Discussion 2: Theories of Work Motivation and Job Attitudes
melodyharris257
ExploringDifferentFormsofJobDisSatisfactionandTheirRelationshipwithWellBeingMotivationandPerformance..pdf
Exploring Different Forms of Job (Dis)Satisfaction and Their Relationship with Well-Being, Motivation
and Performance
Carrie Kovacs*, Barbara Stiglbauer and Bernad Batinic Johannes Kepler University, Austria
Timo Gnambs Leibniz Institute for Educational Trajectories, Germany
Job satisfaction is often treated as a one-dimensional construct. In contrast, Brug- gemann (1974) postulated six distinct forms of (dis)satisfaction: four types of sat- isfaction (progressive, stabilised, resigned, pseudo) and two types of dissatisfaction (constructive, fixated). Despite her theory�s practical relevance, few researchers have explored its assumptions or applications. The current study aimed to charac- terise a German-speaking employee sample (n 5 892) according to Bruggemann�s theory using mixture modelling. We investigated stability of the (dis)satisfaction forms over a five-month period, as well as their relationship with well-being, moti- vation and (self-reported) performance. We found latent clusters corresponding to most Bruggemann types, though no distinction between progressive and stabilised satisfaction was possible. While cluster membership varied over time, some clusters (e.g. resigned satisfaction) were more stable than others (e.g. constructive dissatisfaction). Overall satisfaction level explained 25–51 per cent variance in well-being and motivation, and 13–16 per cent variance in performance. Including forms of satisfaction improved cross-sectional prediction by 2–6 per cent explained variance. Results suggest that unfavourable consequences of job dissat- isfaction may be limited to fixated—not constructive—dissatisfaction, though no consistent longitudinal effects emerged. We argue that exploring qualitative differ- ences in job satisfaction promotes a more nuanced and potentially useful under- standing of the relationship between satisfaction and work outcomes.
INTRODUCTION
Job satisfaction is one of the most prevalent topics in work and organisational psychology research. Numerous empirical studies and meta-analyses explore the relationship between job satisfaction and other work-related predictors or
* Address for correspondence: Dr Carrie Kovacs, Department of Work, Organizational and Media Psychology, Johannes Kepler University Linz, Altenberger Strasse 69, 4040 Linz, Austria. Email [email protected].
VC 2017 International Association of Applied Psychology.
APPLIED PSYCHOLOGY: AN INTERNATIONAL REVIEW, 2018, 67 (3), 523–556 doi: 10.1111/apps.12128
outcomes (e.g. Judge, Heller, & Mount, 2002; Judge, Thoresen, Bono, & Patton, 2001), while the majority of organisations regularly survey their employees with regard to satisfaction at work (e.g. Wiley, 2010). In general, these studies implicitly or explicitly assume that job satisfaction is a single con- struct lending itself to unidimensional quantification; very little research focuses on possible variation in the quality of job satisfaction. Yet the idea of differences in the quality of individuals� satisfaction with their working envi- ronment is not only intuitively attractive, it is also indirectly bolstered by find- ings from standard job satisfaction research. For instance, the fifth European Working Conditions Survey revealed that, on average, over 80 per cent of employees in Europe reported being “very satisfied” or “satisfied” (Eurofound, 2012). At the same time, more than one out of ten of these (very) satisfied employees indicated that they would not want to do the same job at the age of 60 (Eurofound, 2012). While this finding may echo changing expect- ations about lifetime career trajectories (e.g. Sullivan & Arthur, 2006) or job security in general (e.g. Hellgren, Sverke, & Isaksson, 1999), it also implies that job satisfaction can mean different things to different people. Some employees seem to be satisfied with little wish for change, while others predict a period in the future where their current jobs will no longer fulfil their needs (cf. Brown, Charlwood, & Spencer, 2012). In other words, while employees may evaluate their job satisfaction similarly on a one-dimensional (quantitative) scale, the quality of their satisfaction may differ considerably. The following study set out to explore this idea on the basis of Agnes Bruggemann�s (1974, 1976) theory of qualitative job (dis)satisfaction types.
Quantity versus Quality in Organisational Research
The question of quantitative versus qualitative construct distinctions (not to be confused with qualitative versus quantitative data per se) is one which has been particularly fruitful for organisational psychology research in recent years. One prime example of a central organisational construct whose redefini- tion in qualitative terms has stimulated relevant research is that of job insecur- ity. The recognition that job insecurity can refer not only to the likelihood of keeping a position but also to lack of stability in important job characteristics (Greenhalgh & Rosenblatt, 1984) has led researchers to explore differing effects of these “quantitative” versus “qualitative” types of insecurity (Hellgren et al., 1999). Empirical evidence suggests that both types of job insecurity are independent and equally important predictors for a variety of job outcomes (De Witte, De Cuyper, Handaja, Sverke, N€adwall, & Hellgren, 2010), possibly with differential importance for health-related versus attitudinal criteria (Hellgren et al., 1999). Similar shifts from quantitative to qualitative conceptu- alisation have also occurred on the outcome level itself, for instance in the case of employee turnover, organisational commitment, or motivation. Employee
524 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
turnover, measured as a dichotomous outcome variable (leave/stay), has long been a relevant organisational outcome (e.g. Mobley, Griffeth, Hand, & Meglino, 1979). Yet recent approaches have placed an emphasis on type of (non)turnover, noting that “reluctant stayers” or “reluctant leavers” may differ in fundamental dimensions from “enthusiastic” stayers and leavers (Hom, Mitchell, Lee, & Griffeth, 2012). This allows researchers to generate more spe- cific and informative hypotheses about predictor-outcome relationships by taking turnover type into consideration. In the case of organisational commit- ment, the qualitative distinction between affective, normative, and continuance commitment has also become well established, and has already shown itself to be useful both in predicting further outcomes and in providing more nuanced criteria (Meyer, Stanley, Herscovitch, & Topolnytsky, 2002). Similarly, since the introduction of self-determination theory, researchers typically consider qualitatively different types when studying work motivation (e.g. intrinsic, introjected, extrinsic, etc.; Gagn�e & Deci, 2005; Ryan & Deci, 2000).
The question of a qualitative versus quantitative focus in construct defini- tion is also strongly linked—though not synonymous—with the recently revived debate of variable-centred versus person-centred approaches to (quan- titative) data analysis (e.g. Meyer, Stanley, & Vandenberg, 2013; Morin, Morizot, Boudrias, & Madore, 2011; Wang & Hanges, 2011). The roots of this debate can be seen in fundamental discussions of ipsative versus normative analysis (i.e. approaches oriented towards within-subject versus between- subject comparisons; Boverman, 1962), but it has received greater attention and immediate practical relevance through the ease with which appropriate analyses can now be performed using modern software (Meyer et al., 2013). While the variable-centred approach focuses on relationships between variables across the whole sample, the person-centred approach is more interested in identifying subgroups for which those relationships might vary. One advantage of the person-centred approach is that it can identify combinations of variables that are theoretically or empirically relevant in predicting outcomes without recourse to a full variable interaction model. This increases statistical power of analyses while also facilitating interpretation of results and encouraging a more holistic view of variable relationships (Meyer et al., 2013). Though some of the qualitatively focused organisational psychology distinctions mentioned above (i.e. types of job insecurity or organisational commitment) are variable- centred, others—like attempts to define employee turnover types—are clearly person-centred. Some concepts, like the motivation types defined through self- determination theory, can be approached from either a variable-centred or a person-centred perspective. In fact, the boundaries between variable-centred and person-centred approaches are flexible. For instance, the organisational commitment literature has recently elaborated on the affective/normative/con- tinuance distinction by identifying subgroups of individuals showing different profiles of these commitment types (Meyer et al., 2013).
FORMS OF JOB SATISFACTION 525
VC 2017 International Association of Applied Psychology.
Given these trends, it is surprising that there have been so few parallel devel- opments in the field of job satisfaction. In fact, Bruggemann (1974, 1976) pro- posed a process model of job satisfaction including six qualitatively different forms of (dis)satisfaction more than 40 years ago. To date, however, her model has been rarely considered in international research. Yet taking such qualita- tive differences in job satisfaction into account might help explain ostensibly contradictory findings based on one-dimensional job satisfaction measures. Using a person-centred approach could reveal relevant interactions and encourage a more nuanced, holistic view of job satisfaction. Expanding the definition of job satisfaction to include qualitative as well as quantitative differ- ences may also help improve prediction of criterion variables such as well- being, attitudes, or performance (e.g. Judge et al., 2001).
Bruggemann’s Process Model of Job Satisfaction
Bruggemann�s basic assumption is that a person�s job satisfaction or dissatis- faction is not static but the dynamic result of a continual adjustment process (cf. Figure 1): (1) This process is (re)initiated when individuals compare their current job situation with their personal aspirations. A fit between the per- ceived and the desired state leads to stabilising satisfaction, a misfit to indis- tinct dissatisfaction. (2) Individuals then either maintain or change their level
CONSTRUCTIVE
job dissa�sfac�on
FIXATED
job dissa�sfac�on
STABILISED
job sa�sfac�on
PROGRESSIVE
job sa�sfac�on
RESIGNED
job sa�sfac�on
Stabilising
sa�sfac�on
Difference between
actual and ideal
work situa�on
Indis�nct
dissa�sfac�on
Increase
level of
aspira�on
Maintain
level of
aspira�on
Decrease
level of
aspira�on
Maintain
level of
aspira�on
A�empt to
solve problem
Make no a�empt
to solve problem
PSEUDO
job sa�sfac�on
(not opera�onalised)
Distort
percep�on
of situa�on
FIGURE 1. Forms of job satisfaction (translated/adapted from Bruggemann, Groskurth, & Ulich, 1975, pp. 134–135; B€ussing et al., 1999, p. 1002).
526 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
of aspiration. If they maintain their aspiration level, stabilising satisfaction evolves into stabilised satisfaction; if individuals increase their aspirations (e.g. seeking new possibilities or personal growth), stabilising satisfaction becomes progressive satisfaction. Indistinctly dissatisfied individuals can also lower their aspirations in order to achieve a form of resigned satisfaction. (3) Indistinctly dissatisfied individuals who maintain their level of aspiration can have different approaches to solving the misfit problem. If indistinctly dissatisfied employees distort or deny their perception of the actual situation, they may develop pseudo satisfaction. In contrast, if they actively try to change the situation, they will show constructive dissatisfaction. On the other hand, if employees do not make any problem-solving attempts, fixated dissatisfaction will emerge.
Overall, Bruggemann�s theory includes several elements which have been ela- borated in more modern theories. The initial process of comparison between the actual and ideal work situation bears great resemblance to person-environment (P-E) fit approaches, which have generated distinct fit concepts such as fit between personal needs and environmental satisfaction of those needs (i.e. needs- supply fit) or fit between environmental demands and personal abilities (i.e. demands-abilities fit; Edwards, 2008). Yet unlike most theories of P-E fit, Bruggemann defines the “person” variable in the equation as an individual�s ideal expectations about the environment. This encompasses both needs and values, as well as personal perceptions of the appropriateness of environmental demands. Thus, Bruggemann�s theory clearly begins with a P-E fit evaluation, but the type of fit is left open and only the direction—not the extent—of misfit is interpreted.
Similarly, Bruggemann�s model of how individuals respond to misfit has strong links to classic coping theories, including its distinction between problem-focused and perception or appraisal-based (i.e. emotion-focused) coping (Folkman & Lazarus, 1988; Lazarus, 1993). Again, however, Bruggemann�s theory does not fully elaborate the different coping mechanisms it has incorporated; in fact, some of the newer distinctions in the coping litera- ture (e.g. engagement vs. disengagement coping, accommodation coping, or proactive coping; Carver & Connor-Smith, 2010) had not been made at the time the theory came into existence.
