Job Satisfaction, Mental Symptoms, and Well-Being in Adult Workers: A Gender Analysis

Abstract

Although studies have found that job satisfaction has an impact on workers’ physical and mental health, research has generally not focused on the psychological well-being of adult workers and a gender differential analysis has not been conducted. The aim of the current research is to determine the importance of job satisfaction for mental symptoms and well-being among adult working women and men. We also examine gender differences in job satisfaction. A non-probability sample of 1977 Spanish workers (51.6% men and 48.4% women) aged between 36 and 65 years was used in this cross-sectional study. Six questionnaires and self-report scales were used to assess the participants. For both men and women, higher job satisfaction was associated with lower depressive, somatic, anxiety, and social dysfunction symptoms; higher life satisfaction; and greater psychological well-being. Hierarchical multiple regression analyses showed that after controlling for the effects of self-esteem and social support, higher job satisfaction predicted greater life satisfaction, fewer mental symptoms, and greater psychological well-being, although the effect of job satisfaction on psychological well-being was somewhat stronger for men than for women. We conclude that job satisfaction is important for the mental health, psychological well-being, and life satisfaction of adult working women and men.

1. Introduction

Although the term is widely used in the scientific research community as well as in day-to-day life, there is no general consensus on what constitutes job satisfaction [1,2]; it has been conceptualized as a cognitive evaluation, an emotional state, or both [3]. According to Locke [4], job satisfaction refers to a positive or pleasurable emotional state based on the appraisal of a person’s professional experience at work, resulting from the perception that work fulfills or enables the fulfillment of important values that the person holds about work. Intrinsic job satisfaction refers to work activities and tasks, whereas extrinsic job satisfaction refers to rewards associated with work activities and tasks [5]. Although job satisfaction is a multifactorial and intricate concept that can have different meanings for different individuals [1], it refers to a pleasant emotional state that results from a person’s evaluation of his or her work and is tied to well-being at work [6].

The study of job satisfaction, as an employee’s feelings about his or her job and its various aspects, is among the most widely studied phenomenon of the 20th century and is central to many aspects of organizational and industrial psychology [7,8,9]. Despite the numerous studies carried out, it is a fragmented field in which two types of motives could be distinguished that justify the interest in studying it: (1) the relevance of job satisfaction in explaining different variables related to organizational effectiveness, and (2) its moral relevance, as it is an important determinant of the well-being of the individuals, both in terms of his or her physical and mental health [8]. Many studies have been conducted from the first perspective and have found that higher job satisfaction is linked to outcomes such as positive attitudes toward one’s work, work engagement, higher productivity, and lower work absenteeism and job turnover [10,11,12,13,14]. However, the second perspective is critical given the relevance of work to the mental health of employees. There is evidence that job satisfaction is relevant to employees’ mental and physical health and quality of life. Job dissatisfaction has been found to be associated with lower life satisfaction and greater job stress and burnout, as well as with physical health problems and possibly mental health [7,15]. Evidence suggests that employees with higher levels of stress are more likely to develop anxiety symptoms [16], while job security contributes positively to mental health [17]. A systematic review and meta-analysis by Faragher et al. [18] revealed that job satisfaction was associated with mental and physical health, although there was considerable variation in the magnitude of the relationship. The overall combined correlation was r = 0.312, with higher correlations for mental health and psychological problems than for physical health. The highest correlations were for burnout, self-esteem, depression, and anxiety.

Job association is associated with life satisfaction, with positive affect, with absence of negative affect and with happiness. Positive affect and the absence of negative affect, and also with happiness were reported in a meta-analysis by Bowling et al. [19]. Job satisfaction has been shown to have a protective impact on health, self-esteem, happiness, and subjective well-being [20], and has also been associated with a reduced response to and improved recovery from stressors [21]. There are many antecedents of job satisfaction [8,10]. Although many of these antecedents are related to organizational variables, working conditions, and interpersonal relationships at work, in addition to the personal variables of employees [2,10], the relevance of sociodemographic variables, including age, education, marital status, or having children, has also been suggested [13,22,23], although there is no consensus on these findings. Another of the variables examined, which is central to the present study, is gender.

