Professional Self-efficacy and Job Satisfaction: The Mediator Role of Work Design

Introduction

Over the last decades, diverse and profound changes have taken place in organizations and in the way work is structured. This context of uncertain and volatile scenarios demands people to go beyond what their tasks and roles where thought for, therefore being not a surprise that different organizational behaviour related disciplines are more interested in identifying different factors that could facilitate a better adaptation to eventualities. A way to approach this has been focusing on certain constructs related to personal development and wellbeing (Merino, Fernández-Ríos, & Bargsted, 2015), such as the study of work design and person-job fit conditions (Parker, Van den Broeck, & Holman, 2017b).

Within this understanding framework, self-efficacy becomes a key factor since its effect on both workplace and psycho-social wellbeing has been demonstrated (Judge & Bono, 2001; Ventura, Salanova, & Llorens, 2015), which includes major variables such as job satisfaction, performance, work conditions, among others. Traditionally, it has been suggested that self-efficacy plays a mediating role in the relationship between work design and several work outcomes (e.g., Choi, 2016; Judge, Bono, & Locke, 2000; Wang & Netemeyer, 2002). However, Parker, Wall, and Cordery (2001) considered that self-efficacy can be an antecedent of work design – in the expanded model of work design developed by Humphrey, Nahrgang, and Morgeson (2007) they did not consider self-efficacy as a mediator.

Taking this into account, the aim of the present study is to test the mediator role of work design in the relationship between self-efficacy and job satisfaction. With this paper we aim to contribute to the understanding of the relation between professional self-efficacy (a specific kind of work self-efficacy), work design, and job satisfaction. Therefore, the contribution of this research will be to obtain a better understanding about the psychological mechanisms behind the impact of job design dimensions over job satisfaction, which implies taking control over the own behavior when a person’s beliefs about their own skills are strong and realistic.

Self-efficacy as an Antecedent of Work Design and Job Satisfaction

Conceptually, self-efficacy was proposed within the Social Cognitive Theory framework, being understood as “beliefs about one’s skills and abilities to organize and execute the required courses of action in order to achieve certain goals” (Bandura, 1997, p. 3). Self-efficacy provides people with confidence regarding their capacity to take control over different life aspects. In this manner, self-efficacy would be a personal competence that works triggered by stressors, increasing or decreasing the psychological discomfort that they could generate (Meseguer, Soler, & García-Izquierdo, 2014). In fact, people who show high self-efficacy levels are also confident about their ability to respond to external stimuli, influencing their way to perceive and process environmental demands or threats (Salanova, Grau, & Martínez, 2005).

It is worth mentioning that the Social Cognitive Theory sustains that self-efficacy beliefs are specific to certain domain (Bandura, 2001); consequently, an individual can feel less or more effective depending on the activity developed, which implies that the more important the domain is, the more determining self-efficacy beliefs are. Therefore, it is necessary to measure self-efficacy in its particular context, i.e., in the organizational field it is more pertinent to measure professional self-efficacy instead of self-efficacy in general (Salanova et al., 2005). On a practical level, professional self-efficacy would then imply a major self-perceived work activity control (Jones & Fletcher, 2003; Merino et al., 2015).

Accordingly, Bandura’s (1997) theoretical approach sustains that people’s beliefs towards their capacities to carry out their work would have an influence on their motivation to seek or avoid certain tasks. Consequently, Judge et al. (2000) posed that individuals with a positive self-concept should be more willing to assume tasks given that they are more confident in their ability to handle challenges not related to their work activity. Oldham and Fried (2016), in turn, suggested that personal characteristics and skills (like self-efficacy, for example) are frequently being configured by work design characteristics, including attributes such as autonomy and discretion, among others. Moreover, self-efficacy plays a major role in the contemporary study of work motivation and its results, adopting a predictive role in relation with different facets of work activity, and not only including performance but also job satisfaction and wellbeing in the workplace, among others (e.g., Judge, Jackson, Shaw, Scott, & Rich, 2007; Stajkovic & Luthans, 1998). There is scientific evidence available that shows a positive relation of self-efficacy and job satisfaction (Judge & Bono, 2001; Perdue, Reardon, & Peterson, 2007). In this sense, Judge et al (2000) argued that self-efficacy has an impact on work activity through its association with practical job success, mainly because people with high self-efficacy beliefs face difficulties more effectively and pursue their efforts, increasing their job satisfaction. Likewise, in their meta-analysis, Judge and Bono (2001) found out that self-efficacy showed a real estimated correlation of .45 with job satisfaction. Moreover, people with higher professional self-efficacy beliefs have more optimistic thoughts that favors their commitment and job satisfaction (Salanova et al., 2005).