In some ways this lack of elaboration is a disadvantage. For instance, current fit-theories suggest that different facets of the work situation or different types of misfit could moderate individuals� responses. Current coping theories have the potential to provide a more systematic and possibly more comprehensive way of classifying those responses. By not addressing these aspects, the Bruggemann model can be seen as incomplete. At the same time, its relative simplicity and focus on a fairly specific area of application is also an advantage. This reduction allows the model to take a broader perspective, establishing a theoretical link between P-E-fit and specific coping strategies in a plausible, parsimonious, and practically relevant way. Few of the more modern theories have even made this attempt.
FORMS OF JOB SATISFACTION 527
VC 2017 International Association of Applied Psychology.
One of the few attempts to elaborate on Bruggemann�s model on the basis of modern insights was made by B€ussing (1992), who proposed that controll- ability of the work situation would determine whether a person maintained, increased, or decreased his or her level of aspiration. In line with this assump- tion, Inauen, Jenny, and Bauer (2015) found that employees with progressive and stabilised satisfaction reported higher job control (and also lower effort- reward-imbalance) than employees experiencing resigned satisfaction. More- over, constructively dissatisfied employees experienced higher job control than employees with fixated dissatisfaction, which is also in line with the theory.
The dynamic model of job satisfaction has been praised for its practical relevance (e.g. Landy & Conte, 2010); however, its operationalisation has not been without criticism (e.g. Baumgartner & Udris, 2006). Several instruments have attempted to assess the postulated forms of (dis)satisfaction. Originally, Bruggemann (1976) herself developed the “Arbeitszufriedenheit- Kurzfragebogen” (AZK; Work Satisfaction Questionnaire—Short Form), which included one item per (dis)satisfaction form (excluding pseudo satisfac- tion, which, according to Bruggemann, cannot be measured directly due to the underlying implicit processes). Individuals were asked to choose the item that best describes their current (dis)satisfaction with their job. The AZK has been criticised for having somewhat ambiguous items and for not sufficiently capturing the cybernetic aspects of Bruggemann�s model (e.g. Baumgartner & Udris, 2006; Ferreira, 2009). However, some studies have also supported the validity of the scale on the basis of quantitative (Ziegler & Schlett, 2013) as well as qualitative analyses (B€ussing, Bissels, Fuchs, & Perrar, 1999).
Importantly, Bruggemann�s theory helps to explain the high percentage of satisfied employees found in most studies by suggesting that this percentage is actually made up of people experiencing very different forms of satisfaction; in addition to truly satisfied employees (i.e. individuals with stabilised or progres- sive satisfaction), this group also includes employees who are essentially dissat- isfied (i.e. have resigned or pseudo satisfaction). In fact, the few published empirical studies on the different forms of job (dis)satisfaction (e.g. Arnold & Mahler, 2010; Bruggemann, 1976; B€ussing, 1992; B€ussing et al., 1999; Inauen et al., 2015; Wegge & Neuhaus, 2002; Ziegler & Schlett, 2013) have suggested that, on average, about one-third of satisfied employees are actually experienc- ing resigned satisfaction. This result means that considering differences in type of satisfaction, not simply extent of overall satisfaction, is likely to add to the (prognostic) validity of measures of job satisfaction.
Forms of Job (Dis)Satisfaction and Outcomes
Relations between overall job satisfaction and various individual and organisa- tional outcome variables are already well established in the literature: Meta-analyses have revealed medium to strong correlations between job
528 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
satisfaction and general health (r 5 .37, Faragher, Cass, & Cooper, 2005), gen- eral well-being (r 5 .32–.48, Bowling, Eschleman, & Wang, 2010), and job- related well-being (r 5 .49 for positive affect and r 5 2.46 for negative affect, Judge & Ilies, 2004). Job satisfaction has also frequently been found to posi- tively relate to work engagement (r 5 .53, Christian, Garza, & Slaughter, 2011), organisational commitment (r 5 .53, Mathieu & Zajac, 1990; r 5 .65 for affec- tive and r 5 .31 for normative but r 5 2.07 for continuance commitment, Meyer et al., 2002; r 5 .452.62, Tett & Meyer, 1993), and, albeit to a less extent, performance (r 5 .30, Judge et al., 2001) and organisational citizenship behav- iour (OCB; r 5 .36, Lapierre & Hackett, 2007; r 5 .24, LePine, Erez, & Johnson, 2002; r 5 .44, Organ & Ryan, 1995). Furthermore, longitudinal studies have revealed that job satisfaction and general or job-related well-being are recipro- cally related over time (Bowling et al., 2010; Judge & Ilies, 2004) and that prior levels of job satisfaction are more likely to go along with subsequent changes in performance (r 5 .06) than the other way around (r 5 .00, Riketta, 2008).
Studies addressing different forms of job (dis)satisfaction have indicated that these individual and organisational outcomes also depend on the particu- lar (dis)satisfaction form employees are experiencing. Levels of well-being were found to be highest among stabilised satisfied employees, followed by progres- sively and then resignedly satisfied employees; well-being was lowest among employees with fixated dissatisfaction (Inauen et al., 2015; Wegge & Neuhaus, 2002; Ziegler & Schlett, 2013). A similar picture was obtained for overall job satisfaction. These findings are well in line with Bruggemann�s theoretical model: A person with stabilised satisfaction experiences a fit between the work situation and his or her aspirations, while a person with progressive or resigned satisfaction (or any form of dissatisfaction) experiences a misfit. Progressively satisfied employees may be satisfied, but they have increased their aspiration level, which leads to a new discrepancy between the situation and their aspira- tions. Thus, progressively satisfied employees experience a sort of “creative dis- satisfaction” (B€ussing, 1992), even as they continue holding a positive basic attitude towards their job. On the other hand, employees with resigned satisfac- tion are essentially dissatisfied. They have, however, recovered a kind of satis- faction through the process of disengagement or resignation (cf. Carver & Connor-Smith, 2010; Ziegler & Schlett, 2013). Employees with constructive or fixated dissatisfaction have not regained satisfaction; because the former are actively trying to improve the situation, however, they are more likely to reduce the misfit between the actual situation and their aspirations, thus increasing their chances of higher well-being and a stronger subjective sense of overall job satisfaction.
While employees with stabilised satisfaction seem to experience the highest levels of well-being (at least in the short term), progressively satisfied or even constructively dissatisfied employees are those who are especially likely to respond with increased motivation and should therefore show high values on
FORMS OF JOB SATISFACTION 529
VC 2017 International Association of Applied Psychology.
motivational constructs such as work engagement or OCB (cf. Crant, 2000; Parker, Bindl, & Strauss, 2010). Inauen and colleagues (2015) have provided some evidence for this assumption: In their study, work engagement was high- est among progressively satisfied employees, followed by engagement levels among employees with stabilised satisfaction and constructive dissatisfaction; employees with fixated dissatisfaction reported the lowest levels of work engagement. Moreover, in a study by Wegge and Neuhaus (2002), altruism (a dimension of OCB) and organisational identification were highest among employees experiencing progressive satisfaction, stabilised satisfaction, or con- structive dissatisfaction.
Finally, in another cross-sectional study, Arnold and Mahler (2010) investi- gated differences in organisational commitment and intention to quit among employees with different forms of job (dis)satisfaction. The authors hypothes- ised that, due to their “creative dissatisfaction”, progressively satisfied employ- ees would report lower organisational commitment and stronger intention to quit as compared to employees with stabilised satisfaction. The expected differ- ences were statistically significant with regard to turnover intentions but not for commitment. Moreover, commitment was lowest and intention to quit strongest among employees with fixated dissatisfaction, followed by construc- tively dissatisfied employees and employees with resigned satisfaction, though these differences were not statistically significant. These results demonstrate that the different forms of job satisfaction may be more relevant for some out- comes than for others. Though there are only a few studies on qualitatively dif- ferent forms of job (dis)satisfaction—and these are largely cross-sectional or descriptive in nature—as a whole, the results suggest that attitudes, experiences, and outcomes of similarly (dis)satisfied employees are likely to differ depend- ing on which specific form of satisfaction or dissatisfaction they experience.
Present Study
The present study aimed to contribute to job satisfaction research in several ways: First, we explored the extent to which Bruggemann�s theorised job satis- faction types could be empirically identified using mixture modelling. By applying modern person-centred analysis tools to reappraise the robustness of the theory, we hoped to stimulate interest and research in a qualitative concep- tualisation of job satisfaction. The results of this analysis formed the basis for a second step, in which we investigated whether the prediction of individual and organisational outcome variables could be improved when qualitative differen- ces in job satisfaction (i.e. the probabilities of belonging to latent classes) were considered in addition to quantitative differences in overall level of satisfac- tion. We considered three areas broadly relevant to individual and organisa- tional prosperity and chose two representative criterion variables for each area: (a) well-being, measured through general subjective well-being and job-related
530 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
affective well-being; (b) motivation, indicated by work engagement and organi- sational commitment; and (c) performance, consisting of task performance and the OCB dimension performance initiative (hereafter shortened to “OCB”). We expected levels of global job satisfaction to relate positively to these criterion variables (Hypothesis 1), and we expected the inclusion of infor- mation about different forms of job satisfaction to contribute additionally to their prediction (Hypothesis 2). Moreover, we collected data in two waves across a five-month interval in order to be able to examine not only immediate effects of job satisfaction but also effects over time. This approach also gave us the opportunity to explore individual stability and change in latent class mem- bership over time.
METHOD
Participants and Procedure
Data were collected at two time points separated by a five-month interval among 1,208 members of an online survey panel established by Respondi. The panel members were invited via email to take part in the survey and received bonus points (which could be swapped for products) at the first measurement point (T1) in return for their participation. The response rate at T1 after data cleaning was 73 per cent. The cross-sectional analyses are based on a sample of n 5 892 respondents (43.7% female) who provided full data at T1 and also reported working 5 hours a week or more (M 5 37.65, SD 5 9.97; median 5 40.00; maximum 5 80 hours). Their mean age was 41.44 years (SD 5 11.44; range 5 18–65). About half the participants held either a high school diploma (22.5%) or a university degree (32.4%). They were working in a variety of professional fields and companies, where roughly one third (32.0%) held a leadership position.
Of the participants whose responses were analysed at T1, 495 submitted their e-mail addresses to indicate that they were willing to participate in a follow-up survey (T2). The response rate at T2 after data cleaning was 40 per cent. Only individuals who worked at least 5 hours a week and provided full questionnaire data at both time points were included in the longitudinal analy- ses, yielding a sample of n 5 196 individuals (48.5% female). This longitudinal sample was slightly older in age (M 5 44.55 years, SD 5 10.83; range 5 18–64) but similarly educated in comparison with the T1 sample (20.9% holding a high school diploma and 30.1% holding a university degree).
Drop Out. Since attrition from T1 to T2 was considerable, we drew on the data gathered at T1 to analyse possible differences between individuals who participated fully and those who dropped out at T2. We used logistic regression analysis to predict dropout based on age, sex, level of education,
FORMS OF JOB SATISFACTION 531
VC 2017 International Association of Applied Psychology.
number of working hours, holding a leadership position, overall job satisfac- tion, level of each form of job (dis)satisfaction, and the six outcome variables (see Measures section for scale descriptions). Though the omnibus test for all predictors was significant, v2 (18) 5 60.5, p <.01, with a Cox & Snell pseudo- R2 of .07, only the odds ratios for age, task performance, and OCB differed sig- nificantly from 1 at a 5 .10. Full participants were older (44.55 vs. 41.44 years, OR 5 1.03) and reported higher task performance (6.27 vs. 5.84, OR 5 1.46) and higher task initiative (5.28 vs. 4.88, OR 5 1.19) than did individuals who failed to complete the survey at T2. Removing these three variables from the analysis resulted in a statistically non-significant omnibus test, v2 (15) 5 21.3, p 5 .13, a Cox & Snell pseudo-R2 of .02, and no odds ratios below .80 or above 1.13 for any other predictor.