Gender is the socially constructed attributes that define men, women, boys, and girls, according to World Health Organization (WHO) [24]. Gender encompasses the norms, roles, and behaviors associated with being a woman or a man, and the relations between them. The WHO also emphasizes that gender is hierarchical and produces inequalities that intersect with other inequalities including age and socioeconomic status, among other social and economic inequalities that put women’s health and well-being at risk [24]. As Risman [25] argues, gender is more than a personal identity; it is a social structure that has consequences for the person (creating differentiated “selves”), their interpersonal interactions, and social institutions and organizations in which differences in the rights and responsibilities of women and men still exist. Gender is a social structure that limits the development of individuals and assigns them different roles and positions [26]. Gender inequality is produced, maintained, and reproduced at all levels (individual, interactional, and institutional) [27]. Gender roles impose numerous demands on women and men that have an impact on their working and family lives, and organizations, as socially constructed entities, strengthen gender role distinctions [28].

Despite progress in recent decades, gender inequality persists worldwide [29,30,31]. As the United Nations argues, “gender equality is not only a fundamental human right, but a necessary foundation for a peaceful, prosperous and sustainable world” [29]. Gender in-equalities affect the labor market, where gender gaps are widely reported [32] and phenomena such as occupational segregation persist [33]. Globally, women continue to earn, on average, 23% less than men in the labor market, and women spend about three times as many hours on care and unpaid domestic work as men [29]. In addition to being structural, such differences are produced by social norms about the roles assigned to women and men [33], as well as by myths and stereotypes that persist in society. As argued by Barnet et al. [34], despite societal changes, particularly the incorporation of women into the workforce, there are still a number of myths and stereotypes that attempt to justify and naturalize the differences between men and women in family and work roles. Some enduring myths about gender and work include (1) the happy homemaker; (2) the husband is the “breadwinner”; (3) the mother is the natural parent; (4) children need to be cared for at home by their mothers; (5) the boundaries between work and family are less clear for mothers than for fathers; (6) work–family/work–life balance is a “women’s” issue; and (7) women are treated equally at work [34].

Despite gender differences in the labor market, with women on average having lower wages, fewer opportunities for advancement and for career development, poorer working conditions, more workplace discrimination, and fewer benefits than men [32,35,36,37], women have been found to be more satisfied with their jobs than men [32,35,36], a phenomenon referred to as “the gender job satisfaction paradox” [32,33]. However, recent research has shown that job satisfaction among working women has declined in some countries [38] and that differences in job satisfaction between women and men vary according to other variables such as ethnicity or country [35,39]. In some countries and occupations, women are more dissatisfied with their jobs than men [40,41,42], while other studies have found no differences in job satisfaction between women and men [11,43,44], or small differences [45].

Although the reasons for such discrepancies are unknown, they may be due to the fact that many studies have not analyzed gender differences as their main objective, or have not taken into account that gender may interact with other variables. In addition, many studies have been conducted in specific or feminized occupations (e.g., nursing) and do not take into account the relevance of life cycle stage on the job. Therefore, the present work focuses on adulthood, a life stage in which work and family are the two main domains in all societies [46]. We analyze a large group of women and men between the ages of 36 and 65 years and with different occupational levels.

The main aim of the current study is to determine the importance of job satisfaction on mental symptoms, life satisfaction, and psychological well-being in adult men and women employees. First, we analyze gender differences in job satisfaction, analyzing such differences from an intersectional perspective and taking into account educational level, marital status, occupational status, and type of job.

2. Materials and Methods

2.1. Participants and Procedure

The sample comprised 957 women and 1020 men between the ages of 36 and 65 years from the general Spanish population. All participants were employed and active, although they had different occupational categories and different types of jobs, as shown in

Table 1. Sociodemographic and work characteristics of women and men.

