Partner's generalized locus of control and domains of job satisfaction: evidence from Australia

Abstract

Purpose

This study examines the relationship between partners' locus of control and their spouses' domains of job satisfaction (job satisfaction and its domains, personal income and promotion) among Australian couples.

Design/methodology/approach

Data were obtained from the Household, Income and Labor Dynamics of Australia (HILDA) Survey. Various estimation strategies including ordinary least squares (OLS), Mundlak approach and instrumental variable (IV) method are used to reveal the relationship between spouse's locus of control and domains of job satisfaction.

Findings

To reduce sex heterogeneity, the analysis used in this study is disaggregated by sex. In particular, the findings of this study show that wives' locus of control positively influences husbands' satisfaction with pay and working hours, while there is no relationship between husbands' locus of control and wives' domains of job satisfaction.

Social implications

The study's findings emphasize the importance of locus of control in couples. A good work–life balance and a healthy marital relationship potentially facilitate positive effects of characteristics from the partner on employees' job satisfaction. Thus, on the organizational level, employers may consider creating a working environment that promotes a healthy marital relationship for their staff, including flexible working schedules, work from home options, family days or family-extended staff events.

Originality/value

This study is the first to reveal the relationship between spousal locus of control and domains of job satisfaction, enriching the current literature on this topic.

Citation

Tran, D.B. and Tran, H.T.M. (2024), "Partner's generalized locus of control and domains of job satisfaction: evidence from Australia", Journal of Asian Business and Economic Studies, Vol. 31 No. 1, pp. 40-54. https://doi.org/10.1108/JABES-06-2022-0152

1. Introduction

Locus of control is a significant predictor of social and economic success (Almlund et al., 2011). They also appear to have just as strong an effect as cognitive ability (Heckman et al., 2006). Locus of control is defined as “whether or not the person perceives a causal relationship between their own behavior and the reward” (Rotter, 1966) and is polarized as internal and external locus of control. People with an external locus of control often believe that what happens to them is the result of luck, destiny or fate. In contrast, people with an internal locus of control believe that they are responsible for their own successes and failures. The positive association between locus of control and work outcomes has been empirically found in previous literature. For instance, Martin et al. (2005) studied a sample of 404 employees of a large financial services organization and administrative staff members of a head office of a utility company based in the UK Their findings show that people with an internal locus of control develop better quality relations with their manager and this, in turn, results in more favorable work-related factors including job satisfaction, work-related well-being and organizational commitment. Similarly, Chen and Silverthorne (2008) examined the relationship between locus of control and job stress, job satisfaction and job performance among Taiwanese accounting professionals. They concluded that individuals with a higher internal locus of control are more likely to have lower levels of job stress and higher levels of job performance and satisfaction. Padmanabhan (2021) highlighted that those employees with an internal locus of control are more likely to have higher job satisfaction, based on a sample of 65 employees of private sector organizations.

In a broader context, Spector et al. (2002) collected data on work locus of control, job satisfaction, and other work-related characteristics from 5,185 managers from 24 geopolitical entities. The results show that the relationship between work locus of control, job satisfaction and psychological well-being is universal across the countries in the sample. By employing a meta-analysis approach to examine the locus of control at work, Ng et al. (2006) examined the relevant studies of the relationship between locus of control and a wide range of work outcomes, and they found that an internal locus is positively associated with favorable work outcomes, such as positive task and social experiences and greater job motivation. However, beyond the positive impact of a particular personality on one's career, we hypothesize that there is a spillover effect of personality in a marital relationship, which implies that the personality of a partner can possibly influence their spouse's domains of job satisfaction.

Previous studies have revealed that home life can affect work experiences, based on theories of compensation and spillover (Lambert, 1990). The theory of compensation implies that workers can find more satisfaction at work or home as compensation for a lack of satisfaction in the other. The more popular view is that there are spillover effects from home to work (e.g. Demerouti et al., 2010; Jahan Priyanka et al., 2022) and vice versa. When the spillover effect originates from one's partner, this is known as the crossover effect. Relationship crossover (or transmission) effects occur when an individual's personal experiences such as stress and strain (Montgomery et al., 2008), momentary moods (Song et al., 2008), burnout (Westman and Etzion, 1995) and work-to-relationship conflict (Matthews et al., 2006) influence the other partner in the relationship. As an outcome, it is possible that a person can alter their spouse's job satisfaction (Westman and Etzion, 1995), and add to the feelings of work overload, poor role congruence and poor organizational structure (Karabay et al., 2016).