Measures
Job Satisfaction. To assess participants� overall job satisfaction, we asked them to rate their agreement with the statement “I�m satisfied with my job” on a scale from completely disagree (1) to completely agree (7). Five additional statements assessed Bruggemann�s five types of job (dis)satisfaction (published by B€ussing et al., 1999, based on Bruggemann, 1976) using the same response scale: “I�m truly satisfied with my job, especially since I can really progress here” measuring progressive job satisfaction; “I�m truly satisfied with my job and for the near future I would like everything to remain as it is now” meas- uring stabilised job satisfaction; “I�m satisfied with my job—I always say it could be worse” measuring resigned job satisfaction; “Somehow I�m dissatis- fied with my job, but I don�t know what to do” measuring fixated job dissatis- faction; and “I�m dissatisfied with my job but I think that I can change something in the future” measuring constructive job dissatisfaction. Unlike Bruggemann�s (1976) original assessment, we did not force participants to choose the single statement that they agreed with most in order to determine their job (dis)satisfaction type. Instead, this classification was achieved through our statistical analyses based on ratings of all five Bruggemann statements (see Statistical Analyses and Results section).
Well-Being. Both general subjective psychological well-being and job- related affective well-being were measured. General well-being was assessed using the five-item WHO Well-Being Index (WHO-5; WHO Collaborating Centre in Mental Health, 1998). Participants rated how often they had felt dif- ferent aspects of subjective well-being (e.g. feeling “cheerful and in good spi- rits” or experiencing daily life “filled with things that interest me”) over the last two weeks by choosing one of the options at no time (1), some of the time (2), less than half of the time (3), more than half of the time (4), most of the time (5), or all of the time (6); Cronbach�s a 5 .90 at T1. We assessed affective well-being
532 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
using the 12 adjectives of the job-related affective well-being scale developed by Warr (1990), translated into German. Participants were asked to rate how often within the last 30 days their job had made them feel “relaxed”, “anxious”, “depressed”, “enthusiastic,” and so on, on a scale from never (1) to always (6); Cronbach�s a 5 .91 at T1.
Motivation. The current study assessed two forms of work-related moti- vation: work engagement and affective organisational commitment. Work engagement was measured using the nine-item German version of the Utrecht Work Engagement Scale (UWES-9; Schaufeli & Bakker, 2003). Participants were asked to rate how often they experienced engagement at work, described in statements such as “At my job, I feel strong and vigorous.” They indicated extent of engagement by choosing between the options never (1), almost never/ a few times a year or less (2), rarely/once a month or less (3), sometimes/a few times a month (4), often/once a week (5), very often/a few times a week (6), or always/every day (7); Cronbach�s a 5 .96 at T1. Commitment was measured using the eight-item Organizational Affective Commitment subscale of the Employee Commitment Survey developed by Allen and Meyer (1990) and translated by Schmidt, Hollmann, and Sodenkamp (1998; Schmidt and colleagues� use of the word “Betrieb”—i.e. firm/company—was replaced by the more general and accurate German translation “Organisation”). Partici- pants chose between the options disagree completely (1), disagree somewhat (2), neither disagree nor agree (3), agree somewhat (4), and agree completely (5) in response to items like “I really feel as if this organisation�s problems are my own” (Cronbach�s a 5 .78 at T1).
Performance. Performance was assessed using the two 5-item subscales “required work behaviour” (i.e. task performance) and “organisational citizen- ship behaviour: personal initiative” (i.e. OCB) from the FELA-S, a scale meas- uring self-reported achievement behaviour at work (Staufenbiel & Hartz, 2000). Participants were asked to rate statements like “I fulfil my work respon- sibilities appropriately” (task performance) or “I take a regular and active part in meetings at work” (OCB) on a scale consisting of the options disagree com- pletely (1), disagree (2), disagree somewhat (3) neither disagree nor agree (4) agree somewhat (5), agree (6), and agree completely (7); Cronbach�s a 5 .88 for task performance and a 5 .86 for OCB at T1.
Statistical Analyses
Forms of Job Satisfaction. Subgroups of respondents with distinct types of job (dis)satisfaction were identified using finite mixture modelling (see Hallquist & Wright, 2014; Morin et al., 2011). Following Raykov, Marcoulides, and Chang (2016), the number of latent classes were identified by modelling
FORMS OF JOB SATISFACTION 533
VC 2017 International Association of Applied Psychology.
the five Bruggemann job (dis)satisfaction items as latent factors and compar- ing the fit of different mixture models with one to nine classes. Across the dif- ferent classes, strict measurement invariance of the latent factors was enforced. Moreover, the factor variances and covariances were constrained across classes. In this way, the identification of distinct satisfaction types was based on the means of the five items (see Raykov et al., 2016). The number of subgroups (i.e. latent classes) was identified using multiple criteria including the Bayesian Information Criterion (BIC; Schwarz, 1978), measures of entropy, and adjusted likelihood ratio tests (Lo, Mendell, & Rubin, 2001). The BIC indicates better models that more closely approximate the empirical data at lower values, whereas the entropy reflects better classification accuracy when approaching 1. Moreover, the likelihood ratio tests compared each model with a given number of classes to the model with one less class; statistically significant results indi- cate more support for the model with more classes over the model with fewer classes. Finally, in line with prevalent recommendations (see Wright & Hallquist, 2014), the class sizes and class differentiability were used to guide decisions on the practical relevance of a given class solution. Longitudinal sta- bility of cluster membership was examined by extending the previous analyses to a latent transition model (Collins & Lanza, 2010) that estimated the transi- tion probabilities between different clusters at the two measurement occasions. Because only a small subsample (n 5 196) of respondents participated twice, we assumed strict measurement invariance over time and fixed the item param- eters at both time points to those estimated from the full sample at T1.
Prediction of Criterion Variables. In order to test our hypotheses that overall job satisfaction would positively predict our six outcome variables and that the Bruggemann-types of job satisfaction would add to this predictive power, we performed hierarchical linear regression analyses predicting each of the six outcome variables, first using overall satisfaction (Step 1) and then including latent class membership probabilities for five of the six clusters (Step 2) as predictors. Since class membership probabilities for individual partici- pants sum to one, it was not possible to include probabilities for all six clusters; this corresponds roughly to dummy-coding of a six-category variable except that uncertainty in cluster membership is explicitly modelled. Longitudinal analysis models were identical to the cross-sectional analyses except for the inclusion of the values of the outcome variable at T1 in Step 1 as a predictor of the outcome variable at T2. This means that longitudinal results can be inter- preted as predictions of change in the dependent variable over time. All regres- sion analyses were replicated using sex and age as control variables, but since this had minimal impact on the direction or magnitude of effects, we chose to report only the simpler regression model results.
In order to be able to compare coefficients of individual predictors within our regression models, we calculated and interpreted standardised regression
534 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
coefficients. Since conventional formulas for the standard error of standardised coefficients (e.g. Cohen, Cohen, & West, 2002) assume—inaccurately—that the predictor and criterion variances are known population values and not themselves subject to sampling variability (Yuan & Chan, 2011), we used the asymptotic distribution free method suggested by Jones and Waller (2015) and implemented in their R-package fungible (https://cran.r-project.org/web/pack- ages/fungible/index.html) to calculate confidence intervals for our standardised regression coefficients.
RESULTS
Forms of Job Satisfaction
Respondents with qualitatively different types of job (dis)satisfaction were identified by estimating different finite mixture models specifying between 1 and 9 classes (Table 1) on the basis of participants� level of agreement to the five Bruggemann job (dis)satisfaction statements (means and intercorrelations shown in Table 2). BIC decreased as model complexity rose; differences were small between the 6- and 7-cluster solution (DBIC 5 21.015) and negligible between the 8- and 9-cluster solution (DBIC 5 1.852), a finding echoed in the non-significant (p >.05) results of the Lo-Mendell-Rubin Likelihood Ratio Tests for both respective model comparisons. In order to preserve parsimony and because both solutions included five similar cluster profiles matching Bruggemann�s predictions, we selected a 6-cluster solution. Figure 2 shows the cluster profiles, including our labels derived from the Bruggemann terminol- ogy. We identified two theoretically predicted dissatisfaction profiles with high levels of constructive or fixated dissatisfaction, respectively, both of which
TABLE 1 Fit Indices for Different Mixture Models
Model logLik Number of parameters BIC Entropy LRT-p
Smallest class proportion
1 class 28,336.46 20 16,808.79 1.00 – – 2 classes 28,187.72 26 16,552.07 0.88 < .001 34% 3 classes 28,088.14 32 16,393.66 0.88 < .001 11% 4 classes 27,943.25 38 16,144.65 0.90 < .001 9% 5 classes 27,861.58 44 16,022.08 0.86 < .001 9% 6 classes 27,795.12 50 15,929.91 0.86 .023 8% 7 classes 27,764.23 56 15,908.89 0.87 .058 6% 8 classes 27,721.22 62 15,863.64 0.87 .002 6% 9 classes 27,681.75 68 15,825.45 0.89 .353 2%
Note: logLik 5 Logarithm of the model likelihood; BIC 5 Bayesian Information Criterion (Schwarz, 1978); LRT-p 5 p-value associated with the adjusted likelihood ratio test (Lo et al., 2001).
FORMS OF JOB SATISFACTION 535
VC 2017 International Association of Applied Psychology.
T A
B L E
2 D
e s c ri
p ti
v e
S ta
ti s ti
c s
fo r
M e tr
ic V
a ri
a b
le s
a t
T 1
(n 5
8 9 2 )
V ar
ia bl
e M
(S D
) JS
iP S
iS S
iR S
iC D
S iF
D S
G W
B JA
W W
E O
C T
A S
K
O ve
ra ll
jo b
sa ti
sf ac
ti o
n (J
S )
5. 05
(1 .6
9) It
em :P
ro gr
es si
ve sa
ti sf
ac ti
o n
(i P
S )
4. 07
(1 .9
9) .7
4 It
em :S
ta b
il is
ed sa
ti sf
ac ti
o n
(i S
S )
4. 35
(1 .9
9) .7
9 .7
8 It
em :R
es ig
n ed
sa ti
sf ac
ti o
n (i
R S
) 4.
39 (1
.8 0)
.2 0
.0 5
.1 4
It em
:C o
n st
ru ct
iv e
d is
sa ti
sf ac
ti o
n (i
C D
S )
2. 82
(1 .7
9) 2
.5 2
2 .4
5 2
.5 5
.0 1
It em
:F ix
at ed
d is
sa ti
sf ac
ti o
n (i
F D
S )
2. 85
(1 .8
7) 2
.5 9
2 .5
4 2
.5 7
.0 1
.4 9
G en
er al
w el
l- b
ei n
g (G
W B
) 3.
83 (1
.0 8)
.5 0
.5 0
.5 1
.0 6
2 .2
7 2
.4 4
Jo b
-r el
at ed
af fe
ct iv
e w
el l-
b ei
n g
(J A
W )
4. 06
(0 .9
2) .7
1 .6
4 .6
8 .1
2 2
.4 2
2 .6
2 .6
6 W
o rk
en ga
ge m
en t
(W E
) 4.