	Women (n = 957)		Men (n = 1020)		χ2-Value
	n	%	n	%
Education
Primary	283	29.6	325	31.9	19.02 ***
Secondary	263	27.5	350	34.3
University	411	42.9	345	33.8
Occupation
Skilled/unskilled manual	299	31.2	366	35.9	19.55 ***
Non-manual	330	34.5	397	38.9
Professional	328	34.3	257	25.2
Job type
Permanent	650	67.9	792	77.6	49.25 ***
Contracted	268	28	158	15.5
Self-employed	39	4.1	70	6.9
Marital status
Never married	112	11.7	146	14.3	55.87 ***
Married/partnered	650	67.9	786	77.1
Divorced/widowed	195	20.4	88	8.6
	M	SD	M	SD	t-value
Age	48.51	7.05	48.83	7.24	−0.99
Number of children	1.69	1.00	1.61	1.01	1.70

*** p < 0.001.

, which displays the sociodemographic and work characteristics of women and men. One-third of the sample had manual jobs, 36.8% had non-manual jobs, and one-quarter were professionals, a category used here if they had jobs requiring a university degree or diploma. The most common type of employment was permanent, which was the case for 72.9% of the sample; 21.4% had a contract and 5.5% were self-employed. There was also diversity in educational attainment and marital status, although most are married or cohabiting, accounting for 72.6% of the sample. A larger percentage of the sample (84.4%) had at least one child, with two (47.5%) or one (22.3%) being the most common, while 14.6% of the sample had three or more children. There were no statistically significant differences between men and women in age or number of children. However, there were statistically significant differences in other occupational and sociodemographic characteristics. As can be seen in Table 1, women were more likely than men to have a university education and to be employed in jobs requiring tertiary education, while men were more frequently employed in permanent jobs. Women were also more likely to be widowed or divorced, while men were more often married or living with a partner.

All of the participants were volunteers and did not receive any financial remuneration for their involvement in this research. The questionnaires were completed individually in a printed paper format after informed consent was obtained. Informed consent was verbal so that people did not have to sign or leave their personal information in any document.

The sample was recruited through working centers located in different Spanish municipalities and through the social network of students in their last two years of undergraduate or doctoral studies in psychology or sociology. All students who participated in administering the tests were trained previously and received course credit for their assistance. Efforts were made to ensure that students who participated in administering the tests gave them to actual respondents and did not administer the tests themselves, for example, using different pens or different handwriting. The actions taken were as follows: (1) The training prior to administering the tests, which was individualized for each student and conducted by one of the researchers, emphasized the ethical issues, stressed that the data were for research purposes, and limited the number of questionnaires to the number of contacts that the students felt were available. (2) Each questionnaire was checked for consistency in all its answers and also took into account the sociodemographic characteristics of each person. (3) Since one of the questionnaires was open-ended, in addition to a thorough analysis of the content, the handwriting was also carefully analyzed. If there was the slightest doubt that one of the questionnaires might have been falsified, all questionnaires from that participant were eliminated. The number of questionnaires eliminated was less than 1%, although in many cases, the elimination was because the questionnaire was incomplete.

The inclusion criteria for this study were the following: (1) between the ages of 36 and 65 years and (2) employed at the time the tests were administered. This research was performed according to the principles of the Declaration of Helsinki. No name or other information was used to identify participants, and individuals could withdraw from this study at any time. The present research is part of a wider study on gender and well-being, and was approved by the Ethical Committee on Animal Research and Well-Being of the University of La Laguna (number 2015-0170).

2.2. Measures

Job satisfaction was measured by means of the Work Role Satisfaction Questionnaire [47]. This is an open-ended questionnaire consisting of 5 questions about whether the person likes his/her job; whether he/she would have preferred a different job; whether he/she is thinking about changing jobs; to what extent it provides a sense of fulfillment; and whether his/her job makes him/her feel good about him/herself. Responses to each of the open-ended questions were scored quantitatively using a validated code [47]. For the present sample, the Cronbach’s alpha of the 5 items was 0.73.

The Spanish version [48] of the Goldberg General Health Questionnaire GHQ-28 [49] was used to assess mental health symptoms. The GHQ-28 is composed of 28 items divided into four scales, that are not independent, each with 7 items measuring somatic symptoms, anxiety and insomnia, social dysfunction, and severe depression. For the present sample, the Cronbach’s alphas for each of the scales were 0.84, 0.90, 0.73, and 0.89, respectively. The Cronbach’s alpha for the 28 questionnaire items was 0.93.

