The role of affectivity in job satisfaction: a meta-analysis

https://doi.org/10.1016/S0191-8869(99)00192-0Get rights and content

Abstract

Correlations between affectivity and job satisfaction measures were examined by cumulating research findings across studies. Correlations between (1) job satisfaction and positive affectivity, (2) job satisfaction and negative affectivity and (3) job satisfaction and affective disposition, were separately analyzed. The mean correlation corrected for coefficient alpha in the two measures correlated were: 0.49 for positive affectivity (N=3326, k=15), −0.33 for negative affectivity (N=6233, k=27) and 0.36 for affective disposition (N=1415, k=7). Results indicated that 10–25% of variance in job satisfaction could be due to individual differences in affectivity. No strong moderator variables were found. Implications for a dispositional and situational source of job satisfaction are discussed.

Introduction

Job satisfaction has been defined as a pleasurable or positive emotional state resulting from an appraisal of one's job (Locke, 1969). Job satisfaction is one of the most important and widely researched variables in industrial–organizational psychology (Locke, 1976). Hackman and Oldham (1976) identified five job characteristics that determined job satisfaction, which are skill variety, task identity, task significance, autonomy and job feedback. Currently, researchers have identified many more determinants of job satisfaction. Agho, Mueller and Price (1993) discuss several more determinants of job satisfaction. These encompass distributive justice, supervisory support, the internal labor market (respective to the organization), integration or friendships among coworkers and pay. Agho et al. (1993) also identified determinants that led to lower job satisfaction as well. These include opportunity of available jobs outside organization, role ambiguity, role conflict and role overload. Other determinants of job satisfaction include organizational constraints (O'Connor, Peters, Rudolf & Pooyan, 1982), work–family conflict (Bedeian, Burke & Moffett, 1988) and work schedules (Erberhardt and Shani, 1984, Jamal and Baba, 1992, Ralston, 1989, Ronen and Primps, 1981).

Relationships of job satisfaction with experienced strain (Bogg and Cooper, 1995, Brief et al., 1988, Viswesvaran and Deshpande, 1996), job performance (Iffaldano & Muchinsky, 1985) and turnover (Carsten & Spector, 1987) have been widely researched. Job satisfaction has shown to be significantly related to both organizational commitment and employee turnover (Carsten and Spector, 1987, Schlesinger and Zornitsky, 1991). Despite this voluminous research, most studies have focused on situational variables effecting job satisfaction and dispositional factors effecting job satisfaction such as personality have garnered far less attention (Brief & Roberson, 1989).

The recent years have seen attempts to address this dearth of research in the dispositional factors affecting job satisfaction. Researchers (e.g. Watson & Slack, 1993) have attempted to link individual characteristics to job satisfaction. An individual characteristic that has been widely studied is affective disposition (Staw, Bell & Clausen, 1986). Although Agho et al. (1993) focused on situational determinants of job satisfaction, they also found that their model of job satisfaction significantly improved when work motivation, positive affectivity and negative affectivity were added. Furthermore, Arvey, Bouchard, Segal and Abraham (1989) presented evidence that determinants of job satisfaction may be genetically inherited. That is, when monozygotic twins reared apart were examined with the Minnesota Job Satisfaction Questionnaire (MSQ), results indicated that about 30% of the variance in job satisfaction were attributable to genetic components. Additionally, Watson and Slack (1993) found significant correlations for both positive affectivity and negative affectivity with job satisfaction and that both positive affectivity and negative affectivity predicted certain job satisfaction facets even 2 years later. The objective in this study is to meta-analytically integrate the literature on the relationship between affectivity and job satisfaction.

Research on the theoretical explanation of the role of affectivity in job satisfaction is pertinent and can be integrated into theories of job satisfaction. Several researchers have posited a needs theory of job satisfaction (e.g. Wolf, 1970). These theories in part detail psychological needs that must be met in order for the individual to experience job satisfaction. Specifically, affective dispositions provide the perceptual foundation from which these needs are interpreted as being met or not. Hackman and Oldham (1976) present the well known job characteristic theory of job satisfaction. Although, this theory concentrated on the actual tasks performed on the job, Hackman and Oldham added a personality variable, Growth need strength, which moderated the job tasks–job satisfaction relationship. Affective dispositions may provide the cognitive catalyst from which Growth need strength operate.

Indeed, perception and cognition play a significant role in theories of job satisfaction (James & James, 1989). Moyle (1995) has suggested that individuals who are high on the Negative Affectivity Scale perceive their environment generally in negative terms and thus these individuals perceive work as negative, resulting in low job satisfaction. Levin and Stokes (1989) view negative affectivity as resulting from a cognitive process that evaluates negative cues from work more acutely, which manifests eventually into low job satisfaction. The same explanation suggests why individuals with positive affectivity have higher job satisfaction.

In addition, theories of job satisfaction have been advanced that indicate job satisfaction is stable over time (Staw & Ross, 1985) and seems to reflect a genetic source (Arvey et al., 1989). If a correlation exists between stable affective dispositions and job satisfaction, then such a finding explains why job satisfaction is stable over time and covaries within the twin data. That is, the explanation becomes that affective disposition and not job satisfaction per se is inherited. Although the meta-analytically-derived correlation by itself does not prove this causality, such a correlation supports this explanation.