52 (1
.3 6)
.6 9
.6 5
.6 3
.1 3
2 .3
1 2
.4 9
.5 1
.6 5
O rg
an is
at io
n al
co m
m it
m en
t (O
C )
3. 26
(0 .7
4) .6
0 .5
8 .5
6 .0
8 2
.4 0
2 .4
3 .3
2 .4
5 .5
4 T
as k
p er
fo rm
an ce
(T A
S K
) 5.
94 (0
.9 7)
.3 6
.2 0
.2 6
.0 9
2 .2
8 2
.2 8
.1 5
.3 4
.3 0
.2 6
O rg
an is
at io
n al
ci ti
ze n
sh ip
b eh
av io
u r
(O C
B )
4. 97
(1 .2
8) .4
0 .4
1 .3
1 .0
4 2
.1 3
2 .2
8 .2
4 .3
4 .4
7 .4
7 .4
7
N ot
e: C
o ef
fi ci
en ts
w h
o se
ab so
lu te
va lu
e ex
ce ed
s .0
7 ar
e st
at is
ti ca
ll y
si gn
if ic
an t
at a
5 .0
5; va
lu es
ab ov
e .0
9 ar
e si
gn if
ic an
t at
a 5
.0 1.
536 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
showed moderate levels of resigned satisfaction and low levels of progressive and stabilised satisfaction. A third “indistinct” dissatisfaction profile emerged, showing moderate levels of progressive and stabilised satisfaction, fairly high levels of resigned satisfaction, and very high levels of both fixated and con- structive dissatisfaction. There was no indication of separate stabilised and progressive satisfaction profiles, but one profile showed high values for both scores and otherwise low dissatisfaction scores (stabilising satisfaction profile). The fifth cluster showed similar satisfaction and dissatisfaction values, with the exception of resigned satisfaction, which was high as opposed to low—we labelled this profile “resigned satisfaction”. The final cluster showed moderate values for all satisfaction types with a slight peak at resigned satisfaction, lead- ing us to label the cluster “ambivalent/indifferent”. Cluster membership was fairly evenly spread, though the resigned satisfaction (36%) and ambivalent/ indifferent (21%) clusters were more densely populated than the other four clusters (8% to 13% of the sample; see Figure 2 for frequencies).
Stability and Change in Cluster Membership. Stability of cluster mem- bership for individual participants was determined by fitting a latent transition model to the subsample of respondents that participated at both measurement occasions. The overall distribution of the cluster types at T2 stayed roughly comparable to the full-sample T1 frequencies for most clusters (change�3%), though there was a slight rise in proportion of resigned satisfaction (36% vs. 42%) and a decrease in the ambivalent/indifferent cluster (21% vs. 15%) from T1 to T2. Observation of the proportional distribution only among the longi- tudinal sample (n 5 196) at T1, however, showed that these changes were partly due to selective attrition: differences between T1 and T2 disappeared for resigned satisfaction (42% vs. 43%) and diminished in the case of the
1
2
3
4
5
6
7
Stabilising JS (n=112)
Resigned JS (n=320)
Construc�ve JDS (n=73)
Indis�nct JDS (n=95)
Fixated JDS (n=102)
Ambivalent/ indifferent
(n=190)
M ea
n Ra
� ng
(9 5%
C I)
Item: Progressive sa�sfac�on (iPS)
Item: Stabilised sa�sfac�on (iSS)
Item: Resigned sa�sfac�on (iRS)
Item: Construc�ve dissa�sfac�on (iCDS)
Item: Fixated dissa�sfac�on (iFDS)
FIGURE 2. Bruggemann item profiles for the six latent job satisfaction (JS) or dissatisfaction (JDS) clusters.
FORMS OF JOB SATISFACTION 537
VC 2017 International Association of Applied Psychology.
ambivalent/indifferent cluster (17% vs. 15%; all other differences�3%). Table 3 shows the transition probabilities for the six latent clusters from T1 to T2. The stabilising and resigned job satisfaction clusters showed the highest levels of stability (> 74% probability of staying in the same cluster), followed by fixated and indistinct job dissatisfaction (> 40%). Constructive job dissatisfaction and ambivalent/indifferent cluster were less stable, with transition probabilities all less than 35 per cent.
Prediction of Criterion Variables
Descriptive Statistics. Means, standard deviations and intercorrelations of the affective, motivational, and performance-related outcome variables can be found in Table 2. Additionally, Table 4 reports descriptive outcome statistics for employees with different job (dis)satisfaction types for the whole sample at T1 (similar values were observed at T2). All criterion variables reached their highest values among the group of employees with stabilising satisfaction and their second-highest values among the group experiencing resigned satisfac- tion. Differences were less marked between the fixated, constructive, and indis- tinct dissatisfaction clusters.
Well-Being. Overall job satisfaction showed positive relationships with job-related affective well-being cross-sectionally (b 5 .71) and across time after controlling for initial values (i.e. change prediction; b 5 .30). This pattern was echoed when predicting general well-being (b 5 .50 and .20, respectively). Including estimates of the Bruggemann-satisfaction types improved prediction of well-being cross-sectionally by 4–5 per cent explained variance, with stabilis- ing satisfaction, resigned satisfaction, and fixated dissatisfaction being the strongest predictors. Explanatory power of the individual types across time
TABLE 3 Latent Transition Probabilities for Different Latent Types of Job Satisfaction (JS)
or Job Dissatisfaction (JDS) from T1 (Rows) to T2 (Columns)
Latent status Stabilising
JS Resigned
JS Constructive
JDS Indistinct
JDS Fixated
JDS Ambivalent/ indifferent
Stabilising JS .81 .15 .00 .03 .00 .00 Resigned JS .04 .75 .06 .00 .03 .13 Constructive JDS .11 .17 .30 .23 .00 .21 Indistinct JDS .00 .00 .05 .41 .27 .28 Fixated JDS .00 .21 .16 .06 .57 .00 Ambivalent/indifferent .14 .35 .08 .11 .11 .21
Note: Item parameters were fixed to the values calculated in the full T1 sample (n 5 892) and applied to analyses of the longitudinal sample (n 5 196).
538 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
T A
B L E
4 M
e a n
s (S
ta n
d a rd
D e v ia
ti o
n s )
o f
th e
C ri
te ri
o n
V a ri
a b
le s
fo r
E m
p lo
y e e s
w it
h D
if fe
re n
t L a te
n t
T y p
e s
o f
J o
b S
a ti
s fa
c ti
o n
(J S
) o
r J o
b D
is s a ti
s fa
c ti
o n
(J D
S )
a t
T 1
(n 5
8 9 2 )
T yp
e G
en er
al jo
b sa
ti sf
ac ti
on G
en er
al w
el l-
be in
g
Jo b-
re la
te d
af fe
ct iv
e w
el l-
be in
g W
or k
en ga
ge m
en t
O rg
an is
at io
na l
co m
m it
m en
t T
as k
P er
fo rm
an ce
O C
B
S ta
b il
is in
g JS
6. 49
(0 .9
8) 4.
54 (0
.7 8)
4. 84
(0 .5
1) 5.
52 (0
.9 4)
3. 83
(0 .6
2) 6.
42 (0
.7 5)
5. 69
(1 .2
5) R
es ig
n ed
JS 5.
99 (1
.1 6)
4. 18
(0 .9
6) 4.
51 (0
.7 1)
5. 04
(1 .1
8) 3.
56 (0
.6 8)
6. 19
(0 .8
3) 5.
18 (1
.2 4)
C o
n st
ru ct
iv e
JD S
4. 05
(1 .8
2) 3.
66 (0
.9 1)
3. 87
(0 .8
7) 4.
20 (1
.3 7)
2. 93
(0 .8
4) 6.
04 (1
.0 4)
4. 87
(1 .4
1) In
d is
ti n
ct JD
S 3.
86 (1
.5 5)
3. 51
(1 .1
2) 3.
48 (0
.7 8)
4. 08
(1 .2
7) 2.
90 (0
.6 2)
5. 53
(0 .9
4) 4.
87 (1
.1 1)
F ix
at ed
JD S
3. 65
(1 .6
0) 2.
94 (1
.0 9)
3. 13
(1 .0
0) 3.
3 (1
.2 9)
2. 87
(0 .6
6) 5.
88 (0
.9 2)
4. 22
(1 .3
6) A
m b
iv al
en t/
in d
if fe
re n
t 4.
34 (1
.2 3)
3. 51
(0 .9
3) 3.
72 (0
.6 1)
4. 02
(1 .0
9) 2.
96 (0
.5 1)
5. 42
(1 .0
2) 4.
66 (1
.0 6)
FORMS OF JOB SATISFACTION 539
VC 2017 International Association of Applied Psychology.
was modest (1% explained variance, not statistically significant), with no indi- vidual coefficient differing reliably from zero (see Table 5 for full results).
Motivation. Overall job satisfaction was positively related to work engagement cross-sectionally (b 5 .69) and across time after controlling for ini- tial values (change prediction; b 5 .21). The cross-sectional relationship with affective organisational commitment was comparable (b 5 .60) but disappeared almost completely across time (b 5 .06, n.s.; see Table 6). Bruggemann-satisfaction types improved prediction of the motivational out- comes cross-sectionally by 2–3 per cent explained variance while there was no significant improvement in change prediction. Stabilising satisfaction was par- ticularly effective in explaining both work engagement and commitment cross- sectionally (b� .11), while fixated dissatisfaction negatively predicted work engagement (b 5 2.08) but showed no reliable relationship with organisational commitment. Conversely, resigned satisfaction seemed to positively predict organisational commitment (b 5 .15) while not relating significantly to work engagement.
Performance. The relationship between overall job satisfaction and the two performance indicators were positive but smaller in size than relationships with well-being and motivational indicators, with both coefficients sinking below statistical significance in the longitudinal analysis (see Table 7). Adding the Bruggemann-satisfaction types improved the cross-sectional prediction of performance by 3–6 per cent explained variance. Longitudinally, prediction of task performance (but not OCB) was improved by 4 per cent explained var- iance. Four Bruggemann predictors showed a positive relationship with task performance cross-sectionally (.21�b� .28), with only the predictive power of indistinct dissatisfaction failing to reach statistical significance (b 5 .07). Cross-sectional relationships of Bruggemann-types with OCB were more mod- est, with stabilising satisfaction and indistinct dissatisfaction being the strong- est predictors (b 5 .13 and b 5 .11, respectively). Despite the overall increase in explained variance of task performance in the longitudinal analysis, no individ- ual satisfaction predictor (including overall satisfaction) reached statistical sig- nificance in predicting change over time.
DISCUSSION
Forms of Job Satisfaction
In order to replicate Bruggemann�s (1976) classification of employees into one of five forms of job (dis)satisfaction, we asked participants to agree with state- ments reflecting these forms on a seven-point Likert scale. In contrast to the original forced-choice format (choosing the one statement that best matched
540 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
T A
B L E
5 R
e s u
lt s
o f
H ie
ra rc
h ic
a l L in
e a r
R e g
re s s io
n M
o d
e ls
P re
d ic
ti n
g S
u b
je c ti
v e
W e ll -B
e in
g C
ri te
ri a
(S ta
n d
a rd
is e d
C o
e ffi
c ie
n ts
w it
h 9 5 %
C o
n fi
d e n
c e
In te
rv a ls
in B
ra c k e ts
)
G en
er al
w el
l- be
in g
Jo b-
re la
te d
af fe
ct iv
e w
el l-
be in
g
P re
di ct
or s
T 1
T 2
T 1
T 2
S te
p 1
D ep
en d
en t
va ri
ab le
T 1
– .6
0 [.