Psychological well-being was assessed using the Spanish translated and validated version of the Ryff Psychological Well-being Scale [50]. This scale includes 38 items and has been found to have a six-factor structure and a second-order latent construct of psychological well-being [50]. The response scale is in a 6-point Likert format ranging from 1 (strongly disagree) to 6 (strongly agree). The items are structured into the following six scales: self-acceptance, consisting of 6 items whose internal consistency (Cronbach’s alpha) for the present sample was 0.80; positive relationships, consisting of 6 items whose Cronbach’s alpha was 0.79; autonomy, which consisted of 8 items whose Cronbach’s alpha was 0.79; environmental mastery, which consisted of 6 items whose Cronbach’s alpha was 0.64; purpose in life, which consisted of 6 items whose Cronbach’s alpha was 0.83; and personal growth, which consisted of 6 items whose Cronbach’s alpha was 0.71. For the current sample, the Cronbach’s alpha of the 38 items corresponding to the latent construct of psychological well-being was 0.92.

The Satisfaction with Life Scale (SWLS) [51] was used to measure life satisfaction. It consists of 5 items that assess global satisfaction with life, which is considered the cognitive component of subjective well-being. The response scale is in a 7-point Likert format ranging from 1 (strongly disagree) to 7 (strongly agree). The Spanish translation of the SWLS available on the website of the first author of the original study was used. The Cronbach’s alpha for the present sample was 0.86.

Self-esteem was assessed with the Spanish version of the York Self-Esteem Inventory [52]. This 51-item inventory measures global self-esteem, including the personal, interpersonal, family, and achievement self-domains. The response scale is in a 4-point Likert format, from never, scored 0, to always, scored 3. For the present sample, Cronbach’s alpha was 0.94.

The Social Support Scale [53] was used to assess social support. This is a 12-item scale that measures perceived social support. The items are structured into two factors: emotional support, comprising 7 items, and instrumental support, comprising 5 items. The response scale is in a 4-point Likert format, from never, with a score of 0, to always, with a score of 3. For the present sample, the Cronbach’s alpha was 0.87 for emotional support and 0.85 for instrumental support.

2.3. Statistical Analysis

To determine whether there were differences in job satisfaction between women and men with different levels of education, occupations, types of job, and marital status, four analyses of variance (ANOVAs) were conducted with job satisfaction as the dependent variable. The bivariate associations between job satisfaction and the study variables were computed using Pearson’s correlation coefficient. The exception was educational level, which was computed using Spearman’s Rho since it is an ordinal variable. Hierarchical multiple regression analyses were used to investigate the relevance of job satisfaction for the mental symptoms, psychological well-being, and life satisfaction of both women and men, also considering their sociodemographic characteristics, self-esteem, and social support. In each regression analysis, the first step (Model 1) included age and number of children as quantitative variables, as well as educational level as an ordinal variable with 8 levels (from 0 for no studies to 7 for five-year university studies); the second step (Model 2) added the job satisfaction scores; and the third step (Model 3) added the scores on self-esteem, emotional support scores, and instrumental support scores. The dependent variables were the total symptom score (calculated by aggregating the 28 items of the GHQ-28) in the first regression analyses, the psychological well-being score in the second, and the life satisfaction score in the third. The Kolmogorov–Smirnov test was used to assess normality, and Levene’s test of equality of variances was used to assess the equality of variances. As recommended by Tabachnick and Fidell [54], a logarithmic transformation was applied to the mental symptoms score, because it was highly skewed with a predominance of low scores.

Multiple regression analyses and correlations were conducted separately for women and men. Statistical analyses were conducted using SPSS for Windows, version 22.0.

3. Results

Table 2. Means (M), standard deviations (SD), and two-way ANOVA statistics for job satisfaction.

Variable	Men		Women		ANOVA
	M	SD	M	SD	Effect	F Ratio	ηp2
Education
Primary	13.76	4.19	13.31	4.24	Gender	0.92	0.000
Secondary	14.01	4.04	13.70	4.19	Education	20.26 ***	0.020
University	14.22	4.19	15.50	3.41
Interaction Gender × Education					G × E	9.81 ***	0.010
Occupation
Skilled/unskilled manual	13.76	4.16	13.14	4.33	Gender	2.19	0.001
Non-manual	13.89	4.13	13.95	4.00	Occupation	30.92 ***	0.030
Professional	14.51	4.09	15.88	3.17
Interaction Gender × Occupation					G × O	9.76 ***	0.010
Job type
Permanent	13.95	4.12	14.60	3.90	Gender	0.66	0.000
Self-employed	14.90	2.77	15.20	3.27	Job Type	4.87 **	0.005
Contrated	13.85	4.67	13.65	4.28
Interaction Gender × Job Type					G × JT	1.68	0.002
Marital status
Never married	13.84	4.17	14.37	4.36	Gender	0.67	0.000
Married/partnered	13.98	4.20	14.45	3.97	Marital Status	0.09	0.000
Divorced/widowed	14.42	3.41	14.06	3.95
Interaction Gender × Marital Status					G × MS	1.12	0.001

