving direct, measurable changes in client behaviour, however, these same
140
140
behaviours may represent self-protective attempts by clients to manage
psychological impacts such as disappointment. These less pro-active attempts for
control make sense for many clients considering the reduced emotional, financial
and educational resources that many clients’ perceive after facing chronic
stressors.
In conclusion, the results from this thesis suggest a much greater role for
secondary control beliefs and behaviour in adaptive psychological functioning.
This suggests to research psychologists and clinicians alike, the importance of
controlling both the internal self as well as controlling the external environment
for optimal positive psychological adjustment.
141
141
REFERENCES
Affleck, G., Tennen, H., Pfeiffer, C. & Fifield, J. (1987). Appraisals of control
and predictability in adapting to a chronic disease. Journal of Personality and
Social Psychology, 53, 273-279.
Alloy, L.B. & Athrens, A. H. (1987). Depression and pessimism for the future:
Biased use of statistically relevant information in predictions for self versus
others. Journal of Personality and Social Psychology, 52, 366-378.
Anderson, B., Kiecolt-Glaser, J. & Glaser, R. (1994). A biobehavioural model of
cancer stress and disease course. American Psychologist, 49, 389-404.
Argyle, M. (1987). The Psychology of Happiness. London: Methuen.
Argyle, M. & Lu, L. (1990). The happiness of extraverts. Personality and
Individual Differences, 11, 1011-1017.
Argyle, M., Martin, M. & Crossland, J. (1989). Happiness as a function of
personality and social encounters. In J.P. Forgas & J.M. Innes (Eds.), Recent
Advances in Social Psychology: An International Perspective (pp. 189-203).
North Holland: Elsevier Science Publishers.
142
142
Front Pages
Averill, J. (1973). Personal control over aversive stimuli and its relationship to stress.
Psychological Bulletin, 80, 286-303.
Azuma, H. (1984). Secondary control as a heterogeneous category. American
Psychologist, 39, 970-971.
Baron, R. & Rodin, J. (1978). Perceived control and crowding stress. In A. Baum,
Singer, J. & Valine, S. (Eds.), Advances in Environmental Psychology (pp.145-190).
Hillsdale, NJ: Lawrence Erlbaum Associates.
Beck, A.T., Ward, C.H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory
for measuring depression. Archives of General Psychiatry, 4, 561-571.
Bouchard, G., Sabourin, S., Lussier, Y., Wright, J. Richer, C. (1997). Testing theoretical
models underlying the Ways of Coping Questionnaire with couples. Journal of
Marriage and the Family, 59, 409-418.
Bradburn, N. (1969). The Structure of Psychological Well-Being. Chicago: Aldine.
Burger, J.M. & Cooper, H.M. (1979). The desirability of control. Motivation and
Emotion, 3, 381-393.
143
143
Front Pages
Burger, J. & Arkin, R. (1980). Prediction, control, and learned helplessness. Journal of
Personality and Social Psychology, 38, 482-491.
Burke, M., Brief, A., George, J., Roberson, L., & Webster, J. (1989). Measuring affect
at work: Confirmatory analyses of competing mood structures with conceptual
linkage to cortical regulatory systems. Journal of Personality and Social Psychology,
57, 1091-1102.
Burns, D.D. (1980). Feeling Good: The New Mood Therapy. Melbourne: Information
Australia Group.
Carlson, N.R. (1987). Psychology: The Science of Behaviour. Sydney: Allyn &
Bacon, Inc.
Carver, C., Scheier, M. & Weintraub, J. (1989). Assessing coping strategies: A
theoretically based approach. Journal of Personality and Social Psychology, 56, 267-
283.
Carver, C., Pozo, C., Harris, S., Noriega, V., Scheier, M., Robinson, D., Ketcham, A.,
Moffat, F. & Clark, K. (1993). How coping mediates the effect of optimism on
distress: A study of women with early stage breast cancer. Journal of Personality and
Social Psychology, 65, 375-390.