47 ,
.7 4]
– .5
1 [.
35 ,
.6 8]
O ve
ra ll
jo b
sa ti
sf ac
ti o
n .5
0 [.
44 ,.
56 ]
.2 0
[. 07
, .3
2] .7
1 [.
68 ,.
75 ]
.3 0
[. 14
, .4
6] D
R 2
.2 5
.5 2
.5 1
.5 6
S te
p 2
D ep
en d
en t
va ri
ab le
T 1
– .6
0 [.
45 ,
.7 4]
– .4
7 [.
30 ,
.6 4]
O ve
ra ll
jo b
sa ti
sf ac
ti o
n .3
7 [.
28 ,.
46 ]
.1 8
[. 05
, .3
2] .5
5 [.
48 ,.
61 ]
.2 4
[. 08
, .4
1] S
ta b
il is
in g
sa ti
sf ac
ti o
n a
.1 6
[. 09
,. 24
] .0
0 [2
.1 2,
.1 2]
.1 8
[. 11
,. 24
] .0
5 [2
.0 8,
.1 9]
R es
ig n
ed sa
ti sf
ac ti
o n
a .1
0 [.
01 ,.
19 ]
.0 1
[2 .1
5, .1
7] .1
5 [.
08 ,.
22 ]
.1 1
[2 .0
6, .2
7] C
o n
st ru
ct iv
e d
is sa
ti sf
ac ti
o n
a .0
5 [2
.0 2,
.1 1]
.0 5
[2 .0
8, .1
8] .0
7 [.
01 ,.
13 ]
.0 4
[2 .1
1, .1
9] In
d is
ti n
ct d
is sa
ti sf
ac ti
o n
a .0
1 [2
.0 6,
.0 9]
2 .0
5 [2
.1 6,
.0 6]
2 .0
5 [2
.1 0,
.0 1]
2 .0
1 [2
.1 3,
.1 1]
F ix
at ed
d is
sa ti
sf ac
ti o
n a
2 .1
3 [2
.2 0,
2 .0
6] .0
0 [2
.1 3,
.1 3]
2 .1
2 [2
.1 8,
2 .0
7] 2
.0 6
[2 .2
0, .0
9] D
R 2
.0 4
.0 1
n .s
. .0
5 .0
1 n
.s .
T ot
al R
2 .2
9 .5
2 .5
6 .5
8
a L
at en
t cl
u st
er m
em b
er sh
ip p
ro b
ab il
it ie
s.
FORMS OF JOB SATISFACTION 541
VC 2017 International Association of Applied Psychology.
T A
B L E
6 R
e s u
lt s
o f
H ie
ra rc
h ic
a l L in
e a r
R e g
re s s io
n M
o d
e ls
P re
d ic
ti n
g M
o ti
v a ti
o n
a l O
u tc
o m
e s
(S ta
n d
a rd
is e d
C o
e ffi
c ie
n ts
w it
h 9 5 %
C o
n fi
d e n
c e
In te
rv a ls
in B
ra c k e ts
)
W or
k en
ga ge
m en
t O
rg an
is at
io na
l co
m m
it m
en t
P re
di ct
or s
T 1
T 2
T 1
T 2
S te
p 1
D ep
en d
en t
va ri
ab le
T 1
– .6
1 [.
45 ,
.7 7]
– .7
5 [.
65 ,
.8 5]
O ve
ra ll
jo b
sa ti
sf ac
ti o
n .6
9 [.
64 ,.
73 ]
.2 1
[. 04
, .3
7] .6
0 [.
56 ,.
64 ]
.0 6
[2 .0
5, .1
7] D
R 2
.4 7
.5 8
.3 6
.6 3
S te
p 2
D ep
en d
en t
va ri
ab le
T 1
– .6
1 [.
45 ,
.7 7]
– .7
7 [.
66 ,
.8 7]
O ve
ra ll
jo b
sa ti
sf ac
ti o
n .6
2 [.
56 ,.
69 ]
.1 5
[2 .0
1, .3
2] .4
7 [.
41 ,.
54 ]
.0 0
[2 .1
3, .1
3] S
ta b
il is
in g
sa ti
sf ac
ti o
n a
.1 1
[. 04
,. 18
] .0
5 [2
.0 9,
.1 8]
.1 9
[. 12
,. 26
] .0
1 [2
.1 2,
.1 4]
R es
ig n
ed sa
ti sf
ac ti
o n
a .0
8 [.
00 ,.
15 ]
.0 1
[2 .1
5, .1
6] .1
5 [.
08 ,.
22 ]
.0 0
[2 .1
4, .1
4] C
o n
st ru
ct iv
e d
is sa
ti sf
ac ti
o n
a .0
7 [.
01 ,.
13 ]
2 .0
5 [2
.1 7,
.0 7]
.0 0
[2 .0
7, .0
6] .0
9 [2
.0 5,
.2 2]
In d
is ti
n ct
d is
sa ti
sf ac
ti o
n a
.0 8
[. 02
,. 14
] 2
.0 2
[2 .1
4, .0
9] .0
2 [2
.0 4,
.0 8]
2 .0
6 [2
.1 7,
.0 5]
F ix
at ed
d is
sa ti
sf ac
ti o
n a
2 .0
8 [2
.1 3,
-. 02
] 2
.0 5
[2 .1
8, .0
8] .0
2 [2
.0 4,
.0 8]
2 .0
9 [2
.2 0,
.0 1]
D R
2 .0
2 .0
1 n
.s .
.0 3
.0 2
n .s
. T
ot al
R 2
.5 0
.5 9
.3 9
.6 5
a L
at en
t cl
u st
er m
em b
er sh
ip p
ro b
ab il
it ie
s.
542 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
T A
B L E
7 R
e s u
lt s
o f
H ie
ra rc
h ic
a l L in
e a r
R e g
re s s io
n M
o d
e ls
P re
d ic
ti n
g S
e lf
-R e p
o rt
e d
P e rf
o rm
a n
c e
C ri
te ri
a (S
ta n
d a rd
is e d
C o
e ffi
c ie
n ts
w it
h 9 5 %
C o
n fi
d e n
c e
In te
rv a ls
in B
ra c k e ts
)
T as
k pe
rf or
m an
ce O
C B
P re
di ct
or s
T 1
T 2
T 1
T 2
S te
p 1
D ep
en d
en t
va ri
ab le
T 1
– .6
1 [.
48 ,.
74 ]
– .7
9 [.
71 ,
.8 6]
O ve
ra ll
jo b
sa ti
sf ac
ti o
n .3
6 [.
30 ,
.4 2]
.0 6
[2 .0
5, .1
7] .4
0 [.
34 ,.
47 ]
.0 9
[2 .0
1, .1
9] D
R 2
.1 3
.3 9
.1 6
.6 7
S te
p 2
D ep
en d
en t
va ri
ab le
T 1
– .5
8 [.
43 ,.
73 ]
– .7
9 [.
71 ,
.8 7]
O ve
ra ll
jo b
sa ti
sf ac
ti o
n .2
7 [.
17 ,
.3 7]
2 .0
9 [2
.2 6,
.0 7]
.3 7
[. 28
,. 45
] .0
5 [2
.0 6,
.1 6]
S ta
b il
is in
g sa
ti sf
ac ti
o n
a .2
3 [.
15 ,
.3 2]
.1 4
[2 .0
4, .3
2] .1
3 [.
05 ,.
21 ]
2 .0
6 [2
.1 7,
.0 4]
R es
ig n
ed sa
ti sf
ac ti
o n
a .2
8 [.
18 ,
.3 8]
.0 9
[2 .1
3, .3
1] .0
4 [2
.0 4,
.1 2]
2 .0
1 [2
.1 3,
.1 2]
C o
n st
ru ct
iv e
d is
sa ti
sf ac
ti o
n a
.2 1
[. 12
, .2
9] 2
.1 2
[2 .2
4, .0
0] .0
8 [.
01 ,.
15 ]
2 .0
6 [2
.1 3,
.0 1]
In d
is ti
n ct
d is
sa ti
sf ac
ti o
n a
.0 7
[2 .0
1, .1
5] 2
.0 7
[2 .2
0, .0
6] .1
1 [.
04 ,.
17 ]
2 .0
8 [2
.1 7,
.0 2]
F ix
at ed
d is
sa ti
sf ac
ti o
n a
.2 1
[. 14
, .2
9] 2
.0 7
[2 .2
4, .1
0] 2
.0 4
[2 .1
2, .0
4] 2
.0 9
[2 .1
9, .0
1] D
R 2
.0 6
.0 4
.0 3
.0 1
n .s
. T
ot al
R 2
.1 9
.4 3
.1 9
.6 8
a L
at en
t cl
u st
er m
em b
er sh
ip p
ro b
ab il
it ie
s.
FORMS OF JOB SATISFACTION 543
VC 2017 International Association of Applied Psychology.
participants� self-assessment), the Likert-scaled measure allowed participants to agree with multiple statements equally and thus made it possible to search for the postulated satisfaction forms using mixture modelling. This approach produced more nuanced results than the original typology. First, we were unable to corroborate Bruggemann�s assumption that highly satisfied employ- ees fall clearly into a “stabilised” versus “progressive” form of satisfaction; instead, results showed only one group with “stabilising satisfaction” (i.e. high values for both progressive and stabilised satisfaction). This group could, how- ever, be distinguished from a “resigned satisfaction” cluster, whose profile was quite similar but included strong agreement with the resigned satisfaction item. There was also evidence for both postulated dissatisfaction forms, as well as for a third “indistinct dissatisfaction” group. Finally, we identified an “ambivalent/indifferent” cluster with roughly equal values on all Bruggemann items.
It is difficult to say whether the lack of separate stabilised and progressive sat- isfaction profiles should be interpreted substantively or methodologically. Per- haps the formulation of the progressive satisfaction item failed to capture the element of active future change needed to distinguish it from the more passive form of stabilised satisfaction. On the other hand, it is also conceivable that pro- gressive satisfaction develops only in the presence of a general desire to main- tain the current positive situation. In fact, an individual satisfied with the current opportunities for positive change offered by his or her job might be expected to hope that this current situation—in other words the existence of the opportunity for positive change—will be maintained. Thus, the conceptual dis- tinction between progressive and stabilised satisfaction may be less than clear. This argument can be extended by considering that aspiring to positive change requires substantial investment of personal resources (cf. Hobfoll, 1989; or Fay & H€uttges, 2016). Thus, it may be unrealistic to assume that employees con- stantly maintain a state of progressive satisfaction. Perhaps instead, progressive satisfaction is characterised by a more circular process including periods of increased motivation for positive change followed by periods of stabilised satis- faction (i.e. phases of high engagement interspersed by a healthy form of disen- gagement coping). At any given time, there may be few individuals currently experiencing an intense progressive phase, making them difficult to identify. Over longer periods of time and multiple measurements, however, a distinction between employees who experience such progressive phases and those who do not may be possible. Even without assuming a circular stabilised-progressive- stabilised process, however, our current classification may have overlooked progressive versus stabilised trends in aspirational trajectories. Perhaps a dis- tinction between these two clusters could have been made on the basis of more detailed process data, for instance gathered via diary study.