When we look closely at the mechanism of how home life can influence work life, previous studies have proved that the personality of the one's partner can be one factor (Winter et al., 1977; Helson and Roberts, 1992; Solomon and Jackson, 2014; Zheng et al., 2022). More specifically, Westman and Etzion (1995) showed that a spouse's sense of control contributes to the prediction of the other spouse's burnout, highlighting the ability of individuals to empower their spouses with their own sense of control. This relationship can be explained by at least two possible channels. First, personality has been found to have an impact on marital satisfaction (e.g. Dyrenforth et al., 2010) which can potentially produce happiness at home and a positive mood which, in turn, can spill over into the workplace. Empirically, locus of control is found to associate with marital satisfaction (Bugaighis et al., 1983; Lee and McKinnish, 2019). Second, spouses with certain characteristics can provide more support at home, such as domestic tasks, which might save energy and effort for their spouse and allow them to focus more on their work (e.g. Jackson et al., 2010; Solomon and Jackson, 2014).

The current study focuses on one aspect of a partner's personality, namely, locus of control and its relationship with domains of job satisfaction of the other, aiming to reveal the importance of a spouse's locus of control on one's professional accomplishments beyond their own personality traits. Since only a few previous studies have explicitly examined the connection between locus of control and spouses' domains of job satisfaction, we pose the following question for consideration: To what extent is locus of control of partners associated with domains of job satisfaction? The answer to this question significantly contributes to the understanding of the relationship between locus of control and domains of job satisfaction in the context of a marital relationship.

To shed light on this issue, we utilize a rich data set of over 7,000 households throughout Australia over a 15-year period. The findings show that partners' locus of control is significantly associated with domains of job satisfaction, highlighting the value of a positive personality in a marriage or other co-habiting arrangement. In other words, the results provide evidence of the benefits of locus of control on people's lives with respect to marriage. This study, therefore, is the first to reveal the relationship between locus of control and spousal domains of job satisfaction, enriching the current literature on this topic. Moreover, the large panel data used in the study not only addresses econometric problems, such as endogeneity, but also enhances the generalizability of the findings.

The study is structured as follows. The next section summarizes the literature review. The methodology and estimation results are presented in Sections 3 and 4 and Section 5 concludes our study.

2. Literature review

The literature shows that locus of control has a significant impact on economic outcomes such as income, unemployment, health outcomes or the ability to adapt to negative life events (Buddelmeyer and Powdthavee, 2016; Caliendo et al., 2014; Heineck and Anger, 2010; Kesavayuth et al., 2020). Other researchers have observed that individuals' locus of control is a significant factor in the workplace due to its relationship with various career outcomes such as job satisfaction and performance (Aubé et al., 2007; Judge and Bono, 2001; Yousef, 2000). Additionally, previous research has linked having an internal locus of control to higher earnings (e.g. Heineck and Anger, 2010; Piatek and Pinger, 2016; Semykina and Linz, 2007). Furthermore, locus of control is related to occupational attainment (Cobb-Clark and Tan, 2011) and the propensity to become self-employed (Hansemark, 2003; Caliendo et al., 2014).

Cobb-Clark (2015) proposed three potential mechanisms – human capital investments, hiring decisions and optimal incentive contracts – through which locus of control might influence labor market outcomes. First, locus of control is potentially associated with the human capital investments such as academic performance (Heckman and Kautz, 2012; Mendolia and Walker, 2014), health and health-related behaviors (Chiteji, 2010; Cobb-Clark et al., 2014; Kesavayuth et al., 2020). Second, more and more firms use personality tests to gain insights into potential employees (Tran, 2021) in the belief that personality measures may be an important predictor of job performance (e.g. Judge and Bono, 2001). However, the use of personality testing in hiring decisions raises many concerns and is the subject of ongoing debates. Cobb-Clark (2015) suggested that optimal compensation rules might depend on the degree to which workers have internal versus external perceptions of control.