On the other hand, taking into account that self-efficacy is considered part of the Core Self Evaluations (CSE; Judge, Locke, & Durham, 1997), several studies have shown that perceptions of job characteristics are typically found to be higher among individuals with positive CSEs (Akkermans & Tims, 2017; Judge, Van Vianen, & De Pater, 2004; Srivastava, Locke, Judge, & Adams, 2010). Judge et al. (1997) argued that people with high self-efficacy might perceive autonomy in a job where people with low self-efficacy perceive bureaucracy. Thus, professional self-efficacy can help to perceive enriched perceptions of job characteristics. Considering this, we stablished the following hypothesis:

H1: Professional self-efficacy will be positively related to job satisfaction (H1a) and work design characteristics (H1b).

Work Design and Job Satisfaction

Work design as a field of study is enjoying a mayor booming in Applied Psychology (Parker, Morgeson, & Johns, 2017a), which is not a mere temporary answer but a very needed response to changes in the nature of work in a context of contemporary organizations and globalization (Fernández-Ríos et al., 2017). Work design is a process related to how work activity is structured and configured within an organization. It deals with the way that employment, tasks, and roles are represented and modified, thereby showing the impact of structures, representations, and modifications on individual, collective, and organizational results (Grant & Parker, 2009). It is expressed in multiple factors, also known as design characteristics, that can be grouped in three main categories. First, job complexity, expressed in both motivational task characteristics (autonomy, task variety, task significance, task identity, feedback from job) and motivational knowledge characteristics (complexity, information processing, problem solving, skill variety, and specialization). Second, relational work environment, expressed in social characteristics (social support, interdependence, interaction outside organization, and feedback from others). And third, physical demands expressed in physical-contextual characteristics (ergonomics, physical demands, work conditions, equipment use) (Morgeson & Humphrey, 2006).

There is a proliferation of statements that enriched traditional perspectives focused on motivational job factors (Oldham & Fried, 2016). More consistent evidence also shows a major impact of work design on diverse individual, collective, and organizational results (Grant, Fried, & Juillerat, 2010; Parker et al., 2017a). A high-quality work design is key to achieve people’s wellbeing and to adopt positive attitudes at work and ensure a good performance of both individual and organization (Parker et al., 2017b). In this vein, available empirical evidence establishes a positive relationship between work design characteristics and job satisfaction. This is demonstrated by two of the largest meta-analyses in the work design field. The first one, conducted by Fried and Ferris (1987), found out a consistent positive relation between job complexity and job satisfaction. The second and most recent one, conducted by Humphrey et al. (2007), found out that a large part of work design characteristics explained the variance of job satisfaction. In particular, motivational characteristics (task and knowledge) explained 34% of job satisfaction variance, while social and physical-contextual characteristics explain only 17% and 4% respectively. It is worth mentioning that a work design characteristic related to job satisfaction that stands out is autonomy (e.g., Finn, 2001; Saragih, 2011). Consequently:

H2: Work design characteristics are positively related to job satisfaction. Therefore:

H2a: Task characteristics will be positively related to job satisfaction.

H2b: Knowledge characteristics will be positively related to job satisfaction.

H2c: Social characteristics will be positively related to job satisfaction.

H2d: Physical characteristics will be positively related to job satisfaction.

