Fit for the Future
An Exploration of the Relationship between Person-Environment Fit and Job
Satisfaction among Nurses and the Moderating Effects of Social Support and
Intrinsic Motivation
Name: Iris Bolt
Student Number: 11363363
Study Programme: MSc Business Administration
University: University of Amsterdam
Supervisor: Dr. C.T. Boon
Statement of originality
This document is written by Iris Bolt who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
This study aimed to contribute to the literature by investigating the relationship between the different dimensions of PE fit (PJ fit, PO fit and PT fit) and job satisfaction among nurses. In addition, this study also looked at intrinsic motivation and social support as possible moderating variables on the relation between the different dimensions of PE fit and job satisfaction. To collect data, a mixed method was used. Quantitative data was collected by using a survey which was filled out by 497 nurses, covering 32 Dutch healthcare organizations. In addition, to gain more in-depth insights in the responses, additional interviews among 19 nurses were conducted. Quantitative results showed a positive relationship between PJ fit and job satisfaction and PO fit and job satisfaction among nurses. Furthermore, compared to PO fit and PT fit, PJ fit was found to be the strongest predictor of job satisfaction. Moderating effects of social support were found for all dimensions of PE fit and intrinsic motivation positively moderated the relationship between PO fit and job satisfaction. The quantitative findings were further supported by the qualitative findings, which also indicated the need to focus on PO fit in order to enhance job satisfaction among nurses. Implications of these findings are discussed.
Key words: Nursing Shortage, Job satisfaction, Person-job fit, Person-organization fit, Person-team fit, Social support, Intrinsic motivation
Table of contents 1. Introduction ... 4 2. Literature review ... 7 2.1. Job satisfaction ... 7 2.2. Person-environment fit ... 9 2.3. Person-job fit ... 10 2.4. Person-organization fit ... 12 2.5. Person-Team fit ... 14 2.6. Social Support ... 17 2.7. Intrinsic Motivation ... 19 3. Method ... 21
3.1. Overall research design and procedure ... 21
3.2. Quantitative procedure ... 23 3.2.1. Quantitative sample ... 23 3.2.2. Measures ... 24 3.2.3. Control Variables ... 26 3.2.4. Statistical procedure ... 26 3.3. Qualitative Procedure... 28 3.3.1. Qualitative Sample ... 30
3.3.2. Qualitative Analysis Procedure ... 30
4. Results ... 31 4.1. Quantitative results ... 31 4.1.1. Correlation analysis ... 31 4.1.2. Direct effects ... 34 4.1.3. Moderating effects ... 35 4.2. Qualitative results ... 43 5. Discussion ... 48
5.1. Theoretical implications and suggestions for further research ... 48
5.2. Practical implications ... 55
5.3. Limitations ... 56
6. Conclusion ... 58
References ... 60
Appendix 1: Survey (in Dutch) ... 71
Appendix 3: Interview questions (in Dutch) ... 76 Appendix 4: Overview interview participants ... 77
1. Introduction
At the moment, the world is dealing with a shortage of nurses (Kingma, 2007; World Health Organization, 2013). The origin for this shortage is twofold. On the one hand, various demographic changes, such as a less active lifestyle and a higher average life expectancy are driving up the demand for healthcare (Mathers, Stevens, Boerma, White, & Tobias, 2015; Murray et al., 2013; PwC Analysis, 2015; Yach, Hawkes, Gould, & Hofman, 2004). On the other hand, the healthcare sector is dealing with recruitment and retention problems (Lu, While & Barribal, 2005). Compared to 50 years ago, women currently have more choices for a career, which makes it less usual for them to become a nurse. In addition, the job suffers a bad image, since it is known to be very stressful, and working conditions, such as mandatory night shifts, are not ideal (Nevidjon & Erickson, 2001). At the same time, retention of nurses is troublesome: in their study of 2007, Kovner, Fairchild, Poornima, Kim and Djukic found that 13% of the newly licensed registered nurses had already changed jobs within the first year, and 37% intended to search for a new job within the second year.
Since nurses make up a significant part of the healthcare workforce and every loss of this recourse has great impact on the quality of care, the shortage of nurses is of serious concern and an urgent need to focus on recruiting and retaining nurses exists (Duffield, Roche, Homer, Buchan, & Dimitrelis, 2014; Lu, While & Barribal, 2005; Nei, Snyder & Litwiller, 2015). As stated by Lu, While and Barribal “recruitment and retention of nurses are persistent problems associated with job satisfaction” (2005, p. 211). More specifically, dissatisfaction with the job is the most cited reason indicated by nurses for leaving their jobs (Ma, Samuels, & Alexander, 2003; Zangaro & Soeken, 2007). To enable organizations to develop management interventions enhancing recruitment and retention of nurses in the
future, a thorough understanding of what exactly influences nurses’ job satisfaction is of paramount importance.
In their meta-analysis of 2005, Kristof-Brown, Zimmerman and Johnson found that in general, the extent to which a person fits the work environment is positively correlated with job satisfaction. However, this relationship has not been investigated among nurses in specific. Investigating such relationship among nurses may nonetheless be interesting for two reasons. Firstly, the meta-analysis of Kristof-Brown et al. only investigated the aforementioned correlation in general. However, the work environment of nurses is significantly different from, for instance, the work environment of an administrative worker, considering – among others – its stressful nature, high level of responsibility and demanding surroundings (McVicar, 2003). This implies that the effect of person-environment fit on nurses’ job-satisfaction might differ from the general results as found by Kristof-Brown et al. (2005). Secondly, person-environment fit is a multidimensional concept that can be divided into different dimensions of fit, among which a fit with the job, a fit with the organization and a fit with the team the employee works with. In order to be able to establish management interventions aimed at improving job satisfaction among nurses, better insight in how exactly and through which dimension person-environment fit influences job satisfaction is needed. Therefore, the main aim of this study is to investigate whether and how person-environment fit influences job satisfaction among nurses.
As indicated by Kristof-Brown et al. (2005) “there is also need for future research on personal and situational characteristics that moderate fit-outcome relationships (…) this area seems particularly in need of research, as many managers face the challenge of minimizing the effects of poor fit for their employees” (p. 322). Therefore, in this study the moderating effects of two nurse-specific characteristics are tested. One of these characteristics is intrinsic motivation, which is a personal condition. Intrinsic motivation is seen as a common
characteristic of nurses since it is known that, compared to other professions, nurses in general are less driven by external rewards like salary (Speedling, 1990). The other characteristic that will be tested is social support, which is a situational condition and is particularly important for nurses. On a daily basis, nurses have to deal with staff shortage, shiftwork, emotional demands, illness and death, which makes nursing a stressful occupation (McVicar, 2003). In coping with all these environmental stressors, social support plays an important stress-buffering role (Rodwell, Munro & Gulyas, 2012). Therefore, this study will investigate whether social support also act as a moderator of the relationship between person-environment fit and job satisfaction among nurses.