There are only a few studies that examine the association between partner personality and work outcomes. Winter et al. (1977) studied the relationship between husbands' power motivation in 1960, and their wives' careers in 1974, using a sample of 51 male college graduates. The study finds that the career status of a wife is negatively correlated with her husband's power motivation. Helson and Roberts (1992) examined the relationship between the personality of young husbands and wives, and the wives' work history at age 43, in an attempt to reveal the significance of personality characteristics to the women's subsequent careers over the next 15 years among 63 couples from the Mills Longitudinal Study in the US The findings suggest that wives who are successful at work tend to have husbands with higher levels of individuality. Solomon and Jackson (2014) relied on a longitudinal sample of approximately 4,500 married individuals in Australia and found that partners' conscientiousness predicts various work outcomes, such as job satisfaction, income and the likelihood of getting promoted. A more recent study by Zheng et al. (2022) shows that spouse's mindfulness at home on any given day has a positive effect on work engagement the following morning, using data from employees of a large commercial bank in China.

The relationship between one's spouse and work outcomes can potentially be explained by marital satisfaction. In other words, a healthy relationship can have spillover effects in the workplace. Personality has been found to have an impact on marital satisfaction (Bugaighis et al., 1983; Lee and McKinnish, 2019). Moreover, influences from a spouse's personality can be extended beyond the relationship and associated with an individual's overall life satisfaction (Russell and Wells, 1994) and health (Roberts et al., 2009). For example, a happier marriage produces more positive energy and healthier moods, which links to a better work life (Heller and Watson, 2005; Judge and Ilies, 2004). Conscientious partners with higher levels of self-control cause fewer disagreements at home (Donnellan et al., 2004), promoting a more satisfying home life. A supportive romantic partner may be a positive determining factor for women's success (Barth et al., 2016). More recently, the study of Sampath and Baral (2017) showed that work–family experiences of supervisors may influence the perception of a family-supportive work environment and the perceived control over work schedules, which may, in turn, influence the work–family experiences of the subordinates and their subsequent performance.

Another possible factor is the degree to which the home environment is supportive. A conscientious partner is more likely to share the housework, finance decisions, appointments or plan ahead (Jackson et al., 2010), suggesting that a supportive domestic environment is important. For example, if the husband is more involved in household responsibilities, the wife will have more time to focus on her profession. This outsourcing effect can reduce women's personal costs, creating an opportunity for them to advance on their career path. The conceptual framework is presented in Figure 1.

To sum up, there are an increasing number of studies that highlight the role of locus of control, in general, and specifically at the workplace. However, to our understanding, there are only limited empirical studies to reveal the potential role of spouse's locus of control on work outcomes. Thus, this study aims to fill this research gap. Our hypotheses are constructed as follows.

H1.

Wives' locus of control has a positive impact on husbands' domains of job satisfaction.

H2.

Husbands' locus of control has a positive impact on wives' domains of job satisfaction.

3. Methodology

3.1 Data

The data used in this research are obtained from the Household, Income and Labor Dynamics in Australia (HILDA) Survey spanning the period from 2001 to 2015. The data are collected annually over the 15 years, representing 15 waves of data. The data set is designed as panel data that collects information about various aspects of Australian households. We choose this data set for our study because it provides the key variables we require. The data are individually collected from every member of the household aged 15 years and over, which includes their gender, own locus of control, domains of job satisfaction and other related demographic information. We are also able to collect information on the locus of control reported by the partner in the household. Moreover, HILDA is a nationally representative data set, representing the characteristics of the majority of the Australian population. Finally, the quality of the data set and the re-interview rate is very high, so this data set is ideal for our study.

3.1.1 Locus of control

HILDA uses the psychological coping resources component of the mastery module, developed by Pearlin and Schooler (1978). This is intended to capture self-efficacy, rather than locus of control as developed by Rotter (1966). Both scales are closely related and are thus mostly assumed to be interchangeable (Cobb-Clark et al., 2014).

Seven questions concerning the locus of control variable are asked in waves 3, 4, 7, 11 and 15, in which participants were asked to evaluate seven statements.