Mediator Role of Work Design Characteristics

There are studies in which job characteristics and self-efficacy are considered both antecedents and mediators of work outcomes. In line with Job Demands Resources (JD-R) model (Bakker & Demerouti, 2007), work design is viewed as work resource and a source of demands, and self-efficacy is considered as personal resource, and that can explain why both were studied as antecedents and mediator. In order to obtain more evidence, the propose of this research is to test a mediating role of work design in the relationship between self-efficacy and work outcomes, understanding work design as a context variable.

Parker et al. (2001) explicitly proposed a set of variables as antecedents of work design within which is the self-efficacy. Parker and colleagues gave some evidence about it in this way and proposed that individuals with proactive personality may influence their jobs autonomy and can enriched their jobs in a similar way that job crafting. Furthermore, Skaalvik and Skaalvik (2014) found that autonomy at work had an impact on the relation between self-efficacy and job satisfaction in a wide sample of teachers.

According to Judge et al. (1997) and Judge and Bono (2001), self-efficacy is considered a stable personal characteristic related to self-regulation mechanisms, like core self-evaluations. Self-efficacy implies self-confidence in generating effective action-plans, handle difficulties, managing emotions, stress, and anxiety, and exercising control (Bandura, 2012). From this point of view, professional self-efficacy is likely to affect enriched perceptions of job characteristics. In this vein, Judge et al. (2000) tested a model concerning the relation among personal self-evaluations (among them general self-efficacy), intrinsic characteristics of work design, and job satisfaction and found that both perceived intrinsic work design characteristics and job complexity mediated the relation between personal self-efficacy and job satisfaction. Also, Srivastava et al. (2010) found that perceptions of job characteristics mediate the relationship between core self-evaluation and work satisfaction.

This relationship can also be associated with job crafting, where people with high self-efficacy proactively shape their work environment in several ways (Borgogni, Dello Russo, Miraglia, & Vecchione, 2013), contributing to create necessary conditions to meet their own needs, goals, and preference. As Judge and Bono (2001) stated, they can do this because they can manage problematic and challenging situations at work, dealing effectively with personal emotions. Even this study is not focused on job crafting behaviors; it is possible to hypothesize that the same psychological mechanisms can appear when motivational work design characteristics are higher. Then, people with higher self-efficacy in more complex jobs will be more satisfied because they can align proactively their personal and organizational goals. Van den Heuvel, Demerouti, and Peeters (2015) stated that job crafting has been incorporated to Job Demands Resources (JD-R) model. Also, according to Wrzesniewski and Dutton (2001), job crafting is a job redesign approach related to changes that an employee can do on the type and number of tasks, on the interactions that she/he need to make, and on the significance of their work in order to create more meaning. Also, self-efficacy has been included as a personal resource that adds motivation, adaptability, and wellbeing (van den Heuvel, Demerouti, Bakker, & Schaufeli, 2010).

Therefore, it is reasonable to believe that job design can be understood as a work environment variable that has an impact on the relation between professional self-efficacy and job satisfaction. Specifically, according the specific work characteristics related to a job, personal beliefs about personal competencies (self-efficacy) could be a strong motivator of behavior having an impact on work outcomes by making employers more confident about solving conflicts, overcoming frustrations, and persisting against difficulties, and, therefore, being more satisfied with their job performance. Thus, considering that people with high self-efficacy (as a stable personal characteristic) are more likely to perceive their job characteristics (contextual variable) positively and are less likely to focus on negative information, it is possible to assume the mediating role of work design perceptions in the relationship between self-efficacy and job satisfaction. Consequently, we proposed the following hypothesis:

H3. Work design characteristics will mediate the relationship between professional self-efficacy and job satisfaction.

The proposed model and hypotheses are presented in Figure 1.

Method

Participants

A total of 353 Chilean workers participated in this study, from several organizations and careers at 72 different work positions in health (32%), education (27.5%), retail and commerce (14%), mining (10.4%), hospitality (7.4%), transport (4.8%), and construction (3.9%). Sixty four percent of participants worked in the public sector, 30% in private companies, and 6% in NGOs. The average tenure in their respective jobs was 6.25 years (SD = 4.3).