Accordingly, the purpose of this study is twofold (see figure 1. for the conceptual framework). On the one hand, the relation between the three different dimensions of person-environment fit (person-job fit, person-organization fit and person-team fit) and job satisfaction among nurses will be accurately assessed. On the other hand, this study aims to contribute to the literature by investigating the moderating roles of intrinsic motivation and social support on the relation between the different dimensions of person-environment fit and job satisfaction. In order to meet the objectives of this study, quantitative data will be collected by means of a survey. In addition, interviews will be used as follow-up to the survey in order to gather more in-depth insights on the attitudes and thoughts of the respondents.
2. Literature review
This chapter discusses findings from relevant literature and states the hypotheses of this study. The chapter starts with describing the most relevant findings about job satisfaction among nurses. Subsequently, the three different dimensions of person-environment fit are explained and their possible effects on job satisfaction are outlined. Finally, the chapter discusses how social support and intrinsic motivation might moderate the relationship between person-environment fit and job satisfaction.
2.1. Job satisfaction
Since its first introduction in the 1930’s, the concept of job satisfaction has received a lot of attention by both researchers and organizations. Consequently, many definitions regarding this topic exist, ranging from very simplistic, such as “all the feelings that an individual has about his job” (Gruneberg, 1976), to more complex and multidimensional (Hulin & Judge, 2003). However, the most widely used definition, which will also be used in this study, is formulated by Locke (1976), who defines job satisfaction as “a pleasurable or positive emotional state resulting from the appraisal of one’s job or job experiences” (p. 1304). As such, the factors that make a job satisfying do not only depend on the direct feelings of individuals about their jobs, but also on the individuals’ expectations of what their job should provide (Lu, While & Barriball, 2005). Therefore, job satisfaction can be seen as the result of an individual’s comparison of how well outcomes of a job match with those that are desired (Oshagbemi, 1999).
In addition to multiple job satisfaction definitions, various job satisfaction theories can also be derived from the existing literature. In general, these theories can roughly be divided into three different groups. The first group of theories that can be identified is the group of Discrepancy Theories. These theories examine the extent to which the needs of an individual
are met within the workplace. The second group of theories that can be distinguished is the group of Equity Theories, which highlight the social comparisons in the evaluation of job rewards. Finally, there are multiple theories that mainly focus on employee motivation, which are collectively called Expectancy Theories (Adams & Bond, 2000).
Equally to the most used definition, the most well-known theory regarding job satisfaction, which will also be used in this study, is composed by Locke (1976). This theory is called the Affect Theory and belongs to the group of Discrepancy Theories. The Affect Theory states that job satisfaction is determined by a discrepancy between what employees want in a job and what they have in a job: the smaller this discrepancy, the more satisfied with the job one will be. Furthermore, the theory states that how much an employee values a given characteristic of work, moderates how satisfied the employee becomes when such characteristic is met (Locke, 1976; Singh & Sinha, 2013). In other words, when someone values, for example, the possibility to work at home, this person will be more satisfied when the job offers him or her this possibility compared to someone who does not value the possibility of working at home.
Since job satisfaction is related to the effectiveness and vitality of an institution, the topic is considered to be of great importance by organizations (Smerek & Peterson, 2007). As found in previous studies, job satisfaction in general is correlated to the physical and mental well-being of employees. Furthermore, it has considerable effects on job related behaviors, such as productivity, performance, absenteeism and turnover (Oshagbemi, 1999). Job satisfaction itself is in its turn influenced by multiple variables, of which the work itself is found to be the strongest predictor (Smerek & Peterson, 2007).
Job satisfaction has also been frequently studied among nurses. In her meta-analysis, Blegen (1993) found that job satisfaction among nurses was positively correlated with organizational commitment, communication with the supervisor, autonomy, recognition,
communication with peers and fairness. In addition, this author found that job satisfaction was negatively correlated with stress and routinization. Furthermore, other studies found various different variables to have statistically significant net effects on job satisfaction, including co-worker support, supervisory support, pay, workload and job opportunity (Chu, Hsu, Price & Lee, 2003; Seo, Ko & Price, 2004).
Although job satisfaction has frequently been studied among nurses, not all possible predictors of job satisfaction among nurses have yet been researched. An example of such an untested predictor is person-environment fit. As mentioned in the introduction, the extent to which a person fits the work environment is significantly correlated with job satisfaction among employees in general (Kristof-Brown et al., 2005). It would therefore be argumentative to research the correlation between person-environment fit and job satisfaction among nurses.
2.2. Person-environment fit
A work environment that fits is not only of significant importance because it facilitates that employees can perform their jobs; it also has large consequences for both the wellbeing of employees and the extent to which organizations are successful. The theory that is related to this is called Person-environment fit (PE fit) which is often defined as “the compatibility between an individual and a work environment that occurs when their characteristics are well matched” (Kristof-Brown et al., 2005, p. 281). The concept of PE fit has gained great attention by organizations, since it is assumed that in case of a high degree of fit between the individual and the individual’s work environment, individuals are more likely to stay at their jobs (Blau, 1987).
Since the introduction of PE fit, several distinct dimensions of fit have evolved from this theory, including the fit of individuals with the jobs they perform (Person-job fit), the fit
between the individuals and the organization they work for (Person-organization fit) and the individual’s fit with the teams they work with (Person-team fit). Although the names of these dimensions seem related or overlapping, previous research has demonstrated that these three dimensions are statically distinct from each other and all have unique effects on outcome variables (Kristof, 1996; Lauver and Kristof-Brown, 2001).
2.3. Person-job fit
Person-job fit (PJ fit) can be defined as the relationship between the characteristics of an individual and those of the individual’s job (Edwards, 1991). According to the literature regarding this topic, a poor fit or misfit between an employee and his or her job might lead to several negative work outcomes, such as stress and burn-out, whereas a strong fit between an employee and his or her job will lead to positive outcomes such as well-being (Edwards, 1991; Kristof, 1996).
Within PJ fit, two perspectives can be distinguished. The first form is the demands-abilities fit in which it is assumed that the knowledge, skills, and demands-abilities (KSA’s) of employees are compliant with what the environment requires (Edwards, 1991). In line with this conceptualization, a PJ fit will be achieved when an employee has the right knowledge, skills and abilities to meet the demands of the job. The second form is called the needs-supplies fit in which the needs and desires of employees are fulfilled by the environment in which they work (Edwards, 1991). Following this conceptualization, a PJ fit will be achieved when the job meets the needs of the employee. In the literature, the latter type of fit has been the primary focus of theories of satisfaction (Kristof-Brown et al., 2005).