** p < 0.01; *** p < 0.001.

shows the results of ANOVAs with job satisfaction as the dependent variable and gender (men, women) and education (primary, secondary, university) as factors in the first ANOVA; gender and occupation (skilled/unskilled manual, non-manual, professional) in the second ANOVA; gender and job type (permanent, contracted, self-employed) in the third ANOVA; and gender and marital status (never married, married/partnered, divorced/widowed) in the fourth ANOVA The interactions gender × education and gender x occupation were statistically significant (see Figure 1 and Figure 2). Post hoc analyses with Scheffe’s adjustment to determine the groups that had statistically significant differences between them showed that while there were no statistically significant differences in job satisfaction between men with different levels of education or with different occupations, there were significant differences between women. Women with university education were more satisfied with their jobs than women with secondary education (p < 0.001) and women with elementary education (p < 0.001); in addition, women with university education also had a higher job satisfaction than men with elementary education (p < 0.001), men with secondary education (p < 0.001), and men with university education (p = 0.002). Professional women had higher job satisfaction than women with manual occupations (p < 0.001) and non-manual occupations (p < 0.001). Professional women also had higher job satisfaction than men with manual occupations (p < 0.001), men with non-manual occupations (p < 0.001), and professional men (p = 0.005). In addition, professional men had higher job satisfaction than women with manual occupations did (p = 0.007).

In the ANOVA with gender and job type as factors, only the main effect of job type was statistically significant. Post hoc analyses with Scheffe’s adjustment revealed that there were statistically significant differences (p = 0.01) only between the self-employed and contracted employees. As shown in Table 2, the self-employed reported greater job satisfaction than the contracted employees. In the ANOVA in which the factors were gender and marital status, no statistically significant effects were found, nor were there main effects found for gender and marital status or the interaction of gender × marital status.

Table 3. Correlations between job satisfaction and study variables disaggregated by gender.

	Variable	Women	Men
Sociodemographic variables	Age	0.04	0.06
	Number of children	0.02	0.03
	Education a	0.25 ***	0.05
Mental symptoms	Somatic	−0.18 ***	−0.23 ***
	Anxiety and insomnia	−0.23 ***	−0.22 ***
	Severe depression	−0.21 ***	−0.26 ***
	Social dysfunction	−0.13 ***	−0.22 ***
Psychological well-being	Self-acceptance	0.29 ***	0.35 ***
	Positive relationships	0.16 ***	0.24 ***
	Autonomy	0.09 *	0.07 *
	Environmental mastery	0.25 ***	0.33 ***
	Purpose in life	0.30 ***	0.36 ***
	Personal growth	0.16 ***	0.18 ***
Social Support	Emotional	0.17 ***	0.17 ***
	Instrumental	0.17 ***	0.18 ***
	Life satisfaction	0.34 ***	0.35 ***
	Self-esteem	0.23 ***	0.22 ***

Notes. ^a = coefficient calculated using Spearman’s Rho. * p < 0.01; *** p < 0.001.

displays the results of the bivariate correlations between women’s and men’s job satisfaction and the study variables. As can be seen, for both genders, job satisfaction was independent of age and number of children; it was also independent of educational level for men, whereas for women, a higher level of education was associated with greater job satisfaction. In both genders, greater job satisfaction was associated with fewer mental symptoms, although the effect sizes were small, except for the association with social dysfunction in women, where the effect size was very small. Greater job satisfaction was associated with greater life satisfaction, greater self-esteem, greater social support, and greater psychological well-being, with small effect sizes, except for autonomy for both women and men, where the effect size was trivial.

Table 4. Summary of hierarchical regression with mental symptoms (logarithmic transformation) as the dependent variable for women and men.