Fifty six percent of participants were women. The average age was 38.9 years old (SD = 12.46). The distribution by educational degree was as follows: 1.7% of participants had a master degree, 45.3% had a university degree, 27.1% had technical studies, and 25.9% had completed secondary education.

Instruments

Participants answered a survey that included:

Spanish version of the Professional Efficacy Scale (Salanova, Grau, Llorens, & Schaufeli, 2001). This scale has 10 items in a six-point Likert scale. The original authors reported a Cronbach’s alpha (internal consistency) of .70-.86 and in this research we observed a Cronbach’s alpha of .78. An example of items is “I remain calm when I face difficulties in my work because I trust my possibilities”

Spanish version of the Job Descriptive Index (JDI), developed by Smith, Kendall, and Hulin (1969) and updated by Smith et al. (1987), with 72 dichotomic items related to five dimensions of satisfaction: work, income, coworker, supervisor, and career opportunities. In the Spanish version Merino et al (2015) reported an internal consistency of .84, and in this research Cronbach’s alpha was .87. An example of item is “Think about your actual work, how creative is your job the most part of the time?”.

Spanish version of the Work Design Questionnaire by Morgeson and Humphrey (2006), adapted by Fernández-Ríos et al. (2017), with 77 items in a five-point Likert scale. The questionnaire explores 21 dimensions of work design, grouped in 4 main categories: motivational work characteristics (broken down into task characteristics and knowledge characteristics), social characteristics, and contextual characteristics. Total instrument’s internal consistency Cronbach’s alpha was .92 and the various scales’ reliability ranged from .70 to .96, except for three dimensions. CFA results indicated goodness of factor configurations corresponding to each of the four major categories of work characteristics, with CFI and TLI around .90, as well as SRMR and RMSEA below .08. An example of item is “The job allows me to plan how I do my work.”

Procedure

We used non-probabilistic sampling with volunteers. To contact participants, we obtained authorization from their organizations. They received a message indicating the aim of the study, a link to an online anonymous survey, and an informed consent document. The data were collected within a period of three months (May-July 2017). We invited 550 participants, 358 answered the questionnaire, and 5 participants were excluded for incomplete information, the response rate being 65%. All descriptive and regression analyses were made with SPSS 19.0. Because data collection was done through a self-report survey in the same moment, we applied Harman’s one-factor test to verify the common method variance effect (Podsakoff, Mackenzie, Lee, & Podsakoff, 2003). The result of the Harman’s one-factor test showed a factor explaining 14.704% of variance (below 50%). Therefore, the effect of the common method variance does not seem to affect the relationship of the study variables.

Data Analysis

Mediation involves a causal relationship whereby an independent variable (X) impacts on a mediator (M), which in turn affects a dependent variable (Y) (Sobel, 1990). To estimate these relationships, two regression models are needed. First, the mediator (M) is regressed on the independent variable. Second, the dependent variable (Y) is regressed on the mediator (M), controlling for the independent variable (X). Therefore, a relationship is mediated if: X is significantly related to M, M is significantly related to Y after controlling for X, and the mediated effect is statistically significant. In order to contrast H3 of mediation and to estimate the indirect effects, we used Macro Process for SPSS (Hayes, 2013), which gives the confidence intervals (CI) of the bootstrapping of 5,000 samples with a confidence level of 95% (Preacher & Hayes, 2008).

Results

Descriptive Statistics

Table 1 shows means, standard deviations, and correlations among the study variables employed to test hypotheses.

Hypothesis Testing

Hypothesis 1 posits that self-efficacy will be positively related to mediator variables (task, knowledge, social characteristics, and physical conditions). The results obtained by means of regression analysis are displayed in Table 2. Focusing on regression coefficients obtained, self-efficacy has a positive and significant relationship with each mediator (task characteristics: β = .51, p < .001; knowledge characteristics: β = .45, p < .001; social characteristics: β = .28, p < .001; physical characteristics: β = .18, p < .01). In addition, self-efficacy has a positive and significant relationship with job satisfaction (β = .25, p < .001). Therefore, H1 was supported.