Similar arguments apply to the new “Ambivalent/indifferent” cluster we identified. Perhaps consideration of the preceding aspirational trajectories
544 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
would have suggested interpretable differences within this group or revealed similarities to one or more of the other latent clusters. However, given that atti- tudinal ambivalence (e.g. Conner & Armitage, 2008) is well-established in the psychological literature and may in fact moderate the relationship between global job satisfaction and performance (Ziegler, Hagen, & Diehl, 2012; Ziegler, Schlett, Casel, & Diehl, 2012), an extension of the Bruggemann Model by this category is conceptually plausible. If comparisons between the actual and ideal job situation lead to vastly differing evaluations depending on which facet of work is considered, then individuals may not fall into one single job satisfaction cluster; they may show simultaneous but contradictory attitudes toward work. Such ambivalent individuals are difficult to distinguish from par- ticipants who answered carelessly or showed a tendency to agree with all items (e.g. Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). Overall, though, it seems reasonable that not all individuals will show a clear pattern of job (dis)satisfac- tion, and that predictions based on the Bruggemann types may be more precise if these individuals are considered separately. One interesting avenue of future research might be to characterise this new cluster on the basis of scales specifi- cally designed to measure job ambivalence (e.g. Ziegler, Hagen, et al., 2012) or statistical methods designed to identify biased responding (e.g. Rauch, Schweizer, & Moosbrugger, 2007).
A further interesting aspect of the latent cluster results was the large number of individuals with resigned job satisfaction (over 35% of the sample). This proportion is consistent with the 25–45 per cent resigned satisfaction found originally by Bruggemann (1976, p. 72), though it is somewhat higher than the roughly 20 per cent identified in later studies (e.g. B€ussing et al., 1999, p. 1018; Ziegler & Schlett, 2013, p. 68). While the number of resigned employees may truly be quite high, the wording of the resignation item (“it could be worse”) may actually conflate two different types of resignation. On the one hand, agreeing that one�s job “could be worse” may, as Bruggemann argued, be the result of an adaptive coping mechanism aimed at reducing dissonance between one�s aspirations and one�s experience of reality. This may be a healthy way to deal with professional difficulties; it constitutes a positive reframing of poten- tially negative situations that may protect individuals from disappointment or even result in more realistic job expectations. On the other hand, agreement with this statement could also indicate emotional disengagement. Individuals may agree that things “could be worse” because of the relative flippancy of the phrase and its implication that there are more devastating things in life than a mere job. Such emotional disengagement could be a healthy mechanism in principle, but it is likely to be negatively related to investment-based job out- comes (e.g. work engagement, commitment, OCB) and at extreme values it overlaps with the concepts of emotional exhaustion and cynicism characteristic of burnout (e.g. Schaufeli & Bakker, 2004). Thus, the consequences of these two possible sources of resigned satisfaction may be quite different. If these
FORMS OF JOB SATISFACTION 545
VC 2017 International Association of Applied Psychology.
effects really are opposite in direction, then separating these groups could reveal stronger relationships between each type of resigned satisfaction and job outcomes. More detailed exploration of the resigned satisfaction cluster— including questions about the relative roles of cognitive reappraisal and emo- tional disengagement in the “resignation” process—are needed to detangle this possible confound.
Despite such variation from the (dis)satisfaction types postulated by Bruggemann, our outcome patterns did conform to the model�s predictions in interesting ways. Specifically, employees with fixated job dissatisfaction showed the lowest values in well-being and motivation, followed by employees with indistinct dissatisfaction and with constructive dissatisfaction. Resigned satisfaction was associated with slightly lower outcome levels than stabilising satisfaction. One unexpected pattern was the lack of variation in task perform- ance among the job satisfaction types. Since task performance can be under- stood as meeting minimum job requirements, however, this result is not unreasonable: fixated job dissatisfaction does not seem to imply that employ- ees shirk their work, only that they experience lower motivation and well-being while doing it.
In addition to postulating different forms of satisfaction, Bruggemann�s model is theoretically interesting because it assumes that job satisfaction arises through (mis)fit between the given job situation and personal aspirations. This, in turn, implies that type of job satisfaction is not a stable personality trait but a state that may fluctuate or—perhaps—develop along predictable pathways under certain circumstances. Our data suggest that some forms of employees� job (dis)satisfaction are substantially more stable than others. Employees expe- riencing stabilising or resigned satisfaction were extremely likely to remain in their respective clusters after five months. The lower extreme of the satisfaction spectrum, fixated dissatisfaction, was also fairly stable, though individuals had a fair chance of moving from fixated to resigned satisfaction. Movement from the ambivalent/indifferent and the constructive job dissatisfaction clusters was most diffuse: transition probabilities were pretty evenly spread across all the categories. This instability of the ambivalent/indifferent cluster is consistent with research showing that attitude change is more likely when attitudes are ambivalent (Luttrell, Petty, & Bri~nol, 2016; Maio, Bell, & Esses, 1996). If con- flicting attitudes are already present, it seems reasonable that a variety of situa- tional and personal factors might succeed in changing this type of job satisfaction in a variety of directions over the course of five months. In con- trast, it seems likely that the instability of constructive dissatisfaction could be driven by whether or not employees have found that they could actually “change something in the future”. Thus, their movement to more satisfied clus- ters could indicate that their working situation has improved relative to their aspirations, while movement to the indistinct dissatisfaction cluster might indi- cate lack of improvement and a receding hope that they will be able to change
546 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
their situation for the better. Interpreting dissatisfaction from a psychological contract perspective, this argument is in line with Tomprou, Rousseou, and Hansen�s (2015) psychological contract post-violation model, which states that, in addition to self-based resources, organisational resources and organisational responsiveness help determine how employees respond to contract violations. Overall, the results suggest that constructive dissatisfaction might be a cross- roads where employees are particularly receptive to positive (or negative) job interventions. Early identification of these individuals may give organisations a chance to respond to constructive employee criticism in a way that increases sat- isfaction instead of allowing constructive dissatisfaction to slip into more stable and detrimental forms of dissatisfaction like indistinct job dissatisfaction.
Predicting Well-Being, Motivation, and Performance
Overall Job Satisfaction. As expected, overall job satisfaction was a sub- stantial predictor for all the outcome variables in the cross-sectional analyses (explained variances�13%). Its cross-sectional relationship was strongest, accounting for about half of the variance in job-related affective well-being and in work engagement. Overall satisfaction was less effective in predicting changes in the outcomes over time, but its effect was still positive and substan- tial for both forms of well-being, as well as for work engagement. Thus, our first hypothesis was supported fully in the cross-sectional and partly in the lon- gitudinal analysis. One explanation for the decreased explanatory power of job satisfaction for organisational commitment and OCB might be that these scales capture aspects of high-activation, volitional behaviour (i.e. attending meetings, making suggestions for improvement) absent in a general assessment of (comparatively passive) job satisfaction. While the work engagement scale also included items relating to energy and enthusiasm, work engagement�s close ties to the concepts of burnout and well-being might help explain why its relationship to overall job satisfaction more closely echoed those of the explicit well-being scales. In contrast, the lack of long-term relationship between job satisfaction and task performance could be explained by the fact that this scale captures the “minimal” aspects of job performance—that is, fulfilling (only) one�s duties. While dissatisfaction might certainly lead to less conscientious performance of duties, a very substantial level of dissatisfaction must be pres- ent before one would expect an employee to fully neglect his or her contractual duties (this argument could be seen as predicting a ceiling effect for task per- formance; in fact, the mean value was quite high for all groups, though we did find some variation in the measure).
Bruggemann Forms of (Dis)Satisfaction. Despite the sometimes sub- stantial predictive power of overall job satisfaction, adding the latent cluster membership probabilities of the Bruggemann (dis)satisfaction types had a
FORMS OF JOB SATISFACTION 547
VC 2017 International Association of Applied Psychology.
small but consistent positive effect (2–6% explained variance) on the regression models� ability to predict the six work- and health-related outcomes cross- sectionally. Prediction of changes in outcomes over the span of five months was more modest, with only the explained variance for change in task perform- ance reaching statistical significance (4% explained variance); in this case, how- ever, no individual predictor�s regression coefficient differed significantly from zero, making interpretation of the direction of change effects difficult. Thus, our second hypothesis was supported cross-sectionally, but the longitudinal change predictions failed to meet our expectations. Though this may indicate that the Bruggemann types are not particularly predictively useful over time, the relatively small T2-sample may have caused this analysis to be underpow- ered, high measurement error in the one-item scales may have decreased preci- sion of the cluster analysis, or the five-month time lag may be too short or long a timespan to register changes. Further research using multi-item measures and more frequent time sampling designs might be able to reveal stronger lon- gitudinal effects.
Particularly interesting in the cross-sectional analyses were the noticeable differences in predictive power between the different forms of (dis)satisfaction. While fixated dissatisfaction negatively predicted well-being and work engage- ment, constructive dissatisfaction showed positive or neutral relationships. This finding is in line with assumptions underlying research on proactivity and disengagement coping. Disengagement coping (e.g. fixated dissatisfaction) has been shown to be harmful for individuals� well-being (Carver & Connor-Smith, 2010). On the other hand, engagement coping or proactivity (e.g. constructive dissatisfaction) is a more effective way to reduce stress and thus constitutes a more positive long-term coping strategy (e.g. Cangiano & Parker, 2015; Carver & Connor-Smith, 2010). A quiet echo of this hierarchy can be seen in the descriptively higher loadings of the (more active) stabilising satisfaction in comparison with the (more passive) resigned satisfaction predic- tor. Theoretically, however, we would have expected to see this pattern in the longitudinal, not the cross-sectional loadings. Since proactive attempts con- sume resources (Hobfoll, 1989), they are likely to have some negative conse- quences in the short term (Cangiano & Parker, 2015). We found no evidence of this change in effect direction. Possibly our analysis was not sensitive enough to register small changes in the five-month observational period; however, the existing differences in cross-sectional coefficients suggest that some of the advantages of active coping strategies may have already been felt by the time employees were classified as constructively dissatisfied.
Limitations and Future Research
Despite the interesting relationships revealed by questioning participants more closely about their type of job (dis)satisfaction, the current study used a fairly
548 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
crude measure of satisfaction type. Bruggemann has often been criticised for the rough operationalisation of her constructs, and the current scale cannot escape these criticisms. Because each type of (dis)satisfaction was measured using only a single item, no estimate of internal consistency was possible; simi- larly, any attempt to test a five-factor structure on the basis of five items would have been tautological. Though single-item measures have been shown to have psychometrically acceptable properties in other contexts (Gogol et al., 2014) and specifically in the context of global job satisfaction (Wanous, Reichers, & Hudy, 1997), the reduction of each (dis)satisfaction type to one item doubtless also led to a reduction of measurement precision as well as a reduction of breadth of the original constructs (e.g. reducing progressive satisfaction to a desire to personally progress in one�s job). Nevertheless, our mixture modelling approach showed that employee responses to these items did roughly corre- spond to Bruggemann�s typology. Instead of focusing on whether the variables are independent factors across all participants (which Bruggemann never directly claimed), we thus explored the theory on its own terms: as a claim about qualitative differences in variable profiles among individual participants. Indeed, we found that participants with similar overall satisfaction levels could be classified into groups with quite different satisfaction profiles. Thus, we were able to contribute to the recent growth of person-centred research in organisational contexts (e.g. Bennett, Gabriel, Calderwood, Dahling, & Trougakos, 2016; Hom et al., 2012; Meyer et al., 2013; Morin et al., 2011; Wang & Hanges, 2011; Woo & Allen, 2014). Furthermore, by using cluster membership probabilities to predict metric outcomes, we were able to merge the more qualitative with a traditional quantitative approach, demonstrating the potential usefulness of a person-centred typology for variable-based predic- tive models. Though our six-cluster solution, regression analyses, and espe- cially our tentative findings regarding changes in cluster membership over time must be replicated on independent and larger samples in order to be fully trusted, we feel that this approach to corroborating and challenging Bruggemann�s theory is quite promising. Beyond replication with a larger sam- ple, also including multi-item measures of the individual Bruggemann types as well as additional items designed to distinguish between different kinds of res- ignation coping and levels of job ambivalence may further improve the preci- sion of latent cluster boundaries and thus the predictive power of job satisfaction cluster membership.