	Model 1		Model 2		Model 3
	β	t-Value	β	t-Value	β	t-Value
Women
Age	−0.03	−0.95	−0.02	−0.73	−0.04	−1.54
Number of children	0.03	0.89	0.03	1.08	0.06	2.00 *
Education	−0.16	−4.91 ***	−0.11	−3.39 **	−0.08	−2.88 **
Job satisfaction			−0.20	−6.12 ***	−0.09	−3.11 **
Self-esteem					−0.36	−11.51 ***
Emotional support					−0.08	−1.90
Instrumental support					−0.08	−1.89
F change (df1, df2)	8.68 (3, 953) ***		37.50 (1, 952) ***		78.58 (3, 949) ***
Adj. R2	0.024		0.060		0.244
Men
Age	−0.03	−0.83	−0.02	−0.50	−0.03	−1.00
Number of children	−0.02	−0.62	−0.01	−0.28	0.01	0.17
Education	−0.08	−2.70 **	−0.07	−2.49 *	−0.06	−2.42 *
Job satisfaction			−0.26	−8.44 ***	−0.13	−4.87 ***
Self-esteem					−0.43	−15.47 ***
Emotional support					−0.03	−0.83
Instrumental support					−0.12	−3.03 **
F change (df1, df2)	2.95 (3, 1016) *		71.33 (1, 1015) ***		107.18 (3, 1012) ***
Adj. R2	0.006		0.007		0.292

Note: β = standardized regression coefficient. Adj. R² = percentage of explained variance. * p < 0.05; ** p < 0.01; *** p < 0.001.

shows the results of the multiple regression analyses predicting women’s and men’s mental symptoms. For both genders after Step 1, with the sociodemographic characteristics in the equation, the addition of job satisfaction resulted in a significant increase in R², with β = −0.20 for women and β = −0.26 for men. While the magnitude of this coefficient decreased for both women and men after Step 3, when self-esteem and social support were added to the equation, it remained statistically significant and contributed to the prediction of mental symptoms, although the main predictor for both genders was self-esteem, with lower self-esteem and lower job satisfaction in women and men with more symptoms. For women, other statistically significant predictors of more mental symptoms were less education and a greater number of children, while for men, it was less instrumental support and less education.

Table 5. Summary of hierarchical regression with psychological well-being as the dependent variable for women and men.

	Model 1		Model 2		Model 3
	β	t-Value	β	t-Value	β	t-Value
Women
Age	−0.08	−2.54 *	−0.09	−2.89 **	−0.06	−2.87 **
Number of children	0.02	0.584	0.01	0.38	−0.03	−1.37
Education	0.18	5.57 ***	0.12	3.78 ***	0.08	3.97 ***
Job satisfaction			0.24	7.44 ***	0.05	2.68 **
Self-esteem					0.63	29.96 ***
Emotional support					0.15	4.82 ***
Instrumental support					0.13	4.20 ***
F change (df1, df2)	12.52 (3, 953) ***		55.34 (1, 952) ***		511.93 (3, 949) ***
Adj. R2	0.035		0.087		0.650
Men
Age	−0.06	−1.69	−0.07	−2.24 *	−0.05	−2.34 *
Number of children	0.03	0.86	0.01	0.46	−0.01	−0.67
Education	0.11	3.42 **	0.09	3.22 **	0.08	3.83 ***
Job satisfaction			0.32	10.89 ***	0.14	6.79 ***
Self-esteem					0.58	27.40 ***
Emotional support					0.20	6.88 ***
Instrumental support					0.09	3.23 **
F change (df1, df2)	4.87 (3, 1016) *		118.53 (1, 1015) ***		1397.42 (3, 1012) ***
Adj. R2	0.011		0.114		0.592

Note: β = standardized regression coefficient. Adj. R² = percentage of explained variance. * p < 0.05; ** p < 0.01; *** p < 0.001.

shows the results of the multiple regression analyses predicting women’s and men’s psychological well-being. For both genders after Step 1, with sociodemographic characteristics in the regression equation, the addition of job satisfaction resulted in a significant increase in R², with β = 0.24 for women and β = 0.32 for men. Although the magnitude of this coefficient decreased for women and men after Step 3, when self-esteem and social support were added to the equation, it remained statistically significant and contributed to the prediction of psychological well-being for women and men, although in the final model, the relevance of job satisfaction was greater for men (β = 0.14) than for women (β = 0.05). According to the final model, with all variables in the equation, the most relevant predictor of greater psychological well-being for both genders was greater self-esteem, followed by greater emotional support. For men, the third most important predictor was greater job satisfaction, followed by greater instrumental support and greater education. For women, the third most relevant predictor of greater psychological well-being was greater instrumental support, followed by higher education and younger age, with greater job satisfaction being the least relevant predictor and making a modest contribution to predicting psychological well-being for women.