1. “I have little control over the things that happen to me”

2. “There is really no way I can solve some of problems I have”

3. “There is little I can do to change many of the important things in my life”

4. “I often feel helpless in dealing with the problems of life”

5. “Sometimes I feel that I am being pushed around in life”

6. “What happens to me in the future mostly depends on me”; and

7. “I can do just about anything I really set my mind to do”.

The responses are marked on a seven-point Likert scale, with answers from “strongly disagree” to “strongly agree”. To check on the internal consistency of these questions, we deploy Cronbach's alpha. First, we reverse the scores of items 1 to 5 so that answers to all the questions on locus of control are increasing in internal control tendencies. Second, a Cronbach's alpha of 0.83 is calculated, implying a high level of internal consistency among the seven items.

To construct our locus of control measure, we follow Piatek and Pinger (2016), who recommend using a predicted locus of control score based on factor analysis. Accordingly, as stated above, we reverse the scores of items 1 to 5 so that our locus of control index is increasing in internal control tendencies. We then generate the predicted factor obtained from principal factor analysis. The economic literature on locus of control has largely focused on using a predicted locus of control factor based on a factor analysis, rather than a simple sum of items, as this prevents the risk of measurement error and attenuation bias as it allows the data to determine how each item is weighted in the overall index (Preuss and Hennecke, 2018; Xue et al., 2020).

3.1.2 Domains of job satisfaction

A study by Heslin (2005) aimed to reveal the nature of career success, both objective and subjective success. Objective career success can be measured by factors such as pay levels and the rate of promotion, while subjective career success, which has been studied over the last couple of decades, focuses on people's reactions to their own careers, most often indicated by job satisfaction. However, several factors tend to be ignored, such as work engagement, work dedication, absorption at work, affective commitment, job title, authority and autonomy. In this paper, based on the availability of the data and the reasonable scope of the study, we follow the suggestions of Heslin (2005) and Solomon and Jackson (2014) in choosing job satisfaction and its domain as intrinsic occupational outcomes and personal income and job promotion as extrinsic occupational outcomes.

The indicators of intrinsic occupational outcomes include overall job satisfaction, which was determined using the question “All things considered, how satisfied are you with your job?’. In addition, we use responses regarding five aspects of job satisfaction. They are as follows:

1. Total pay satisfaction.

2. Job security satisfaction.

3. Work satisfaction.

4. Working hours satisfaction and

5. Flexibility to balance work and nonwork commitment satisfaction.

The respondents score each one between 0 and 10 to indicate how dissatisfied (0) or satisfied (10) they feel. Job satisfaction is a key concept in industrial and organizational psychology, and it has been linked to a variety of desirable outcomes including job performance, organizational citizenship behavior, absenteeism and life satisfaction (Heller et al., 2002; Erdogan et al., 2012). The majority of definitions of job satisfaction center on how employees feel and think about their jobs (Weiss, 2002). Job satisfaction and its domains are chosen because it is easy to use and captures many aspects of a person's work situation (Ribar and Wooden, 2020). Moreover, it has been previously used in research using the HILDA data set (e.g. Tran, 2021; Wooden and Warren, 2004).

We also consider extrinsic occupational outcomes in this paper, represented by income and job promotion. In particular, we use the logarithm of real annual personal disposable income by first obtaining the nominal annual personal disposable income from the data set, and then deflate it using the averaged quarterly CPI index with a reference base of September, 2012. In terms of promotion, one of the answers to the question, “Did any of these happen to you in the past 12 months?” is “Promoted at work”; respondents who select this option are marked as having been promoted.

3.1.3 Estimation sample

Since the key variable, locus of control, is available in five waves, namely 3, 4, 7, 11 and 15, we extract data from those waves and focus on respondents aged 15–65. For the control demographic variables, we control for age, age squared, education, total resident children, residential states and territories and waves. The final sample consists of 12,940 observations of males and 10,705 observations of females. Table 1 presents descriptive statistics by sex.

3.2 Methodology

The relationship between spouse's locus of control and domains of job satisfaction is laid out as follows.