It will not be the first time that the relationship between PJ fit and job satisfaction will be investigated. In 1994, Bretz and Judge published a study in which they found a positive moderate correlation between PJ fit and satisfaction. They concluded that individuals should
be conscious about the degree to which they fit in their organization, since a strong PJ fit leads to higher levels of both satisfaction and extrinsic success and therefore, the “consequences of not fitting may be quite serious” (Bretz & Judge, 1994, p. 49). A few years later, Kristof-Brown et al. (2005) and Lauver and Kristof-Brown (2001) found a strong correlation between PJ fit and job satisfaction.
The literature is currently lacking evidence for a direct positive relationship between PJ fit and job satisfaction among nurses. However, in this study it is expected that this direct positive relationship exists among all professions, and therefore also among nurses. In line with the needs-supplies fit, PJ fit in particular is achieved when the needs and desires of an employee are fulfilled by the job. In other words: a fit between a nurse and his or her job exists when what the nurse wants is met by what the job offers. At the same time, the Affect Theory of job satisfaction states that the smaller the discrepancy is between what employees want in a job and what the job offers, the more satisfied with the job one will be. Since both theories imply the same, that is, an overlap between what the employee desires and what the job offers, it is expected that a fit between a nurses and their work-environment (e.g. a smaller discrepancy between what someone wants and what a job offers) will have a positive effect on nurses’ job satisfaction.
Based on these theoretical arguments and the previously reported positive relationships between PJ fit and job satisfaction among employees other than nurses, the following hypothesis is conducted:
Hypothesis 1: There is a positive relationship between person-job fit and job satisfaction among nurses.
2.4. Person-organization fit
The second dimension of PE fit concerns the compatibility between employees and the organizations they work for and is called the Person-organization fit (PO fit). According to Kristof (1996), this type of fit is achieved when both the employee and the organization share the same fundamental characteristics or when at least one of the two provides something that is needed by the other entity.
In this dimension of fit, especially the values of both individuals and organizations play an important role (Kristof-Brown et al. 2005), since a high congruence between these values can lead to favorable outcomes which are relevant to both parties (Edwards & Cable, 2009). Individual values are general beliefs about the importance of normatively desirable behaviors or end states (Edwards & Cable, 2009). Based on these values, individuals make decisions. This can be demonstrated by the example of sustainability. When someone values sustainability, he or she will take this in consideration when buying a car and will therefore be more likely to choose a sustainable car. Organization values “provide norms that specify how organizational members should behave and how organizational resources should be allocated” (Edwards & Cable, 2009, p. 655). Although the origin of the values differs, both types of values focus on behavior and the desirability thereof. When there is a high degree of overlap between individual and organizational values, this is referred to as Value Congruence (Chatman, 1989; Kristof, 1996).
PO fit also plays an important role during the recruitment and selection process. According to the attraction-selection-attrition theory of Schneider (2001), employees feel more attracted to organizations they perceive to fit with and are, therefore, more likely to seek employment with organizations that exhibit characteristics desired by them. In addition, organizations are in turn more likely to select future employees who are similar to the organization.
The relationship between PO fit and job satisfaction has been extensively investigated. The majority of studies of the relationship between PO fit Job Satisfaction found that PO fit is strongly positively related with job satisfaction (e.g. Kristof-Brown et al., 2005; Verquer, Beehr & Wagner, 2003). Kristof (1996) theorized that the greater the value congruence between the employee and the organization, the higher the degree to which an employee is satisfied. In addition, although weaker than the positive relationship, Wheeler, Buckley, Halbesleben, Brouer and Ferris (2005) also found the opposite effect to exist: poor PO fit leads to decreased job satisfaction among employees. Finally, with respect to job satisfaction among nurses, a recent study by Risman, Erickson and Diefendorff (2016) investigated whether the positive relationship between PO fit and job satisfaction existed among nurses of the United States and found that PO fit as perceived by nurses was indeed a significant predictor of job satisfaction.
In line with these findings it is expected that also in this study PO fit will positively influence job satisfaction among nurses. PO fit implies a state of congruence between the values of an employee and the values of an organization. In their study of 2009, Edwards and Cable investigated how exactly value congruence leads to positive outcomes among which job satisfaction. According to them, value congruence leads to job satisfaction because of three constructs, which are communication, attraction and trust. First of all, value congruence enhances communication within organizations, because when employees share the same values they have similar ideas of what is important and they will interpret events in the same way. Communication, in turn, is an important predictor of job satisfaction, because it provides information and helps employees with achieving particular goals, which will lead to increased feelings of satisfaction. Secondly, attraction, which refers to mutual liking and friendship among employees, is enhanced by value congruence since agreement on what is important promotes harmony and cooperation and avoids conflict. When employees experience mutual
liking and friendship this will lead to positive feelings about the job which increases job satisfaction. Finally, shared values between employees and the organization promotes employees’ trust in the organization and trust enhances job satisfaction because it implies for example, that other members of the organization act on one’s behalf (Edwards and Cable, 2009). Since especially trust and communication have found to be strongly related job satisfaction among nurses (Blegen, 1993; Velez & Strom, 2012) it is expected that PO fit will have a positive effect on job satisfaction among nurses. Therefore, the following is hypothesized:
Hypothesis 2: There is a positive relationship between person-organization fit and job satisfaction among nurses.
2.5. Person-Team fit
The third dimension of PE fit is called Person-Team fit (PT fit). This type of fit emphasizes the compatibility between an individual and the team he or she works with (Judge & Ferris, 1992). In other words, PT fit exists when an employee is interpersonally compatible with other members of his or her team, based on different criteria such as values or KSA’s (Adkins et al., 1996; Kristof-Brown et al., 2005; Kristof-Brown & Stevens, 2001). In general, it has been found that PT fit can enhance group performance and compliance with group norms. Furthermore, this dimension of fit leads to more effective interaction with colleagues (Feldman, 1984; Werbel & Johnson, 2001).
Although PT fit has not gained as much attention as the previous two forms of PE fit, the relationship between PT fit and job satisfaction in general has been investigated before. In their meta-analysis, Kristof-Brown et al. (2005) found that PT fit has a positive moderate correlation with job satisfaction. Furthermore, although no evidence has been found for the
direct relationship between PT fit and job satisfaction among nurses, Adams and Bond (2000) found high correlations between of job satisfaction among nurses and cohesion of the ward nursing team, which highlight the importance of fitting into a team. In addition, several studies support that the quality of teamwork is related to job satisfaction of nurses (Ernst, Messmer, Franco & Gonzalez, 2004; Foley, Kee, Minick, Harvey & Jennings, 2002).