Hypothesis 2 predicted that mediator variables would be positively related to job satisfaction. In this case, the results presented in Table 3 after controlling for self-efficacy (β = .05, p > .05), show that some mediators are significantly and positively related to job satisfaction (outcome variable): task characteristics (β = .30, p < .001) and social characteristics (β = .23, p < .001). However, knowledge characteristics (β = -.01, p > .05) and physical characteristics (β = -.04, p > .05) do not present a significant relationship with job satisfaction. Therefore, these results partially supported Hypothesis 2, supporting hypotheses H2a and H2c.

Regarding hypothesis 3 (see Figure 2), self-efficacy has a positive and significant relationship with each mediator (task characteristics: β = .51, p < .001; knowledge characteristics: β = .45, p < .001; social characteristics: β = .28, p < .001; physical characteristics: β = .18, p < .001). After controlling the effect of self-efficacy on job satisfaction, (β = .05, p > .05), task characteristics (β = .30, p < .05), and social characteristics (β = .23, p < .001) have a significant and positive relationship with job satisfaction. However, knowledge characteristics (β = -.01, p < .001) and physical characteristics (β = -.04, p > .05) do not present a significant relationship with job satisfaction.

In H3 we stated that task, knowledge, social characteristics, and physical conditions will partially mediate the relationship between self-efficacy and job satisfaction. Because only task and social characteristics were positively and significantly related to job satisfaction after controlling self-efficacy (see Table 3), the mediated relationships further tested were those involved in these mediator variables. The indirect effect, based on bootstrap procedure, is significant for a mediating role of task characteristics (B = 2.23, boot ET = 0.60, 95% IC [1.03, 3.44]) and for social characteristics (B = 0.91, boot ET = 0.30, 95% IC [0.43, 1.66]). Furthermore, although there is a total effect of self-efficacy on job satisfaction (β = .25, p < .001), the direct effect of the model is not significant (B = 0.66, p > .05). Therefore, the results show a complete mediation of task and social characteristics of work design in the relationship between self-efficacy and job satisfaction (see Figure 2), partially supporting hypothesis 3 of this study.

Discussion

The main goal of this study was to test the mediator role of work design characteristics in the relationship between self-efficacy and job satisfaction. The results obtained supported the mediational relationship between self-efficacy and job satisfaction through task and social characteristics as job design characteristics.

Firstly, following mediational steps, the results showed that self-efficacy was positively related to work design characteristics, which is congruent with the statements of Bandura (1997), Judge et al. (2000), Judge et al. (2004), and Srivastava et al. (2010), who established that some personal characteristics, such as self-efficacy, are related with the perception of enriched jobs.

Secondly, regarding to different work design characteristics proposed as mediators, after controlling for self-efficacy we found that task and social work design characteristics positively influence job satisfaction. Nevertheless, knowledge and physical work characteristics were not related to job satisfaction. These results are consistent with previous studies that show how motivational characteristics of job design explained an important percent of variance of job satisfaction. In Humphrey et al.’s (2007) meta-analysis, the dimension of motivational work characteristics was the one that most explained job satisfaction variance followed by social characteristics, contextual characteristics being the dimension that least explained job satisfaction variance.

Thirdly, although self-efficacy was significantly related to job satisfaction, which is congruent with previous studies (e.g. Judge & Bono, 2001; Judge, Hulin, & Dalal, 2012), when job design characteristics were considered, self-efficacy was not significantly related with job satisfaction. This demonstrates that task and social characteristics fully mediate the relationship between self-efficacy and job satisfaction. These results are in line with Parker et al.’s (2011) proposal, and with the results of Skaalvik and Skaalvik (2014) found that autonomy influences the relationship between self-efficacy and job satisfaction. Moreover, similar results were found by Judge et al. (2000) using job complexity as mediator and core-self-evaluations as predictor (including self-efficacy), and Frese, Garst, and Fay (2007) with regard to how job complexity improves perceived control and self-efficacy. Likewise, our results are consistent with what was formulated in the Job Demands Resources (JD-R) model (Bakker & Demerouti, 2007), referring to the fact that work design characteristics can be potential resources that affect the relationship between personal resources and attitudinal results, as in the case of professional self-efficacy and job satisfaction, respectively.