(1)JSDit=α1SpouseLCit+β1LCit+γ1Xit+ui+εit

where JSDit denotes the domains of job satisfaction (intrinsic outcomes include job satisfaction and its domains and personal income and job promotion representing extrinsic outcomes) of individual i at time t; SpouseLCit is spouse's locus of control while LCit is own locus of control; Xit represents the control variables; ui is the individual-specific effects; and εit is the idiosyncratic error. The parameter α1 captures the effect of spouse's locus of control on job-related outcomes. The data are analyzed using the Stata 14 software.

To reduce sex heterogeneity, the analysis is distinguished by sex [1]. Spouse's locus of control and own locus of control are the key independent variables. Domains of job satisfaction variables, namely, job satisfaction and its domains, income and job promotion are used as dependent variables. We standardize these dependent variables to mean 0 and standard deviation 1. Age, age squared, total resident children, education, residential states and territories and waves are controlled in the equation.

4. Estimation results

4.1 Estimation with the Mundlak approach

To obtain the correct coefficients in equation (1), the general assumption is that the ui should be uncorrelated with Xit in the model. However, there are unobserved individual characteristics that might correlate with both locus of control and domains of job satisfaction. Thus, we apply the Mundlak approach (Mundlak, 1978) to estimate equation (1). This approach applies random effects regression models with group-means of independent variables which vary within groups. This method produces results that are identical to the fixed effects estimator (Yang, 2022). It not only relaxes the assumption in the random effects estimator that the observed variables are uncorrelated with the unobserved variables, but it also allows adding time-invariant variables into the model. Due to its advantages, this approach has been previously used in studies regarding locus of control (Cobb-Clark et al., 2014; Kesavayuth et al., 2020).

Table 2 presents the relationship between wives' locus of control and the domains of job satisfaction of husbands, while Table 3 provides the results of the relationship between husbands' locus of control and wives' domains of job satisfaction. Regarding spouse's locus of control, it is notable that wives' locus of control positively associates with husbands' higher pay satisfaction and working hours satisfaction at p-value <0.01, while the remaining coefficients are insignificant. On the other hand, husbands' locus of control is uncorrelated with any measures of the wives' domains of job satisfaction.

The estimation results using the ordinary least squares (OLS) and instrumental variable (IV) method are presented in the Appendix.

4.2 Discussion

We find that a more internal locus of control of partners is strongly associated with their spouses' domains of job satisfaction. In particular, the results indicate that a wife's locus of control is associated with a higher level of satisfaction in terms of pay and working hours for the husband, while there is no relationship between husbands' locus of control and wives' domains of job satisfaction. There are potential explanations for the findings. Firstly, partners with an internal locus of control tend to create ideal domestic conditions, allowing their spouses to focus on their work. Regarding our results, if wives are supportive of domestic tasks, their husbands can have more time for work, leading to a reduction of stress levels and an increase in the level of satisfaction with working hours. Moreover, men have more resources to advance their careers if they are well supported at home. Secondly, personality traits, such as locus of control, can have a positive impact on marital satisfaction, creating more positive feelings and moods, which can have a spillover effect at the workplace. Explicitly testing these hypotheses in future research is likely to reveal the potential mechanisms of how spouse's locus of control impacts domains of job satisfaction, which will be an important contribution to the literature.

In addition, the results show that personal locus of control positively correlates with job satisfaction and its domain for both men and women, after being corrected for endogeneity problem, which is consistent with previous studies (Chen and Silverthorne, 2008; Martin et al., 2005; Spector et al., 2002). Possible explanations for this relationship can be found in the psychology literature. Workers with an internal locus of control seek out more complex jobs (Judge et al., 2000), perform better at work (Judge and Bono, 2001) and effectively manage stress (Chen and Silverthorne, 2008). They also tend to set more challenging goals, persevere in the face of adversity, experience less job stress and achieve greater overall success (Ng et al., 2006; Wang et al., 2010).

5. Conclusion

This study reveals the relationship between partners' locus of control and spouses’ domains of job satisfaction among Australian couples. The results indicate that wives' locus of control has an impact on husbands' satisfaction level with regard to pay and hours, highlighting the potential benefits of locus of control in the marital relationship.