A further limitation of our study and possible direction for future research arises from our sample composition. Drawing on an online panel, we recruited a broad spectrum of employed individuals from different professions and at different points in their careers. Though the online survey mode may have increased the proportion of participants working in an office, on the whole we had a highly heterogeneous sample. On the one hand, this makes it more likely that our sample contained most job satisfaction types likely to emerge in the
FORMS OF JOB SATISFACTION 549
VC 2017 International Association of Applied Psychology.
course of an average working career. At the same time, this heterogeneity makes it more difficult to explore the effects of specific situations on the devel- opment of job satisfaction. In a relatively homogeneous sample of newly hired employees, for instance, we might be able to see effects of differences between the ideal and actual work situation more clearly, since such differences are likely to be greatest before employees have had the chance to adjust their expectations to a given working environment. This in turn means that it should be easiest to identify adjustments and to observe whether they have the postu- lated consequences for satisfaction during the first weeks and months follow- ing job entry. Thus, future research looking into the development of job satisfaction in a homogeneous sample just entering the workforce is particu- larly interesting from a Bruggemann perspective.
A more general problem with our operationalisation of the Bruggemann construct is that laying our focus on the current form of (dis)satisfaction neglects the opportunity to test the model�s assumptions about how these forms come to be. Bruggemann�s model is not only a typology, but also a pro- cess model, and its assumptions about the antecedents and causes of the differ- ent types of (dis)satisfaction are the theoretical aspects most likely to be of practical use in promoting the more positive types of satisfaction and prevent- ing the more negative types from developing. Thus, in order to determine whether a discrepancy between the ideal and actual work situation truly does interact with level of aspiration to produce different forms of (dis)satisfaction, it is necessary to question participants about their ideal and actual work situa- tions. Research on person-environment fit has made many relevant methodo- logical contributions that could be a starting point for a more sophisticated operationalisation of the dynamic comparison process at the root of the model (e.g. Edwards, Cable, Williamson, Lambert, & Shipp, 2006).
In order to track the shape of developments over time, it is also necessary to study participants on more than two occasions. Tracking subjective ideals and perceived situation across multiple occasions (e.g. using diary study methodol- ogy) offers a direct way of measuring changes in level of aspiration and of test- ing Bruggemann�s process assumptions without relying on (presumably biased) post-hoc judgments of those changes and without being forced into a rough linear extrapolation based only on two time points. Linking subjective estimates of the actual work situation with objective indicators—for instance, those available through modern tracking technologies (e.g. Remijn, Stembert, Mulder, & Choenni, 2015; Swan, 2013) or other multimethod approaches (e.g. Eid & Diener, 2006)—may even provide a way to capture the “pseudo- satisfaction” construct, which Bruggemann (1976) wrote off as unmeasurable.
In summary, going beyond a single, overall satisfaction measure seems a promising way to help increase our understanding of the mechanisms driving positive as well as negative work outcomes. While overall satisfaction is impor- tant, additional aspects of engagement, resignation, and problem-solving
550 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
attempts accompanying that self-estimate of “job satisfaction” may help explain or even bring to light inconsistencies in the overall measure�s ability to predict practically relevant outcomes. Ultimately, being (dis)satisfied with work can be a critical liability; it can also be a source of change and improvement. Bruggemann�s model offers a way of differentiating between types of (dis)satis- faction in order to help researchers understand the processes involved and to help practitioners ensure better outcomes for employee and employer alike.
REFERENCES
Allen, N.J., & Meyer, J.P. (1990). The measurement and antecedents of affective, con- tinuance and normative commitment to the organization. Journal of Occupational Psychology, 63, 1–18. doi:10.1111/j.2044-8325.1990.tb00506.x
Arnold, A., & Mahler, P. (2010). Effects of different forms of job satisfaction on commitment and intention to quit. Discussion Paper No. 15, University of Zurich. Retrieved from https://www.yumpu.com/en/document/view/10633434
Baumgartner, C., & Udris, I. (2006). Das “Z€urcher Modell” der Arbeitszufrieden- heit: 30 Jahre “still going strong” [The Zurich model of job satisfaction]. In L. Fischer (Ed.), Arbeitszufriedenheit: Konzepte und empirische Befunde (pp. 111–134). G€ottingen: Hogrefe.
Bennett, A.A., Gabriel, A.S., Calderwood, C., Dahling, J.J., & Trougakos, J.P. (2016). Better together? Examining profiles of employee recovery experiences. Journal of Applied Psychology, 101, 1635–1654. doi:10.1037/apl0000157
Boverman, D.M. (1962). Normative and ipsative measurement in psychology. Psychological Review, 69, 295–305. doi:10.1037/h0045573
Bowling, N.A., Eschleman, K.J., & Wang, Q. (2010). A meta-analytic examination of the relationship between job satisfaction and subjective well-being. Journal of Occupational and Organizational Psychology, 83, 915–934. doi:10.1348/ 096317909X478557
Brown, A., Charlwood, A., & Spencer, D.A. (2012). Not all that it might seem: Why job satisfaction is worth studying despite it being a poor summary measure of job quality. Work, Employment & Society, 26, 1007–1018. doi:10.1177/ 0950017012461837
Bruggemann, A. (1974). Zur Unterscheidung verschiedener Formen von Arbeitszu- friedenheit [The differentiation of different forms of work satisfaction]. Arbeit und Leistung, 28, 281–284.
Bruggemann, A. (1976). Zur empirischen Untersuchung verschiedener Formen von Arbeitszufriedenheit [Empirical investigation of different forms of work satisfac- tion]. Zeitschrift f€ur Arbeitswissenschaft, 30, 71–74.
Bruggemann, A., Groskurth, P., & Ulich, E. (1975). Arbeitszufriedenheit [Job satis- faction]. Bern: Huber.
B€ussing, A. (1992). A dynamic view of job satisfaction in psychiatric nurses in Germany. Work and Stress, 6, 239–259. doi:10.1080/02678379208259956
B€ussing, A., Bissels, T., Fuchs, V., & Perrar, K.-M. (1999). A dynamic model of work satisfaction: Qualitative approaches. Human Relations, 52, 999–1028. doi:10.1177/ 001872679905200802
FORMS OF JOB SATISFACTION 551
VC 2017 International Association of Applied Psychology.
Cangiano, F., & Parker, S.K. (2015). Proactivity for mental health and well-being. In S. Clarke, T.M. Probst, F. Guldenmund, & J. Passmore (Eds.), The Wiley- Blackwell handbook of the psychology of occupational safety and workplace health (pp. 228–250). Chichester: Wiley Blackwell.
Carver, C.S., & Connor-Smith, J. (2010). Personality and coping. Annual Review of Psychology, 61, 679–704. doi:10.1146/annurev.psych.093008.100352
Christian, M.S., Garza, A.S., & Slaughter, J.E. (2011). Work engagement: A quanti- tative review and test of its relations with task and contextual performance. Personnel Psychology, 64, 89–136. doi:10.1111/j.1744-6570.2010.01203.x
Cohen, J., Cohen, P., & West, S.G. (2002). Applied multiple regression/correlation analysis for the behavioral sciences (3rd edn). Mahwah, NJ: Taylor & Francis.
Collins, L.M., & Lanza, S.T. (2010). Latent class and latent transition analysis: With applications in the social, behavioral, and health sciences. Hoboken, NJ: Wiley.
Conner, M., & Armitage, C.J. (2008). Attitudinal ambivalence. In W.D. Crano & R. Prislin (Eds.), Attitudes and attitude change (pp. 261–286). New York: Psychology Press.
Crant, J.M. (2000). Proactive behavior in organizations. Journal of Management, 26, 435–462. doi:10.1177/014920630002600304
De Witte, H., De Cuyper, N., Handaja, Y., Sverke, M., N€aswall, K., & Hellgren, J. (2010). Associations between quantitative and qualitative job insecurity and well- being. International Studies of Management & Organization, 40, 40–56. doi: 10.2753/IMO0020-8825400103
Edwards, J.R. (2008). Person-environment fit in organizations: An assessment of the- oretical progress. The Academy of Management Annals, 2, 167–230. doi:10.1080/ 19416520802211503
Edwards, J.R., Cable, D.M., Williamson, I.O., Lambert, L.S., & Shipp, A.J. (2006). The phenomenology of fit: Linking the person and environment to the subjective experience of person-environment fit. Journal of Applied Psychology, 91, 802–827. doi:10.1037/0021-9010.91.4.802
Eid, M., & Diener, E. (2006). Handbook of multimethod measurement in psychology. Washington, DC: American Psychological Association.
Eurofound (2012). Health and well-being at work: A report based on the fifth European Working Conditions Survey. Retrieved from http://www.eurofound. europa.eu/publications/report/2013/working-conditions/health-and-well-being-at- work
Faragher, E.B., Cass, M., & Cooper, C.L. (2005). The relationship between job satis- faction and health: A meta-analysis. Occupational and Environmental Medicine, 62, 105–112. doi:10.1136/oem.2002.006734
Fay, D., & H€uttges, A. (2016). Drawbacks of proactivity: Effects of daily proactivity on daily salivary cortisol and subjective well-being. Journal of Occupational Health Psychology, Advance online publication. doi:10.1037/ocp0000042
Ferreira, Y. (2009). FEAT–Fragebogen zur Erhebung von Arbeitszufriedenheit- stypen: Zukunftsperspektive f€ur das Z€uricher Modell [Questionnaire for assess- ment of types of job satisfaction: Future prospects of the Zurich Model]. Zeitschrift f€ur Arbeits- und Organisationspsychologie [German Journal of Work and Organizational Psychology], 53, 177–193. doi:10.1026/0932-4089.53.4.177
552 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
Folkman, S., & Lazarus, R.S. (1988). Coping as a mediator of emotion. Journal of Personality and Social Psychology, 54, 466–475. doi:10.1037/0022-3514.54.3.466
Gagn�e, M., & Deci, E.L. (2005). Self-determination theory and work motivation. Journal of Organizational Behavior, 26, 331–362. doi:10.1002/job.322
Gogol, K., Brunner, M., Goetz, T., Martin, R., Ugen, S., Keller, U., . . . Preckel, F. (2014). “My questionnaire is too long!” The assessments of motivational-affective constructs with three-item and single-item measures. Contemporary Educational Psychology, 39, 188–205. doi:10.1016/j.cedpsych.2014.04.002
Greenhalgh, L., & Rosenblatt, Z. (1984). Job insecurity: Toward conceptual clarity. The Academy of Management Review, 9, 438–448. doi:10.2307/258284
Hallquist, M.N., & Wright, A.G.C. (2014). Mixture modeling methods for the assess- ment of normal and abnormal personality, Part I: Cross-sectional models. Journal of Personality Assessment, 96, 256–268. doi:10.1080/00223891.2013.845201
Hellgren, J., Sverke, M., & Isaksson, K. (1999). A two-dimensional approach to job insecurity: Consequences for employee attitudes and well-being. European Journal of Work and Organizational Psychology, 8, 179–195. doi:10.1080/135943299398311
Hobfoll, S.E. (1989). Conservation of resources: A new attempt at conceptualizing stress. American Psychologist, 44, 513–524. doi:10.1037/0003-066X.44.3.513
Hom, P.W., Mitchell, T.R., Lee, T.W., & Griffeth, R.W. (2012). Reviewing employee turnover: focusing on proximal withdrawal states and an expanded criterion. Psychological Bulletin, 138, 831–858. doi:10.1037/a0027983
Inauen, A., Jenny, G.J., & Bauer, G.F. (2015). Discriminating five forms of job satis- faction: Investigating their relevance for occupational health research and prac- tice. Psychology, 6, 138–150. doi:10.4236/psych.2015.62013
Jones, J.A., & Waller, N.G. (2015). The normal-theory and asymptotic distribution- free (ADF) covariance matrix of standardized regression coefficients: Theoretical extensions and finite sample behavior. Psychometrika, 80, 365–378. doi:10.1007/ s11336-013-9380-y
Judge, T.A., Heller, D., & Mount, M.K. (2002). Five-factor model of personality and job satisfaction: A meta-analysis. Journal of Applied Psychology, 87, 530–541. doi: 10.1037//0021-9010.87.3.530
Judge, T.A., & Ilies, R. (2004). Affect and job satisfaction: A study of their relation- ship at work and at home. Journal of Applied Psychology, 89, 661–673. doi: 10.1037/0021-9010.89.4.661
Judge, T.A., Thoresen, C.J., Bono, J.E., & Patton, G.K. (2001). The job satisfaction– job performance relationship: A qualitative and quantitative review. Psychological Bulletin, 127, 376–407. doi:10.1037/0033-2909.127.3.376
Landy, F.J., & Conte, J.M. (2010). Work in the 21st century: An introduction to industrial and organizational psychology (3rd edn). Hoboken, NJ: Wiley- Blackwell.