Table 6. Summary of hierarchical regression with life satisfaction as the dependent variable for women and men.

	Model 1		Model 2		Model 3
	β	t-Value	β	t-Value	β	t-Value
Women
Age	−0.03	−1.00	−0.04	−1.43	−0.01	−0.43
Number of children	0.06	1.79	0.05	1.59	0.02	0.82
Education	0.21	6.80 ***	0.14	4.55 ***	0.11	4.00 ***
Job satisfaction			0.31	9.96 ***	0.19	7.04 ***
Self-esteem					0.35	12.12 ***
Emotional support					0.15	3.55 ***
Instrumental support					0.12	2.93 **
F change (df1, df2)	16.16 (3, 953) ***		99.26 (1, 952) ***		118.22 (3, 949) ***
Adj. R2	0.045		0.135		0.368
Men
Age	0.00	0.09	−0.01	−0.40	0.01	0.24
Number of children	0.06	1.97 *	0.05	1.61	0.03	1.21
Education	0.07	2.37 *	0.06	2.11 *	0.05	1.79
Job satisfaction			0.35	11.83 ***	0.24	8.71 ***
Self-esteem					0.29	10.27 ***
Emotional support					0.17	4.31 ***
Instrumental support					0.09	2.31 *
F change (df1, df2)	3.36 (3, 1016) *		139.96 (1, 1015) ***		80.06 (3, 1012) ***
Adj. R2	0.007		0.126		0.292

Note: β = standardized regression coefficient. Adj. R² = percentage of explained variance. * p < 0.05; ** p < 0.01; *** p < 0.001.

shows the results of the multiple regression analyses predicting women’s and men’s life satisfaction. For both genders, after Step 1, with sociodemographic characteristics in the regression equation, the addition of job satisfaction resulted in a significant increase in R², with β = 0.31 for women and β = 0.35 for men. Although the magnitude of this coefficient decreased for women and men after Step 3, it remained statistically significant when self-esteem and social support were added to the equation, contributing to the prediction of women’s and men’s life satisfaction. According to the final model, with all variables in the equation, the most important predictor of higher life satisfaction for both genders was higher self-esteem, followed by higher job satisfaction, greater emotional support, and greater instrumental support. In addition, for women, another statistically significant predictor of higher life satisfaction was higher education.

4. Discussion

The present study, conducted with a sample of 1977 adult women and men with different sociodemographic and occupational characteristics, found that although there are more similarities than differences between women and men in terms of job satisfaction, there are some gender differences. It has been argued that job satisfaction is a key variable for many important social and organizational outcomes, as people who are more satisfied with their jobs are not only more motivated and productive at work, but also healthier and happier outside the workplace [55]. The main aim of the present study was to determine the relevance of job satisfaction for mental symptoms, psychological well-being, and life satisfaction for women and men. Although effect sizes were generally small, bivariate correlations indicated that higher job satisfaction was associated with greater life satisfaction, greater psychological well-being, greater self-esteem, greater social support, and fewer mental symptoms for both men and women, consistent with the literature [2,7,18,19,20].