This study has certain limitations. Firstly, the results are obtained among Australian couples, which should be extended to other countries to examine the generalizability of the findings. Secondly, as mentioned above, the instruments in this research might not strictly meet the requirements of the IV approach, especially the exclusion restriction. Thirdly, future research can pay more attention to the mechanisms of how spousal locus of control affects partners' domains of job satisfaction, including the outsourcing effect, emulation effect and spillover effect. Fourthly, gender-related research is increasingly being criticized for its binarism (men and women). Furthermore, research addressing partners requires the inclusion of nontraditional family structures (other than husband and wife). However, such other genders were not considered in this study due to a lack of sufficient sample size from the survey. As a result, there is a need to broaden our topic to other genders and partnership types in order to contribute to such advancements in the field of gender research studies. Lastly, although locus of control is being challenged for its binary perspective (external and internal), the majority of research continues to consider locus of control as a bi-polar construct. However, for other cases such as a balanced locus of control (neither external nor internal) or bi-local locus of control, more complete constructs must be considered.

Our findings highlight the value of locus of control to couples, supporting and expanding the theoretical framework on this subject, especially in the context of marriage. At the organizational level, many employers share the objective of ensuring the well-being of employees and managers should consider creating a working environment that promotes healthy marital relationships for their staff. For example, they might allow for a more flexible working schedule, encourage the practice of working from home and organize family days or family-extended staff event.

On the other hand, Almlund et al. (2011) proposed three possible policy implications for creating a more internal locus of control. First, parents play a critical part in encouraging noncognitive skills in their children; the more able and engaged the parents, the more effectively they can support their children in obtaining these skills. Second, educational attainment influences locus of control. Finally, for those children born in unfavorable family conditions, early interventions, such as child-care centers or frequent home visits by teachers, could alleviate some of the initial disadvantages for those children which are attributable to influencing locus of control. Thus, education, parental involvement and policy interventions all have an impact on the locus of control, demonstrating the importance of early-childhood support in increasing the productivity of future investments in human capital. In other words, interventions that enhance internal locus of control are a potential way to increase the chances of positive domains of job satisfaction. Although these three implications are not empirically examined in this paper, further research can look into these ways to affect children's locus of control in a way that will be beneficial to them and their future partners.

Figures

Figure 1

Conceptual framework of partners' locus of control and occupational outcomes

Table 1

Descriptive statistics

	Men				Women
	Mean	S.D.	Min	Max	Mean	S.D.	Min	Max
Locus of control
Little control	2.56	1.48	1	7	2.54	1.46	1	7
No ways to solve problems	2.33	1.40	1	7	2.38	1.47	1	7
Cannot change important things in life	2.36	1.37	1	7	2.37	1.40	1	7
Feel helpless	2.25	1.36	1	7	2.34	1.43	1	7
Pushed around	2.45	1.48	1	7	2.49	1.52	1	7
Future depends on me	5.59	1.49	1	7	5.56	1.49	1	7
Can do just about anything	5.45	1.37	1	7	5.44	1.38	1	7
Spouse's locus of control
Little control	2.58	1.50	1	7	2.57	1.49	1	7
No ways to solve problems	2.42	1.50	1	7	2.34	1.42	1	7
Cannot change important things in life	2.42	1.44	1	7	2.37	1.38	1	7
Feel helpless	2.39	1.48	1	7	2.26	1.37	1	7
Pushed around	2.51	1.55	1	7	2.43	1.48	1	7
Future depends on me	5.52	1.52	1	7	5.59	1.50	1	7
Can do just about anything	5.41	1.42	1	7	5.43	1.39	1	7
Negative life events	0.52	0.78	0	11	0.54	0.78	0	7
Spouse's negative life events	0.54	0.80	0	11	0.54	0.79	0	11
Age	42.58	11.24	16	65	41.28	11.19	16	65
Total resident children	1.17	1.19	0	11	1.13	1.17	0	7
Intrinsic job outcomes
Job satisfaction	7.59	1.65	0	10	7.79	1.64	0	10
Pay satisfaction	6.98	2.06	0	10	7.07	2.11	0	10
Job security satisfaction	7.85	2.12	0	10	8.06	2.07	0	10
Work satisfaction	7.69	1.71	0	10	7.72	1.80	0	10
Hours satisfaction	7.10	2.04	0	10	7.39	2.08	0	10
Balance satisfaction	7.36	2.27	0	10	7.64	2.22	0	10
Extrinsic job outcomes
Real personal income	64,789.58	55,886.88	34	906,040	42,614.46	39,690.70	57	906,040
Job promotion	0.10	0.30	0	1	0.09	0.28	0	1
Educational attainment
Postgraduate degree	0.06	0.24	0	1	0.05	0.23	0	1
Graduate diploma/Certificate	0.06	0.24	0	1	0.10	0.29	0	1
Bachelors or honors	0.16	0.37	0	1	0.21	0.41	0	1
Advanced diploma, diploma	0.10	0.30	0	1	0.10	0.31	0	1
Certificate III or IV	0.31	0.46	0	1	0.16	0.37	0	1
Year 12	0.11	0.31	0	1	0.15	0.35	0	1
Year 11 or below	0.18	0.38	0	1	0.23	0.42	0	1
Observations	12,940				10,705