In this study it is assumed that PT fit is positively related to job satisfaction. Within the dimension of PT fit, authors often refer to the distinction between a complementary and a supplementary fit. A complementary fit occurs when the employee possesses KSA’s that supplement the team, because these KSA’s are not possessed by other members of the team. A supplementary fit, instead, occurs when an employee possesses characteristics that are similar to the characteristics of his or her team (Muchinsky & Monahan, 1987; Young Seong & Kristof-Brown, 2012). This study will focus on supplementary fit. In line with the supplementary fit, PT fit exists when there are shared values between an individual and his or her team. Sharing the same values implies a degree of similarity which leads to attraction and positive feelings among which sense of belonging (Byrne, 1971). Such attraction is also described by the earlier mentioned attraction-selection-attrition theory of Schneider, which theory confirms the tendency of employees to be attracted to teams whose members share values similar to their own values. In their study among nurses, Winter-Collins and McDaniel (2000) found that sense of belonging is positively related to nurses’ job satisfaction. Based on this argument and the previously found results on the relationship between PT fit and Job satisfaction, it is expected that a strong fit between a nurse and his or her team will have a positive effect on job satisfaction, and therefore the following is hypothesized:
Hypothesis 3: There is a positive relationship between Person-team fit and job satisfaction among nurses.
Despite the fact it is hypothesized that all three dimensions of PE fit will be related to job satisfaction among nurses, this does not mean that all dimensions are expected to equally contribute to the extent to which nurses are satisfied with their jobs. Although the three types of PE fit are often found to be related, all types are distinct dimensions. Each dimension of fit reflects different aspects of the work environment and each dimensions of fit has a unique impact on outcome variables (Cable & DeRue, 2002; Kristof-Brown, 2000; Kristof-Brown et al., 2005). This means, for instance, that when an employee perceives to fit with his or her job, this does not automatically imply that he or she also perceives to fit with the organization or team.
In this study it is assumed that PJ fit will be most strongly related to job satisfaction among nurses. First of all, this is expected because of a study by Smerek and Perterson (2007), who found that the work itself is the strongest predictor of job satisfaction. The assumption is further supported by previous research that showed that “job-related constructs are most strongly associated with attitudes about the job, whereas organization-related constructs are more closely related to organizational attitudes” (Kristof-Brown et al., 2005, p. 294; Shore & Martin, 1989). Since PJ fit is a job-related construct and job satisfaction can be seen as an attitude about ones’ job, PJ fit is expected to be stronger related to job satisfaction than PO or PT fit.
Furthermore, various studies investigated the predictors of job satisfaction among nurses and reported that job satisfaction among nurses was predicted by multiple variables. However, variables that are often reported in these studies include autonomy, job stress and work schedule (Blegen, 1993; Bratt et al., 2000; Leveck & Jones, 1996; Shader et al, 2001), which are all variables that can be linked to the job dimension of PE fit. Considering these theoretical arguments, the following is hypothesized:
Hypothesis 4: Compared to Person-organization fit and Person-team fit, Person-job fit will be stronger related to job satisfaction.
.
2.6. Social Support
Social support can be defined as “information leading the subject to believe that he is cared for and loved, esteemed, and a member of a network of mutual obligations” (Cobb, 1976, p. 300), and is concerned with actions of others that are intended to be helpful (Deelstra et al. 2003). The concept of social support is very broad, since it can have multiple forms among which an advice, a sense of belongingness and assistance with tasks (Harris, Winskowski & Engdahl, 2007). Although social support can have multiple forms, this study is restricted to workplace social support. Workplace social support consists of task support, career mentoring, coaching and collegial social support and is particularly provided by colleagues and supervisors (Harris et al., 2007). Considering the work environment of nurses and their close cooperation with both supervisors and colleagues, this study focuses on both suppliers of workplace social support.
Over the past couple of years, workplace social support has received increased attention and by now, many studies have proven that social support has a significant influence on the health and well-being of workers. It has been found that a lower level of social support is associated with a higher prevalence of diseases. Furthermore, social support plays an important role in coping with stress, which is often referred to as the stress-buffering role of social support (Viswesvaran, Sanchez & Fisher, 1999). This stress-buffering role implies that social support reduces or buffers stress by decreasing the harmful effects of stress on health outcomes (Standsfeld, 2005).
The relationship between social support at work and job satisfaction has also been investigated before. In 1964, Vroom already identified social support as a predictor of job satisfaction. Since then, the majority of the studies regarding this topic have been reporting that a higher level of social support is related to a higher level of job satisfaction among workers (Decker, 1985; Harris, Mortizen, Robitschek, Imhoff, & Lynch, 2001). More specifically, in a study conducted by Harris, Winskowski and Engdahl (2007), social support accounted for 17,6% of the variance of job satisfaction. In the nursing literature, social support is also found to be an important factor in determining job satisfaction among nurses (Fagin, Brown, Bartlett, Leary & Carson, 1995; Bradley & Cartwright, 2002). According to AbuAlRub (2004) social support can enhance the quality of care and decrease the level of stress among nurses which is likely to result in higher rates of nurse retention.
The moderating role of social support has rarely been reported in the nursing literature. Moreover, no results have been found regarding the moderating effect of social support on the relationship between PE fit and job satisfaction. In this study, it is expected that social support will moderate the relationship between the different dimensions of PE fit and job satisfaction, so that this relation will be stronger when more social support is received. When employees perceive social support, this will lead to positive outcomes. Because of the stress-buffering role of social support, employees that receive social support are better able to cope with stress and experience less detrimental outcomes of stress. Consequently, these employees will attach more positive feelings and motives to their work-environment. Since job satisfaction can be described as “a pleasurable or positive emotional state resulting from the appraisal of one’s job or job experiences” (Locke, 1976, p. 1304), the positive feelings and motives can be defined as job satisfaction. In contrast, employees that receive low levels of social support will perceive more stress and strain and are therefore more likely to associate their work-environment with more negative feelings and motives instead of positive ones. As such, it is
possible that, compared to employees that receive a lot of support, for employees that receive lower levels of social support a fit with the work environment will lead to less job satisfaction, even when there is a high degree of value congruence.
Hypothesis 5a: The positive relationship between person-job fit and job satisfaction is moderated by social support, so that this relationship is stronger for higher values of social support.
Hypothesis 5b: The positive relationship between person-organization fit and job satisfaction is moderated by social support, so that this relationship is stronger for higher values of social support.
Hypothesis 5c: The positive relationship between person-team fit and job satisfaction is moderated by social support, so that this relationship is stronger for higher values of social support.