Regression analyses also showed that greater job satisfaction was associated with better mental health and well-being for both women and men, although its relevance depended on the type of mental health indicator used and on gender. Because previous research has shown that job satisfaction is associated with self-esteem and social support [2,18,20,56], to determine whether the effects of job satisfaction on mental symptoms, psychological well-being, and life satisfaction persisted beyond the effects of self-esteem and social support, both variables were included in the regression equation (in Model 3), whereas only job satisfaction was included in Model 2. Although the magnitude of the effect decreased considerably after self-esteem and social support were included in the regression, job satisfaction remained a statistically significant predictor in all regression analyses. In predicting mental symptoms and life satisfaction, job satisfaction was the second most influential predictor after self-esteem for both genders, although its importance in predicting life satisfaction was greater than for mental symptoms. For men, the role of job satisfaction in predicting life satisfaction was quite similar to that of self-esteem (β = 0.29 for self-esteem and β = 0.24 for job satisfaction). For women, however, the beta weight of self-esteem was almost twice that of job satisfaction (β = 0.35 for self-esteem and β = 0.19 for job satisfaction). In predicting mental symptoms for women, the beta weight for self-esteem was four times greater than that for job satisfaction, whereas for men, it was 3.3% greater. For both groups of women and men, greater emotional support was more important than job satisfaction in predicting greater psychological well-being. However, the importance of greater job satisfaction in predicting greater psychological well-being for men was stronger than that of greater instrumental support, higher education, and younger age. For women, although greater job satisfaction was a statistically significant predictor of greater psychological well-being, greater emotional and instrumental support, more education, and younger age were more relevant in predicting psychological well-being.

We also analyzed gender differences in job satisfaction from an intersectional perspective, which included sociodemographic factors that have been shown to be relevant to job satisfaction in previous studies, such as educational level and marital status [13,22,23], and work-related factors such as occupational status and job type. The results of the present study showed that the differences in job satisfaction between women and men varied according to women’s educational level and occupational status. While neither educational level nor occupational status were differential factors in job satisfaction for men, women with higher educational level and higher occupational status had higher job satisfaction than other women and also higher job satisfaction than men. Although job satisfaction has been studied extensively, the findings on the existence of gender differences in job satisfaction are not consistent. While some studies have found that women are less satisfied with their jobs than men [41,42], others have found no differences [11,38,43], and still others have found that gender differences in job satisfaction vary by country [32,35,40] or intersect with other variables such as race/ethnicity [39].

The results of the present study only partially confirm the existence of the “gender job satisfaction paradox”. This paradox has been proposed in recent decades, when it was found that, although employed women have worse working conditions, greater discrimination, fewer opportunities for advancement, and lower salaries, their job satisfaction is higher than that of men [57]. Although our study did not directly analyze whether there are differences between women and men in terms of working conditions, the sociodemographic data suggest that there are. While women and men in the current study do not differ in age, and women are more likely than men to have a university education, they are less likely than men to have a permanent job. In any case, the higher job satisfaction of professional and tertiary educated women compared to the other groups suggests that high levels of education and employment promote women’s well-being.

The analysis of the relevance of gender and job type (permanent, contracted, or self-employed) on job satisfaction showed that only job type was relevant for job satisfaction. Self-employed workers were more satisfied than contracted employees, a result that is consistent with previous findings [58,59]. In addition, marital status was not related to job satisfaction, as found in previous research [11]. Correlational analyses showed that while women’s educational level was associated with higher job satisfaction, men’s job satisfaction seemed to be independent of their educational level. These findings are consistent with those of other studies that found that the association between education level and job satisfaction was found to be dependent on other variables, including gender [22,23].

This study advances the current state of knowledge by increasing the evidence for the relevance of job satisfaction to mental health, a relevance that is maintained even when controlling for the effects of two variables that have been shown to be highly relevant to mental health: self-esteem and social support. Although these are important findings, this study has a number of limitations. Among them are the following: The design is cross-sectional, so causal inferences cannot be made. The sample is large but not probabilistic. All the measures used are self-reports, which are subject to a number of biases, the most notable of which is social desirability. In addition, all participants lived in Spain, so the results may not be generalizable to other countries. Future research should use probability samples, multi-method assessments, and longitudinal designs to allow for causal inference. It would also be important to use samples from different countries.

5. Conclusions

The results of the present study enable us to conclude that job satisfaction is important for the mental health and well-being of women and men, as it is associated with greater life satisfaction, fewer mental symptoms, and greater psychological well-being. Gender is relevant to job satisfaction, which is greater in women with higher occupational levels and university education, so higher education for women and quality jobs should be promoted. The results of the current study are relevant for the design of policies and programs, both at the organizational and public health levels, aimed at promoting job satisfaction among workers, thus contributing to improving the health and well-being of the population and gender equality.