Further, affective dispositions can generally be mapped into the Big Five framework. For example, measures of negative affectivity have been classified as measures of emotional stability or neuroticism (Watson & Clark, 1984), whereas positive affectivity has been linked to extraversion (George, 1992). Thus, although we focus in this study on the relation between affective dispositions and job satisfaction, the results reported can also be interpreted within the Big Five framework. This is important to the extent that this research can be added to the vast research on personality and organizational constructs, as the Big Five framework has established a unifying framework to organize the research on the role of personality at work. Linking positive affect to extraversion and negative affect to neuroticism establishes the theoretical and empirical foundation for the meta-analysis. Both extraversion and neuroticism have been identified as major personality traits in all theories and models of personality.

Linking affective disposition to personality traits is important because such links provide the theoretical basis for the hypothesis that negative and positive affect are two separate construct dimensions and not bipolar ends of the same dimension. Whether or not positive and negative affectivity are the two polar ends of the same construct has been a source of debate among researchers (Schuabroeck, Ganster & Fox, 1992). Factor analytic studies (Watson, 1988, Watson and Pennebaker, 1989) have shown the presence of two factors. However, as the works of Spector and his colleagues indicate, the emergence of two factors may be due to item wording. Thus, it is possible that all positively phrased items load onto a factor whereas all negatively oriented items load onto a second factor. Perhaps, a more persuasive line of evidence to examine whether positive and negative affectivity are the polar ends of the same construct is to examine the correlation between them. A high magnitude of the correlation indicates they are bipolar; a low value for the correlation suggests they are not bipolar. The correlation between measures of negative affectivity and measures of positive affectivity as reported in the extant literature varies from a low of r=−0.05 (Brief & Roberson, 1989) to a high of r=−0.39 (Judge & Locke, 1993). Note, however, that these are observed correlations uncorrected for any artifacts. To investigate whether measures of positive affectivity and measures of negative affectivity are assessing the same bipolar construct the true score correlation is perhaps more appropriate. Connolly and Viswesvaran (1999) report a moderate correlation of r=−0.30 between positive and negative affect, which suggest that the two are also empirically distinct (in addition to the theoretical and conceptual differences presented between them). These empirical results coupled with the research on the independence of extraversion and neuroticism dimensions support the independence of positive and negative affect.

Several measures of negative and positive affectivity exist in the literature. The Positive and Negative Affect Schedule (PANAS) scale developed by Watson, Clark and Tellegen (1988) is a commonly used measure. Another measure is the Multidimensional Personality Questionnaire (MPQ; Tellegen, 1982). The PANAS and the MPQ measure both positive affectivity and negativity affectivity. However, there exist more measures of negative affectivity than there are measures of positive affectivity. Perhaps this is a reflection of the fact that affectivity was studied more in the literature on depression in clinical settings before their relevance in explaining normal behavior was recognized. Specifically, Watson and Clark (1984) identified 18 measures of negative affectivity. These 18 scales encompass a wide variety of personality measures, variously labeled as trait anxiety, neuroticism, ego strength, general maladjustment and repression–sensitization. Recently, Viswesvaran and Sanchez (1998) demonstrated the measurement equivalence of the 18 scales using the principle of tetrad differences. The principle of tetrad differences implies that the second order determinants should vanish (Spearman, 1904) and this was found to be the case with the 18 measures of negative affect (Viswesvaran & Sanchez, 1998). This measurement equivalence suggests that these scales can be grouped together in a meta-analysis as measures of the same construct. In addition, Judge and Hulin (1993) investigate the effects of affective disposition defined as a tendency to respond to classes of environmental stimuli in a pre-determined, affect based manner. In the paper, we meta-analyzed the correlations between such measures of affective disposition (Judge and Hulin, 1993) and job satisfaction separately from studies reporting the correlation between job satisfaction and positive or negative affect.

Several Studies have examined the affectivity and job satisfaction relation. For example, Levin and Stokes (1989) using a sample of 315 participants from a large international professional service organization found negative affectivity to be a significant predictor of job satisfaction. Judge (1993) found using a sample of medical personnel that the job dissatisfaction and turnover relationship was stronger for individuals high on positive affectivity. However, Necowitz and Roznowski (1994) failed to find significant correlates between negative affectivity and job satisfaction for those who were not in aversive task conditions.

The objective in this study is to cumulate, using the principles of psychometric meta-analysis (Hunter & Schmidt, 1990), the extant literature examining the correlation between measures of affective dispositions and measures of job satisfaction. Three separate meta-analytic cumulations were undertaken. The correlations between (1) measures of negative affectivity and job satisfaction, (2) measures of positive affectivity and job satisfaction and (3) measures of affective disposition and job satisfaction, were meta-analytically cumulated. We expected negative affectivity to correlate negatively with job satisfaction and positive affectivity to correlate positively with job satisfaction. Further, we expected the positive affectivity–job satisfaction relation to be stronger than the negative affectivity–job satisfaction relation. The prediction that positive affectivity will correlate more with Job Satisfaction than does negative affectivity is based on research on the biological determinants of extraversion and neuroticism (Eysenck and Pervin, 1990, Strelau and Eysenck, 1987). The correlation between extraversion and physiological responses has been found to be more marked than the correlation between neuroticism and physiological responses. Coupled with research linking positive affectivity to extraversion and negative affectivity to neuroticism (cf. George, 1996), we expect positive affectivity to have stronger relationships with outcomes such as job satisfaction than the relationship of negative affectivity to job satisfaction.