2.7. Intrinsic Motivation
Motivation is a concept that is often used to explain human behavior; it provides a reason for why someone acts in a certain way (Elliott & Dweck, 1988). In the construct of motivation, two types can be distinguished, which are called extrinsic motivation and intrinsic motivation. According to Ryan and Deci (2000), extrinsic motivation refers to “the performance of an activity in order to attain some separable outcome” (p. 71), whereas intrinsic motivation refers to “doing an activity for the inherent satisfaction of the activity itself” (p. 71). In other words, where someone who is extrinsically motivated will perform a task because he or she will be
rewarded afterwards, someone who is intrinsically motivated will perform a task just because he or she likes it.
Starting with an investigation of animal behavior, the concept of intrinsic motivation has been studied since the 1970’s. Previous studies found that intrinsic motivation was positively related to job satisfaction and therefore can enhance the level of job satisfaction among employees (Luo, 1999; Masvaure, Ruggunan & Maharaj, 2014). Compared to employees that are not intrinsically motivated, intrinsically motivated employees are more involved with their jobs and see more meaning in their work. Furthermore, they are more persistent, more self-driven and more interested in their jobs (Ryan & Deci, 2000; Dysvik & Kuvaas, 2011; Gagne & Deci 2005).
If employees see more meaning, and are more interested and involved in their work, it is expected that they attach more value to a fit with the environment, compared to employees that are not or less intrinsically motivated. This is expected because a fit with the environment leads to positive work outcomes such as job performance, which is valued by employees that are more involved and see more meaning in their work, because for them it is important that the work is performed well and leads to the desired results (Edwards, 2007). As mentioned before, the Affect Theory states that the extent to which an employee values a facet of work will determine how satisfied he or she becomes when this facet is met. More specifically, when an employee values a certain facet, his or her satisfaction will be more greatly impacted when this facet is met, compared to someone that does not value that specific facet. In line with this theory, when there is a fit with the environment, this will lead to more job satisfaction among nurses that are intrinsically motivated than among nurses that are less or not intrinsically motivated, because this fit is more valued by intrinsically motivated nurses. Based on these argumentation, the following is hypothesized:
Hypothesis 6a: The positive relationship between person-job fit and job satisfaction is moderated by intrinsic motivation, so that this relationship is stronger for higher values of intrinsic motivation.
Hypothesis 6b: The positive relationship between person-organization fit and job satisfaction is moderated by intrinsic motivation, so that this relationship is stronger for higher values of intrinsic motivation.
Hypothesis 6c: The positive relationship between Person-team fit and job satisfaction is moderated by intrinsic motivation, so that this relationship is stronger for higher values of intrinsic motivation.
3. Method
In this chapter the research methodology utilized in this study is explained. First, the overall research design and procedure are described. Then, the quantitative part of this study is discussed, in which the sample, measures and statistical procedure are explained. Afterwards, the qualitative part of the study is described, in which the procedure, the sample and the analysis method are explained.
3.1. Overall research design and procedure
This study aims to contribute to the literature by investigating the relationship between three different dimensions of person-environment fit, which are person-job fit, person-organization fit and person-team fit (independent variables) and job satisfaction (dependent variable) among nurses. In addition, this study also looks at intrinsic motivation and social support as
possible moderating variables on the relation between the different dimensions of PE fit and job satisfaction.
In this study, a mixed method was used to test the hypotheses and answer the research question. First of all, quantitative data was collected by using an online survey that was made with Qualtrics, which is a well-known survey design program. Using a survey provides the ability to do quantitative analysis and to include many people in a short period of data collection. In addition, in order to gather more in-depth insights in the attitudes and thoughts of the respondents, interviews by telephone and mail and were used to collect qualitative data. Interviews are suitable for getting the story behind experiences of respondents and therefore an interview may be useful as follow-up to questionnaires to further investigate the responses (McNamara, 1999).
For both the quantitative and the qualitative part of the research, the population of interest consisted of Dutch nurses. Currently, in the Netherlands more than 160.000 physical and mental health nurses are registered (Berends, 2014). For this large population the sampling frame was unknown and therefore non-probability convenience sampling was used to collect the data. In order to make the quantitative sample as representative as possible, both mental health nurses and physical health nurses from a broad variety of healthcare organizations were approached.
During both the quantitative and qualitative data collection ethical standards were pursued. The respondents were informed about the aim of the research in advance and anonymity was provided. Furthermore, before the interviews took place, the respondents were asked for permission to record the interview.
3.2. Quantitative procedure
As mentioned above, for this study a survey was constructed and used to collect the quantitative data. In total, the survey consisted of 34 questions and the time required to complete it was approximately five minutes (see Appendix 1 for the entire questionnaire). The respondents were not forced to answer the questions, since forcing respondents to answer questions is likely to result in cancelling the survey earlier and more often (Décieux, Mergener, Neufang & Sischka, 2015; O’Neil, Penrod & Bornstein, 2003).
The data collection started on the 27th of February and ended ten weeks later on the 11th
of May. The survey was distributed online by using mail, LinkedIn and Facebook. Most of the nurses were directly approached and asked by the researcher to fill in the survey. However, some of the nurses that were approached, were also asked to forward the link of the survey to their colleagues (snowball sampling). Due to this individually approach of respondents, no reminders were send during the period of data collection.
3.2.1. Quantitative sample
In total, 497 participants filled out the survey. After cleaning the data in SPSS, the sample consisted of 400 respondents, among which 283 females (70,7%) and 72 males (18%). 45 respondents (11,3%) did not fill in their gender. The respondents worked at 32 different organizations (see Appendix 2 for an overview of the different organizations). Among the respondents that filled out the survey, 265 respondents worked as physical health nurse (67,3%) and 129 worked as mental health nurse (32,3%). The average age of the respondents was 43 years (SD = 13.142; range 19-67). 33,8 % of the respondents succeeded in intermediate vocational education (MBO) and 61,8 % of the respondents succeeded at least in higher vocational education (HBO, WO Bachelor or WO Master). Furthermore, the average
organizational tenure was 14,93 years (SD = 11.075; range = 0.20-46) and the average working hours per week was 29,56 (SD = 6.049; range = 8-40).
3.2.2. Measures
Different scales were used in order to test the different hypotheses. The internal consistency of each scale was tested by using Cronbach’s Alpha (α). In line with the scientific norm, a threshold of 0.7 was used (Pallant, 2007). A seven-point Likert-type scale was used for all items, ranging from 1 (strongly disagree) to 7 (strongly agree). Furthermore, to ensure better understanding, all items were directly translated into Dutch.
The three dimensions of person-environment fit were measured using twelve items of Cable & DeRue (2002). Person-job fit was measured by six items (α = 0.907). An example item is “There is a good fit between what my job offers me and what I am looking for in a job”. To measure Person-organization fit (α = 0.927) three items were used among which “The things I value in life are very similar to the things that my organization values”.