Lapierre, L.M., & Hackett, R.D. (2007). Trait conscientiousness, leader-member exchange, job satisfaction and organizational citizenship behaviour: A test of an integrative model. Journal of Occupational and Organizational Psychology, 80, 539–554. doi:10.1348/096317906X154892
Lazarus, R.S. (1993). Coping theory and research: past, present, and future. Psychosomatic Medicine, 55, 234–247.
FORMS OF JOB SATISFACTION 553
VC 2017 International Association of Applied Psychology.
LePine, J.A., Erez, A., & Johnson, D.E. (2002). The nature and dimensionality of organizational citizenship behavior: A critical review and meta-analysis. The Journal of Applied Psychology, 87, 52–65. doi:10.1037//0021-9010.87.1.52
Lo, Y., Mendell, N.R., & Rubin, D.B. (2001). Testing the number of components in a normal mixture. Biometrika, 88, 767–778.
Luttrell, A., Petty, R.E., & Bri~nol, P. (2016). Ambivalence and certainty can interact to predict attitude stability over time. Journal of Experimental Social Psychology, 63, 56–68. doi:10.1016/j.jesp.2015.11.008
Maio, G.R., Bell, D.W., & Esses, V.M. (1996). Ambivalence and persuasion: The processing of messages about immigrant groups. Journal of Experimental Social Psychology, 32, 513–536. doi:10.1006/jesp.1996.0023
Mathieu, J.E., & Zajac, D.M. (1990). A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment. Psychological Bulletin, 108, 171–194. doi:10.1037/0033-2909.108.2.171
Meyer, J.P., Stanley, D.J., Herscovitch, L., & Topolnytsky, L. (2002). Affective, con- tinuance, and normative commitment to the organization: A meta-analysis of antecedents, correlates, and consequences. Journal of Vocational Behavior, 61, 20– 52. doi:10.1006/jvbe.2001.1842
Meyer, J.P., Stanley, L.J., & Vandenberg, R.J. (2013). A person-centered approach to the study of commitment. Human Resource Management Review, 2, 190–202. doi: 10.1016/j.hrmr.2012.07.007
Mobley, W.H., Griffeth, R.W., Hand, H.H., & Meglino, B.M. (1979). Review and conceptual analysis of the employee turnover process. Psychological Bulletin, 86, 493–522. doi:10.1037/0033-2909.86.3.493
Morin, A.J.S., Morizot, J., Boudrias, J.-S., & Madore, I. (2011). A multifoci person- centered perspective on workplace affective commitment: A latent profile/factor mixture analysis. Organizational Research Methods, 14, 58–90. doi:10.1177/ 1094428109356476
Organ, D.W., & Ryan, K. (1995). A meta-analytic review of attitudinal and disposi- tional predictors of organizational citizenship behavior. Personnel Psychology, 48, 775–802. doi:10.1111/j.1744-6570.1995.tb01781.x
Parker, S.K., Bindl, U.K., & Strauss, K. (2010). Making things happen: A model of proac- tive motivation. Journal of Management, 36, 827–856. doi:10.1177/0149206310363732
Podsakoff, P.M., MacKenzie, S.B., Lee, J.-Y., & Podsakoff, N.P. (2003). Common method biases in behavioral research: A critical review of the literature and rec- ommended remedies. Journal of Applied Psychology, 88, 879–903. doi:10.1037/ 0021-9010.88.5.879
Rauch, W.A., Schweizer, K., & Moosbrugger, H. (2007). Method effects due to social desirability as a parsimonious explanation of the deviation from unidimensional- ity in LOT-R scores. Personality and Individual Differences, 42, 1597–1607. doi: 10.1016/j.paid.2006.10.035
Raykov, T., Marcoulides, G.A., & Chang, C. (2016). Examining population heterogeneity in finite mixture settings using latent variable modeling. Structural Equation Modeling: A Multidisciplinary Journal, 23, 726–730. doi:10.1080/10705511.2015.1103193
Remijn, L., Stembert, N., Mulder, I.J., & Choenni, S. (2015). Emergent technologies in multimethod and mixed methods research: Incorporating mobile technologies.
554 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
In S.N. Hesse-Biber & R.B. Johnson (Eds.), The Oxford handbook of multimethod and mixed methods research inquiry (pp. 548–560). Oxford: University Press.
Riketta, M. (2008). The causal relation between job attitudes and performance: A meta-analysis of panel studies. Journal of Applied Psychology, 93, 472–481. doi: 10.1037/0021-9010.93.2.472
Ryan, R.M., & Deci, E.L. (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25, 54–67. doi:10.1006/ceps.1999.1020
Schaufeli, W.B., & Bakker, A.B. (2003). UWES Utrecht Work Engagement Scale: Preliminary manual. Occupational Health Psychology Unit, Utrecht University, Utrecht. Retrieved from http://www.beanmanaged.com/doc/pdf/arnoldbakker/ articles/articles_arnold_bakker_87.pdf
Schaufeli, W.B., & Bakker, A.B. (2004). Job demands, job resources, and their relationship with burnout and engagement: A multi-sample study. Journal of Organizational Behavior, 25, 293–315. doi:10.1002/job.248
Schmidt, K.-H., Hollmann, S., & Sodenkamp, D. (1998). Psychometrische Eigenschaf- ten und Validit€at einer deutschen Fassung des “Commitment”-Fragebogens von Allen und Meyer (1990) [Psychometric properties and validity of a German version of Allen and Meyer�s (1990) questionnaire for measuring organizational commit- ment]. Zeitschrift f€ur Differentielle und Diagnostische Psychologie, 19, 93–106.
Schwarz, G. (1978). Estimating the dimension of a model. The Annals of Statistics, 6, 461–464. doi:10.1214/aos/1176344136
Staufenbiel, T., & Hartz, C. (2000). Organizational Citizenship Behavior: Entwicklung und erste Validierung eines Meßinstruments. [Organizational Citi- zenship Behavior: Development and validation of a measurement instrument.]. Diagnostica, 46, 73–83. doi:10.1026//0012-1924.46.2.73
Sullivan, S.E., & Arthur, M.B. (2006). The evolution of the boundaryless career con- cept: Examining physical and psychological mobility. Journal of Vocational Behavior, 69, 19–29. doi:10.1016/j.jvb.2005.09.001
Swan, M. (2013). The quantified self: Fundamental disruption in big data science and biological discovery. Big Data, 1, 85–99. doi:10.1089/big.2012.0002
Tett, R.P., & Meyer, J.P. (1993). Job satisfaction, organizational commitment, turnover intention, and turnover: Path analyses based on meta-analytic findings. Personnel Psychology, 46, 259–293. doi:10.1111/j.1744-6570.1993.tb00874.x
Tomprou, M., Rousseau, D.M., & Hansen, S.D. (2015). The psychological contracts of violation victims: A post-violation model. Journal of Organizational Behavior, 36, 561–581. doi:10.1002/job.1997
Wang, M., & Hanges, P.J. (2011). Latent class procedures: Applications to organiza- tional research. Organizational Research Methods, 14, 24–31. doi:10.1177/ 1094428110383988
Wanous, J.P., Reichers, A.E., & Hudy, M.J. (1997). Overall job satisfaction: How good are single-item measures? Journal of Applied Psychology, 82, 247–252. doi: 10.1037/0021-9010.82.2.247
Warr, P. (1990). The measurement of well-being and other aspects of mental health. Journal of Occupational Psychology, 63, 193–210. doi:10.1111/j.2044-8325. 1990.tb00521.x
FORMS OF JOB SATISFACTION 555
VC 2017 International Association of Applied Psychology.
Wegge, J., & Neuhaus, L. (2002). Emotionen bei der B€uroarbeit am PC: Ein Test der “affective events”-Theorie [Emotions and office work with computers: A test of the affective events theory]. Zeitschrift f€ur Arbeits- und Organisationspsychologie [German Journal of Work and Organizational Psychology], 46, 173–184. doi: 10.1026//0932-4089.46.4.173
WHO Collaborating Centre in Mental Health (1998). WHO-5 Fragebogen zum Wohlbefinden [WHO-5 Well-Being Index]. Retrieved from https://www.psykiatri- regionh.dk/who-5/Pages/default.aspx
Wiley, J. (2010). Strategic employee surveys: Evidence-based guidelines for driving organizational success. San Francisco, CA: Jossey-Bass.
Woo, S.E., & Allen, D.G. (2014). Toward an inductive theory of stayers and seekers in the organization. Journal of Business and Psychology, 29, 683–703. doi:10.1007/ s10869-013-9303-z
Wright, A.G.C., & Hallquist, M.N. (2014). Mixture modeling methods for the assess- ment of normal and abnormal personality, Part II: Longitudinal models. Journal of Personality Assessment, 96, 269–282. doi:10.1080/00223891.2013.830262
Yuan, K.-H., & Chan, W. (2011). Biases and standard errors of standardized regres- sion coefficients. Psychometrika, 76, 670–690. doi:10.1007/s11336-011-9224-6
Ziegler, R., Hagen, B., & Diehl, M. (2012). Relationship between job satisfaction and job performance: Job ambivalence as a moderator. Journal of Applied Social Psychology, 8, 2019–2040. doi:10.1111/j.1559-1816.2012.00929.x
Ziegler, R., & Schlett, C. (2013). Formen der Arbeitszufriedenheit: Untersuchungen zur Validit€at der Selbstzuordnungsmethode sowie zu Unterschieden in der Valenz der Arbeit, der Arbeitssituation und der dispositionalen Affektivit€at [Forms of job satisfaction: Studies on the validity of the self-classification method and on differences in job valence, job situation, and dispositional affectivity]. Zeitschrift f€ur Arbeits- und Organisationspsychologie [German Journal of Work and Organizational Psychology], 57, 51–76. doi:10.1026/0932-4089/a000107
Ziegler, R., Schlett, C., Casel, K., & Diehl, M. (2012). The role of job satisfaction, job ambivalence, and emotions at work in predicting organizational citizenship behavior. Journal of Personnel Psychology, 11, 176–190. doi:10.1027/1866-5888/ a000071
556 KOVACS ET AL.
VC 2017 International Association of Applied Psychology.
Copyright of Applied Psychology: An International Review is the property of Wiley- Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