Contrary to studies that indicate that self-efficacy can act as a mediator of the relationship between job characteristics and job satisfaction (Choi, 2016; Judge et al, 2000; Wang & Netemeyer, 2002), our results indicated that self-efficacy was not related to job satisfaction when considering job characteristics variables. This result indicates that self-efficacy does not have a mediator role between work design characteristics and job satisfaction.

This study presents several key implications. Firstly, we used an expanded work design model proposed by Morgeson and Humphrey (2006), and our results showed the potentially critical role of job enrichment as a mechanism to promote job satisfaction influenced by self-efficacy. In this sense, as Parker (1998) established, job enrichment increases control and perception of internal forces, which in turn increases focus and creativity and affects the type of goals established and regulates effort, all of them key elements for the emergence of self-efficacy. In addition, taking the relational job design perspective into account, work employees’ prosocial motivation can provide more opportunities to interact with others (co-workers, clients, customers, etc.), which affects their motivation and attitudes such as job satisfaction (Grant et al., 2010).

Secondly, our results shed light on the role of professional self-efficacy as antecedent of work design. Since self-efficacy influences how environment and social support are perceived (Bandura, 2001), people who display high levels of self-efficacy tend to interpret task and social characteristics as a challenge that promotes their attitudes and behaviors, contributing to positive job outcomes, such as job satisfaction.

Thirdly, considering previous studies that explain that work design can promote “can do”, “reason to”, and “energized to” motivational states that in turn stimulate self-efficacy (Parker, 1998, 2014), our results go beyond showing how people with a high level of professional self-efficacy are involved in job enrichment. These results contribute to the emergence of a proactive perspective on work design, in which reciprocal relationships between job characteristics and individual attributes are a central assumption (Grant & Parker, 2009). In this sense, personal resources (such as self-efficacy) not only enable individuals to effectively deal with problems and achieve goals, but the experience of these outcomes through enriched job, in terms of task and social characteristics, can reinforce and strengthen these personal resources and affect positively job satisfaction.

From a practical point of view, job enrichment improves psychosocial work conditions and has been related to positive attitudes and job performance. Nevertheless, our results showed that, in order to improve job satisfaction through self-efficacy, it must be through work design and not only by measures associated with motivation without the context of work or focused exclusively on extrinsic elements. Thus, the interventions to improve job satisfaction must be accompanied by job enrichment interventions based on increasing autonomy, identity and variety of tasks, feedback, and social support. This is in line of what Parker et al. (2017a) stated about a view of work design and re-design more integrated, expanded, and context-related.

Our study has several limitations to keep in mind when interpreting the results. First, this research presents the classic limitations of cross-sectional studies, and then it is not possible to establish causal relationships between variables. Despite this limitation, our findings are congruent with previous literature. Nevertheless, future research should conduct longitudinal studies to investigate the relationship addressed here. Second, there is a risk related to common method variance, because data were obtained by self-report measures. In this case, this possible effect in an inflated observed relationship was non-significant, as we tested. Future research with longitudinal design may help to improve the conclusions on the relationship evaluated in this study. However, correlations among the study variables differed in size and, as published studies show, “using a self-report methodology is no guarantee of finding significant results, even with very large samples” (Spector, 2006, p. 224). Moreover, our results show that self-efficacy related differently to the mediation variables considered and those mediation variables related differently to the criterion variable presenting meaningful specificities for each work characteristics. Third, although the sample is composed by diverse work positions of public and private sector, this does not reflect the general population, which limits the generalizability of our results.

In conclusion, our results provide certain insight concerning the relationship between self-efficacy and job satisfaction. This study evidences the importance of motivational and social work design characteristics in this relationship. Accordingly, we propose that future research consider another criterion variable as job performance.