Person-Team fit (α = 0.902) was also assessed by three items. An example item of this third type of
fit is “My teams’ values and culture provide a good fit with the things that I value in life”.
Job satisfaction was assessed by using the three-items measure of Cammann, Fichman,
Jenkins and Klesh (1979). An example item is “All in all, I am satisfied with my job” (α = 0.737).
Intrinsic motivation was assessed by using Grant’s (2008) four-item measure, including
“I am motivated to do my work because it is fun” (α = 0.899).
Social Support was measured by using eight items of Schreurs, Van Emmerik, Günter
and Germeys (2012). Example items are “My colleagues show that they like me” and “My supervisor show that he or she likes me” (α = 0.860).
In order to test the goodness of the scales, principal axis factoring analyses (PAF) were conducted. Except the social support scale, all scales had only one eigenvalue above one, which means only one component was extracted. Therefore, these solutions could not be rotated. For the social support scale, the Kaiser-Meyer-Olkin measure verified the sampling adequacy for the analysis, KMO = 0.807. Bartlett’s test of sphericity x2 (28) = 2080.160, p < 0.001, indicated that correlations between items were sufficiently large for PAF. An initial analysis was run to obtain eigenvalues for each component in the data. Two components had eigenvalues over Kaiser’s criterion of 1 and in combination explained 74,36% of the variance. In agreement with Kaiser’s criterion, examination of the scree plot revealed a levelling off after the second factor. Thus, two factors were retained and rotated with a Oblimin with Kaiser normalization rotation. Table 1 shows the factor loadings after rotation. The items that cluster on the same factors suggest that factor 1 represents perceived social supervisor support, and factor 2 perceived social colleague support. As can be seen in the table, the items show no cross-loadings. Since there was a clear distinction between the factors, the items were divided into two different scales that were labeled colleague social support (items 19, 20, 21 & 22; α = 0.817) and supervisor social support (items 23, 24, 25 & 26; α = 0.929) respectively.
Table 1: Factor analysis results of social support items
Rotated Factor Loadings Supervisor
Support
Colleague Support 19. Mijn collega’s laten merken dat ze me aardig vinden -0.062 0.867
20. Mijn collega’s laten zien dat ze de manier waarop ik mijn werk doe waarderen
-0.038 0.885
21. Mijn collega’s geven me advies over hoe ik dingen moet aanpakken
0.109 0.680
22. Mijn collega’s helpen me als het nodig is 0.009 0.799
dingen moet aanpakken
24. Mijn leidinggevende helpt me als het nodig is 0.932 0.012
25. Mijn leidinggevende laat zien dat hij/zij de manier waarop ik mijn werk doe waardeert
0.931 -0.005 26. Mijn leidinggevende laat merken dat hij/zij me aardig
vindt
0.868 0.014
3.2.3. Control Variables
Literature on job satisfaction among nurses have showed consistent significant relationships with age, gender (1= male, 2= female), level of education (1= middelbare school, 2= MBO, 3= HBO, 4= universiteit (Bachelor), 5= universiteit (master), 6= PhD, 7= Anders) organizational tenure in years and weekly working hours (Blegen, 1993; Pfeffer & O’Reilly, 1989, Lu et al. 2002). Therefore, these variables have been included in this study as well. Furthermore, both mental health nurses and physical health nurses participated in this study. Since these two types of nurses perform different tasks, their type of specialization (1= physical health nurse, 2 = mental health nurse) was also included as a control variable.
3.2.4. Statistical procedure
To analyze the data and test the hypotheses, the Statistical Software Package for Social Sciences (SPSS) was used. After downloading the data file, the data was checked for missing values. The respondents with more than six missing answers were deleted and all the other missing values were coded as missing. In this way, only the items that were answered were taken into account during the analysis. The original data file consisted of 497 respondents. After removing the respondents with too many missing values, the data file consisted of 400 respondents. In addition, a check of frequencies was run to examine whether the data contained deviant values. However, no deviant values were found.
counter-scales, the Cronbach’s alpha was computed. All scales had a Cronbach’s alpha > 0.7, which means that the level of internal consistency was high. The corrected item-total collations were all above 0.30 which indicated that all the items had a good correlation with the total score of the associated scales and since none of the items would substantially affect reliability if they were deleted, no items had to be removed.
After measuring scale reliability, the different scale means were computed in order to test the hypotheses. To check the normality of the distributions of the scales, the Skewness and Kurtosis values of the scales were analyzed. The Skewness value provides information about the asymmetry of the distribution, whereas the Kurtosis value provides information about the sharpness of the peak of the distribution (Field, 2013). As exhibited in Table 2, none of the scales were normally distributed. For PO fit, the distribution was moderately skewed. For PT fit, PJ fit, colleague social support, supervisor social support, intrinsic motivation and job satisfaction the distribution was highly skewed. All values were negative, which indicates that there are a lot of high scores in the distributions. Furthermore, the Kurtosis values of all scales except person-organization fit were highly positive, which indicates the peaks of the distributions of these scales were pointy and heavy-tailed (Field, 2013). However, “in large data sets typical in public health research, most statistical methods rely on the Central Limit Theorem, which states that the average of a large number of independent random variables is approximately normally distributed around the true population mean” (Lumley, Diehr, Emerson & Chen, 2002, p.152). Therefore, if the sample is large enough, the shape of the data does not affect significance tests (Field, 2013). Considering the large size of the sample used in this study (N=400), all variables were used and none of the variables was transformed.
Table 2: Skewness & Kurtosis
Stat. Statistic Std. Error Statistic Std. Error
P-J fit 399 -1.195 0.122 1.267 0.244
P-O fit 399 -0.775 0.122 -0.367 0.244
P-T fit 399 -1.229 0.122 1.704 0.244
Colleague social support 397 -1.879 0.122 7.098 0.244 Supervisor social support 397 -1.017 0.122 0.507 0.244
Intrinsic motivation 395 -1.079 0.123 2.333 0.245
Job satisfaction 393 -1.201 0.123 1.916 0.246
In order to determine whether relationships between the different variables were present, a correlation analysis was done. In Table 3 the mean, standard deviation and correlations of the variables of interest (including control variables) are provided. Then, hierarchical multiple regression was performed to investigate the ability of PO fit, PT fit and PJ fit to predict levels of job satisfaction, after controlling for gender, age, level of education, organizational tenure, working hours per week and job specialization. Finally, in order to test the moderating effects of intrinsic motivation and social support on the relationship between the different dimensions of PE fit and job satisfaction another hierarchical multiple regression was performed. After entering the control variables gender, age, level of education, organizational tenure, working hours per week and job specialization, the independent, moderation and interaction variables (multiplying the standardized independent and moderation variables) were added to the equation.