Table 2

Locus of control and husbands' domains of job satisfaction using Mundlak method

Husbands' domains of job satisfaction	Job satisfaction	Pay satisfaction	Job security satisfaction	Work satisfaction	Hours satisfaction	Balance satisfaction	Real personal income	Job promotion
Wives' locus of control	0.0221*	0.0347***	0.00194	−0.00276	0.0373***	0.000347	−0.00808	−0.0032
	(0.0131)	(0.0132)	(0.0132)	(0.0132)	(0.0131)	(0.0130)	(0.0072)	(0.0046)
Locus of control	0.126***	0.0713***	0.0825***	0.128***	0.0685***	0.0596***	0.0160**	0.00853*
	(0.0132)	(0.0133)	(0.0133)	(0.0133)	(0.0132)	(0.0131)	(0.0073)	(0.0047)
Age	−0.0238**	0.0156	−0.0193	−0.0103	−0.0229*	−0.00547	0.0841***	−0.0111***
	(0.0119)	(0.0120)	(0.0119)	(0.0120)	(0.0119)	(0.0118)	(0.0065)	(0.0042)
Age squared	0.0303**	−0.000863	0.0137	0.0126	0.0388***	0.0175	−0.0715***	0.00425
	(0.0130)	(0.0132)	(0.0131)	(0.0131)	(0.0131)	(0.0130)	(0.0072)	(0.0046)
Total resident children	0.0163	0.0165	−0.00853	0.012	−0.00913	−0.0166	0.0162*	−0.00581
	(0.0152)	(0.0153)	(0.0153)	(0.0153)	(0.0152)	(0.0151)	(0.0083)	(0.0054)
Postgraduate degree	0.0124	−0.174	−0.13	0.24	−0.0207	0.188	0.136	−0.150**
	(0.1920)	(0.1940)	(0.1930)	(0.1940)	(0.1930)	(0.1910)	(0.1060)	(0.0677)
Graduate diploma/Certificate	−0.0561	−0.0577	−0.322*	0.0563	0.114	0.295*	0.034	−0.0803
	(0.1690)	(0.1710)	(0.1700)	(0.1710)	(0.1700)	(0.1680)	(0.0930)	(0.0596)
Bachelors or honors	−0.0604	−0.119	−0.22	−0.103	−0.00384	0.272*	−0.013	0.00651
	(0.1650)	(0.1670)	(0.1660)	(0.1660)	(0.1650)	(0.1640)	(0.0906)	(0.0581)
Advanced diploma, diploma	−0.101	−0.113	−0.274*	−0.0694	−0.0153	0.164	−0.0542	−0.065
	(0.1410)	(0.1430)	(0.1420)	(0.1420)	(0.1420)	(0.1410)	(0.0776)	(0.0497)
Certificate III or IV	−0.0829	0.00893	−0.225**	0.000104	0.0441	0.118	−0.0802	−0.0398
	(0.1090)	(0.1100)	(0.1090)	(0.1100)	(0.1090)	(0.1080)	(0.0598)	(0.0383)
Year 12	−0.161	−0.337**	−0.316**	−0.162	−0.172	0.11	−0.0951	−0.148***
	(0.1470)	(0.1490)	(0.1480)	(0.1490)	(0.1480)	(0.1470)	(0.0810)	(0.0519)
State of residence dummies	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Wave dummies	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Means of all independent variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	12,940	12,940	12,940	12,940	12,940	12,940	12,940	12,940