144
144
Front Pages
Cattell, R.B. (1946). The Description and Measurement of Personality. Yonkers, NY:
World Book.
Collins, D., Baum, A. & Singer, J. (1983). Coping with chronic stress at Three Mile
Island: Psychological and biochemical evidence. Health Psychology, 2, 149-166.
Corah, N.L. & Boffa, J. (1970). Perceived control, self-observation, and response to
aversive stimulation. Journal of Personality and Social Psychology, 16, 1-4.
Costa, P. & McCrae, R. (1980). Influence of extraversion and neuroticism on subjective
well-being: Happy and unhappy people. Journal of Personality and Social
Psychology, 38, 668-678.
Costa, P. & McCrae, R. (1984). Personality as a lifelong determinant of well-being. In
C. Malatesta & C. Izard (Eds.), Affective Processes in Adult Development and
Ageing (pp. 141-157). Beverly Hills, CA: Sage.
Costa, P. & McCrae, R. (1992). Revised NEO-Personality Inventory
(NEO PI-R) and the NEO Five Factor Inventory: Professional Manual.
B. & Sinar, E. (1998). Primary and secondary control over age-related changes in
physical appearance. Journal of Personality, 66, 583-605.
157
157
Front Pages
Vitaliano, P.P., Russo, J., Carr, J., Maiuro, R. & Becker, J. (1985). The Ways of Coping
Checklist: revision and psychometric properties. Multivariate Behavioural Research,
20, 3-26.
Vollrath, M., Torgersen, S. & Alnaes, R. (1995). Personality as a long-term predictor of
coping. Personality and Individual Differences, 18, 117-125.
Wallston, K. (1989). Assessment of control in health care settings. In Steptoe, A. &
Appels, A. (Eds.), Stress, Personal Control and Health (pp. 85-105). Brisbane: John
Wiley and Sons.
Ware. J.E. & Sherbourne, C.D. (1992). The MOS 36-item short-form health survey
(SF-36): Conceptual framework and item selection. Medical Care, 30, 473-481.
Warr, P., Barter, J. & Brownbridge, G. (1983). On the independence of positive and
negative affect. Journal of Personality and Social Psychology, 44, 644-651.
Watson, D., Clark, L.A. & Tellegen, A. (1988). Development and validation of brief
measures of positive and negative affect: The PANAS scales. Journal of Personality
and Social Psychology, 54, 1063-1070.
158
158
Front Pages
Weinstein, N. (1980). Unrealistic optimism about future life events. Journal of
Personality and Social Psychology, 39, 806-820.
Weinstein, N. (1984). Why it won't happen to me: Perceptions of risk factors and
susceptibility. Health Psychology, 3, 431-457.
Weinstein, N. (1989). Optimistic biases about personal risks. Science, 246, 1232-1233.
Weisz, J. (1990). Development of control related beliefs, goals and styles in childhood
and adolescence: A clinical perspective. In Rodin, J., Schooler, C. & Schaie, K.
(Eds.), Self-directedness: Cause and Effects Throughout the Life Course (pp. 103-
145). Hillsdale NJ: Lawrence Erlbaum Associates.
Weisz, J., Rothbaum, F. & Blackburn, T. (1984). Standing out and standing in: The
psychology of control in America and Japan. American Psychologist, 39, 955-969.
Weisz, J., Thurber, A., Sweeney, L., Proffitt, V. & LeGagnoux, G. (1997). Brief
treatment of mild-to-moderate child depression using primary and secondary control
enhancement training. Journal of Consulting and Clinical Psychology, 65, 703-707.
Yik, S., Russell, J.A. & Barret, L.F. (1999). Structure of self-reported affect: Integration and beyond. Journal of Personality and Social Psychology, 77, 600-619.
159
159
Front Pages
Appendix A1
Information Sheet and Covering Letter Sent to participants for Study 1