In addition to a meta-analysis of the job satisfaction–affectivity relationship, an analysis of potential moderators of the relationship between job satisfaction and affectivity was undertaken (see Table 1). Several potential moderators such as type of job satisfaction measure used, organizational tenure of the participants, organizational sector (i.e. for profit and non-profit), organizational size in terms of the number of associates employed and age of the participants were examined. A list of potential moderators and their levels are summarized in Table 1. Note, however, that the list of potential moderators and their levels were limited to the data available in the studies. Additionally, the potential moderator analyses were undertaken more in an exploratory spirit.

Section snippets

Database

A search to locate all studies reporting a correlation between job satisfaction assessments and measures of affective disposition was conducted. First, the PsycLit database was electronically searched. The studies found were then examined to reveal more leads for additional studies to be included in the meta-analysis. Conference presentations were also included. A total of 27 articles were marked as relevant and were included in the database. The measures of affective dispositions that went

Results

Table 4 provides the frequency weighted (i.e. unweighted by sample size) mean and standard deviation of the square root of the reliability estimates. Coefficient alphas for all three distributions of affectivity and one distribution of job satisfaction are summarized. Across the four distributions, the mean value of the reliability estimates is highest for job satisfaction and the lowest for affective disposition. This probably reflects the fact that scales of job satisfaction have been in use

Discussion

The correlations of −0.33 for negative affectivity, 0.49 for positive affectivity and 0.36 for affective disposition suggest that 10–25% of the variance in general job satisfaction can be explained by affectivity or affective disposition. Although there is a substantially significant affective component to general job satisfaction, it is important to note that a sizeable percent of the total variance remains unexplained. There also are other factors that contribute to job satisfaction.

It

References (66)

  • M.R. Barrick et al.

    The Big Five personality dimensions and job performance

    Personnel Psychology

    (1991)
  • A.G. Bedeian et al.

    Outcomes of work-family conflict among married male and female professionals

    Journal of Management

    (1988)
  • J. Bogg et al.

    Job satisfaction, mental health and occupational stress among senior civil servants

    Human Relations

    (1995)
  • A.P. Brief et al.

    Should negative affectivity remain an unmeasured variable in the study of job stress?

    Journal of Applied Psychology

    (1988)
  • A.P. Brief et al.

    Job attitude organization: an exploratory study

    Journal of Applied Social Psychology

    (1989)
  • J.M. Carsten et al.

    Unemployment, job satisfaction and employee turnover: a meta-analytic test of the Muchinsky model

    Journal of Applied Psychology

    (1987)
  • P.Y. Chen et al.

    Negative affectivity as the underlying cause of correlations between stressors and strains

    Journal of Applied Psychology

    (1991)
  • Connolly, J. J. & Viswesvaran, C. (1999). Positive and negative affect: a meta-analysis. Paper presented at the...
  • B.J. Erberhardt et al.

    The effects of full-time versus part-time employment status on attitudes toward specific organizational characteristics and overall job satisfaction

    Academy of Management Journal

    (1984)
  • H.J. Eysenck

    Biological dimensions of personality

  • H.J. Eysenck et al.

    Manuel of the Eysenck Personality Questionnaire

    (1975)
  • T.P. Flannery et al.
  • J.M. George

    The role of personality in organizational life: issues and evidence

    Journal of Management

    (1992)
  • J.M. George

    Leader positive mood and group performance: the case of customer service

    Journal of applied Social Psychology

    (1995)
  • J.M. George

    Trait and state affect

  • R.L. Heneman
  • J.E. Hunter et al.

    Methods of meta-analysis: correcting error and bias in research findings

    (1990)
  • M.T. Iaffaldano et al.

    Job satisfaction and job performance: a meta-analysis

    Psychological Bulletin

    (1985)
  • M. Jamal et al.

    Shiftwork and department-type related to job stress, work attitudes and behavioral intentions: a study of nurses

    Journal of Organizational Behavior

    (1992)
  • L.A. James et al.

    Integrating work environment perceptions: explorations into the measurement of meaning

    Journal of Applied Psychology

    (1989)
  • L.A. James et al.

    Perceived job characteristics and job satisfaction: an examination of reciprocal causation

    Personnel Psychology

    (1980)
  • T.A. Judge

    Does affective disposition moderate the relationship between job satisfaction and voluntary turnover?

    Journal of Applied Psychology

    (1993)
  • T.A. Judge et al.

    Effect of dysfunctional thought processes on subjective well-being and job satisfaction

    Journal of Applied Psychology

    (1993)
  • Cited by (327)

    View all citing articles on Scopus
    View full text