Note(s): *p < 0.1 **p < 0.05 ***p < 0.01. Robust standard errors are in parentheses

Table 3

Locus of control and wives' domains of job satisfaction using Mundlak method

Wives' domains of job satisfaction	Job satisfaction	Pay satisfaction	Job security satisfaction	Work satisfaction	Hours satisfaction	Balance satisfaction	Real personal income	Job promotion
Husbands' locus of control	0.00718	−0.00163	0.000961	−0.00738	0.0234	−0.0079	−0.00557	0.00815
	(0.0150)	(0.0147)	(0.0150)	(0.0153)	(0.0159)	(0.0154)	(0.0096)	(0.0050)
Locus of control	0.114***	0.0599***	0.0628***	0.0790***	0.0766***	0.0770***	0.00708	0.00472
	(0.0152)	(0.0149)	(0.0151)	(0.0154)	(0.0160)	(0.0155)	(0.0097)	(0.0050)
Age	−0.0145	0.0245*	−0.0121	0.00458	−0.0301**	−0.0351**	0.0801***	−0.00297
	(0.0137)	(0.0134)	(0.0137)	(0.0139)	(0.0145)	(0.0140)	(0.0088)	(0.0045)
Age squared	0.0122	−0.0142	0.00468	−0.00225	0.0414**	0.0406**	−0.0616***	−0.00277
	(0.0155)	(0.0151)	(0.0154)	(0.0157)	(0.0163)	(0.0158)	(0.0099)	(0.0051)
Total resident children	0.0297	0.000565	−0.0452**	0.0175	0.0366*	0.0492***	−0.102***	−0.0143**
	(0.0184)	(0.0181)	(0.0184)	(0.0188)	(0.0195)	(0.0189)	(0.0118)	(0.0061)
Postgraduate degree	0.15	0.13	0.275	0.398**	0.139	−0.304	0.311***	−0.0052
	(0.1820)	(0.1780)	(0.1810)	(0.1850)	(0.1920)	(0.1860)	(0.1160)	(0.0600)
Graduate diploma/Certificate	0.148	−0.00378	0.172	0.208	0.12	−0.145	0.131	−0.0647
	(0.1590)	(0.1550)	(0.1580)	(0.1610)	(0.1670)	(0.1620)	(0.1010)	(0.0524)
Bachelors or honors	−0.11	0.0186	−0.041	0.201	−0.0499	−0.304**	0.206**	−0.0159
	(0.1440)	(0.1410)	(0.1440)	(0.1470)	(0.1520)	(0.1480)	(0.0922)	(0.0477)
Advanced diploma, diploma	−0.0889	−0.0312	−0.183	−0.126	0.0315	0.122	−0.00225	−0.0906**
	(0.1270)	(0.1250)	(0.1270)	(0.1300)	(0.1350)	(0.1300)	(0.0813)	(0.0421)
Certificate III or IV	−0.0843	−0.106	−0.055	0.015	0.02	−0.148	0.0153	−0.0203
	(0.0903)	(0.0884)	(0.0900)	(0.0919)	(0.0954)	(0.0924)	(0.0577)	(0.0298)
Year 12	−0.0337	−0.0532	−0.00668	0.0176	0.00897	−0.0833	−0.185**	−0.039
	(0.1270)	(0.1240)	(0.1260)	(0.1290)	(0.1340)	(0.1300)	(0.0809)	(0.0418)
State of residence dummies	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Wave dummies	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Means of all independent variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	10,705	10,705	10,705	10,705	10,705	10,705	10,705	10,705

Note(s): *p < 0.1 **p < 0.05 ***p < 0.01. Robust standard errors are in parentheses

Note

1. In our data set, the information of gender – male or female – is provided by the respondents. However, the information of the partner can be either married or de facto (live together as a couple but are not married) and include same-sex couples. However, the number of observations for same-sex couples is very small: male–male is 127 (0.97% of total observations of males), and female–female is 205 (1.88% of total observations of females). Thus, we make a judgment call to exclude same-sex couples from the research parameters due to the small sample size. However, for the development in the field of gender studies, this group should not be neglected in future research.

Appendix

The supplementary material for this article can be found online.