3.3. Qualitative Procedure
At the end of the survey the respondents were asked whether they were open to participate in an additional interview. Out of the 68 respondents that indicated they wanted to participate, twenty were randomly selected for telephone interviews and received an invitation by mail.
Seven of them reacted and were interviewed by phone. Furthermore, the other 48 respondents were asked to answer additional open questions by mail. In total, twelve of them reacted by providing their answers on the questions.
The main purpose of the interviews was to gain more insights in the answers given in the survey to be able to interpret the results of the quantitative part of the study. Therefore, the interview questions were mainly based on the survey. For all respondents, the same interview protocol was used. After the introduction, the interview started with questions to collect some demographic information, such as age and years worked for the organization. Next, a question was asked to identify the reason why someone had ever chosen to become a nurse. Then, the first part of the interview started in which the interviewee was asked about his or her perceived fit with the organization, the job and his or her team. The main aim of this part of the interview was to discover the reasons behind their perceptions of fit. It was focused on the reasons why a nurse thought he or she fitted. The second part of the interview was mainly focused on job satisfaction. The interviewees were asked whether they were satisfied with their jobs and what needed to be changed in order to become more satisfied.
The telephone interviews were semi-structured, all in Dutch and lasted between ten and twenty minutes on average. The interviews have been held in random order and have all been recorded and transcribed in Microsoft Word. Since the privacy of all respondents was guaranteed by conducting anonymous interviews, interviewee numbers will be used (i.e. interviewee 1 – 7; see Appendix 4) instead of names.
The other 48 respondents that indicated they wanted to participate in an additional interview received the interview questions by mail. The interviews were all structured: every respondent received the same mail with the same eleven questions. All correspondence was in Dutch. In total, twelve respondents reacted by providing their answers on the questions. The questions were all open-ended and were similar to the questions of the interviews by phone
(see Appendix 3 for an overview of the interview questions). The interviews have all been transcribed in Microsoft Word. Since the privacy of all respondents was guaranteed by conducting anonymous interviews, interviewee numbers will be used (i.e. interviewee 8 – 19; see Appendix 4) instead of names.
3.3.1. Qualitative Sample
Of the respondents that participated in the additional interviews, two were male (10,5%) and seventeen were female (89,5%). The average age was 51 years (SD = 12.943; range = 23-64) and the average organizational tenure was 25.7 years (SD = 9.775; range = 1-39). Among the respondents that participated in the interviews, thirteen respondents worked as physical health nurse (68,4%) and five worked as mental health nurse (26,3%). For one nurse the job specialization was unknown (see Appendix 4 for an overview of the respondents).
3.3.2. Qualitative Analysis Procedure
After all the interviews were transcribed in Microsoft Word, the data was systematically analyzed and coded. During this analysis process, the data of both types of interviews (telephone and mail) were analyzed simultaneously and with the same approach. In other words: no differentiation between the data collected by telephone interviews and the data collected by mail interviews was made. For the coding process a more deductive approach was used. However, while reading the transcriptions, the researcher was also open to and alert for new emerging common themes and patterns. As a first step, based on the variables used in the survey and the interview questions a start list of starter codes was created. The composed codes were all related to one of the variables that were used in the survey. Examples of starter codes that were used are: Reason to become a nurse, Motivation for the job, Most important
aspect of the job, Person-job fit, Peron-organization fit, Person-team fit, Job satisfaction and Change. During the coding process, meaningful sentences of the transcriptions were
identified and assigned to the different codes. Next, all parts with the same code were grouped together and analyzed again. Based on this last analysis a list of most important quotes per theme was conducted and an overview of the most important topics was created.
4. Results
In this chapter the results of both the quantitative and qualitative analyses are discussed. The first paragraph focuses on the quantitative results, at which the correlation analysis, the direct effects and the moderating effects will be described. In the second paragraph the qualitative results are discussed, by describing the most important information collected from the interviews.
4.1. Quantitative results 4.1.1. Correlation analysis
In Table 3 an overview of the descriptive statistics, correlations and scale reliabilities is presented. Among the correlations that were found, some appeared in line with the expectations derived from the literature. First of all, statistically significant positive correlations were found between the different dimensions of fit. However, these correlations ranged from low to moderate, which implies that the respondents were able to make a distinction between the three dimensions of fit.
Based on the results of the correlation analysis, it can be concluded that there is a significant relationship between all three dimensions of PE fit and job satisfaction. Correlations appeared to be the strongest for PJ fit. PO fit is moderately correlated to job satisfaction. Against expectations, the correlation between and PT fit and job satisfaction is
positive but weak. Furthermore, the moderating variables intrinsic motivation and both types of social support significantly correlate with the independent variables as well as with the dependent variable.
Moreover, the correlation matrix shows that there is no significant correlation between the control variables and the dependent variable job satisfaction, except for working hours per week and job specialization. However, these last two correlations were significant but weak. This indicates that the control variables and the dependent variable are hardly related and thus also indicates that the variance in job satisfaction is not likely to be explained by these variables.
Beside these expected correlations, some unexpected however interesting correlations presented themselves. As can be found in Table 3, job specialization (1 = physical health nurse, 2 = mental health nurse) negatively correlates with PJ fit, PO fit, intrinsic motivation and job satisfaction. Although these correlations range between weak and moderate, these results indicate the presence of differences between mental health nurses and psychical health nurses with regard to these variables.
Finally, other striking results were that a strong positive correlation was found between intrinsic motivation and job satisfaction, a moderate negative correlation was found between age and PO fit and working hours per week positively correlates with job satisfaction.
Table 3. Means, Standard Deviations & Correlation Mean Std. Dev 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 1.Gender 1.800 0.403 2.Age 43.140 13.142 -0.125* 3.Education 2.830 0.984 -0.140** 0.016 4.Organizational tenure 14.930 11.075 -0.029 0.704** -0.016 5.Hours/week 29.560 6.049 -0.310** -0.174** 0.183** -0.264** 6.Job specialization 1.330 0.479 -0.240** 0.227** 0.157** 0.085 0.147** 7.PJ fit 5.339 1.030 0.067 -0.015 -0.062 0.016 0.062 -0.164** (0.907) 8.PO fit 4.596 1.338 0.037 -0.227** 0.006 -0.166** 0.056 -0.292** 0.446** (0.927) 9.PT fit 5.337 0.940 0.053 0.023 -0.082 0.063 -0.06 0.029 0.278** 0.153** (0.902)
10.Colleague social support 5.806 0.758 0.041 -0.095 -0.091 0.012 0.072 -0.05 0.249** 0.142** 0.392** (0.817)
11.Supervisor social support 5.042 1.368 0.019 -0.137** -0.096 -0.098 0.099* -0.087 0.406** 0.391** 0.163** 0.359** (0.929)
12.Intrinsic motivation 5.809 0.884 0.029 -0.031 -0.045 -0.019 0.039 -0.208** 0.547** 0.399** 0.174** 0.282** 0.416** (0.899)
13.Job satisfaction 5.768 0.889 0.017 -0.071 -0.005 -0.059 0.108* -0.137** 0.637** 0.482** 0.133** 0.159** 0.381** 0.695** (0.737)
* Correlation is significant at the 0.05 level (2-tailed).
4.1.2. Direct effects
Hypotheses 1, 2 and 3 proposed that person-job fit, person organization-fit and person-team fit would be positively related to job satisfaction. Hypothesis 4 proposed that compared to person-organization fit and person-team fit, person-job fit would be stronger related to job satisfaction.
In order to test these hypotheses hierarchical multiple regression was performed (see Table 4). In the first step of the regression analysis, the following six predictors were entered: gender, age, level of education, organizational tenure (in years), the amount of working hours per week and job specialization. This model was not statistically significant F(6,336) = 1.792;
p > 0.05. Therefore, it can be concluded that most variance of the variable job satisfaction is
explained by other factors, which was also indicated by the results of the correlation analysis. After the entry of PJ fit, PO fit and PT fit at step 2 the total variance explained by the model as a whole was 47,4% (F(9,333) = 33.306, p < 0.001). The introduction of PJ fit, PO fit and PT fit explained an additional 44,3% of the variance in job satisfaction after controlling for gender, age, level of education, organizational tenure (in years), the amount of working hours per week and job specialization (R2 change = 0.443; F(3,333) = 93.379, p < 0.001).
In the overall model, two out of three predictor variables were significant, with job fit recording a higher regression coefficient (ß = 0.541, p < 0.001) than person-organization fit (ß = 0,264, p < 0.001). In other words, if person’s person-job fit increases for one, his or her job satisfaction will increase for 0.541. On the other hand, if person’s person-organization fit increases for one, his or her job satisfaction will increase for 0.264.
Therefore, hypothesis 1 and hypothesis 2 are accepted: there is a positive relation between person-job fit and job satisfaction among nurses as well as there is a positive relation between person-organization fit and job satisfaction among nurses. This means that the more a nurse fits with the job or the organization, the more satisfied with the job he or she will be.
Furthermore, since there is no statistical significant relation between person-team fit and job satisfaction among nurses, hypothesis 3 is rejected. Finally, hypothesis 4 is accepted as well: compared to person-organization fit and person-team fit, person-job fit has indeed a stronger effect on job satisfaction. Thus, a good fit with the job has a stronger effect on job satisfaction among nurses than a good fit with the organization or the team have.
Table 4: Hierarchical Regression Model of Job Satisfaction
R R2 R2 Change B Std. Err ß t Step 1 0.176 0.031 Gender 0.076 0.129 0.034 0.584 Age 0.000 0.005 0.000 0.002 Education -0.006 0.051 -0.007 -0.120 Organizational Tenure 0.000 0.007 -0.003 -0.043 Hours/Week 0.023 0.009 0.154* 2.548 Job specialization -0.211 0.110 -0.110 -1.918 Step 2 0.688 0.474** 0.443** Gender -0.002 0.096 -0.001 -0.018 Age 0.003 0.004 0.038 0.638 Education -0.003 0.038 -0.004 -0.092 Organizational Tenure -0.001 0.005 -0.013 -0.213 Hours/Week 0.009 0.007 0.061 1.341 Job specialization 0.061 0.084 0.032 0.726 P-J fit 0.473 0.040 0.541** 11.768 P-O fit 0.179 0.031 0.264** 5.676 P-T fit -0.031 0.040 -0.033 -0.781 Statistical significance: *p <0.05; **p <0.001. 4.1.3. Moderating effects
Hypotheses 5a, 5b and 5c proposed that the direct relation between job fit, person-organization fit, person-team fit and job satisfaction would be moderated by social support, so that these direct relationships are stronger for higher values of social support. Furthermore, hypotheses 6a, 6b and 6c proposed that the direct relation between job fit,
person-organization fit, person-team fit and job satisfaction would be moderated by intrinsic motivation, so that these direct relationships are stronger for higher values of intrinsic motivation.
In order to test these hypotheses hierarchical multiple regression was performed (see Table 5). In the first step of the regression analysis, the following six predictors were entered: gender, age, level of education, organizational tenure (in years), the amount of working hours per week and job specialization. This model was not statistically significant (F(6,335) = 1.835, p > 0.05). After the entry of the standardized variables PJ fit, PO fit, PT fit, intrinsic motivation, colleague social support, supervisor social support and the nine standardized interaction terms at step 2, the total variance explained by the model as a whole was 65,7% (F(21,320) = 29.129, p < 0.001). The introduction of the independent, moderation and interaction variables explained an additional 62,5% of the variance in job satisfaction after controlling for gender, age, level of education, organizational tenure (in years), the amount of working hours per week and job specialization (R2 change = 0.625; F(15,320) = 38.804, p <
0.001).
Hypothesis 5a predicted that the positive relationship between PJ fit and job satisfaction would be moderated by social support, so that this relationship would be stronger for higher values of social support. Results (see Table 5) showed that the regression coefficient for the interaction variable PJ fit x Colleague social support is 0,186 and is statistically different from zero, t(320) = 4.218, p < 0.001. Figure 2 provides a graphical presentation of the moderating effect of colleague social support on the relationship between PJ fit and job satisfaction. In this figure it can be seen that the lines do intersect and both have a positive slope, however the line for high values of colleague social support is steeper than the line for low values of colleague social support, which implies that the effect of PJ fit on job satisfaction is stronger for higher values of social colleague support.
Table 5: Moderating effects of Intrinsic motivation and Social support R R2 R2 Change B Std. Err ß t Step 1 0.178 0.032 Gender 0.070 0.130 0.032 0.543 Age 0.000 0.005 -0.003 -0.043 Education -0.004 -0.051 -0.004 -0.067 Organizational Tenure 0.000 0.007 -0.003 -0.039 Hours/Week 0.023 0.009 0.154* 2.536 Job specialization -0.218 0.110 -0.113* -1.974 Step 2 0.810 0.657** 0.625** Gender 0.001 0.080 0.000 0.011 Age -0.002 0.003 -0.032 -0.647 Education -0.014 0.032 -0.016 -0.453 Organizational Tenure 0.002 0.004 0.023 0.461 Hours/Week 0.006 0.006 0.041 1.070 Job specialization 0.121 0.071 0.063 1.705 P-J fit 0.428 0.236 0.475 1.810 P-O fit 0.180 0.037 0.199*** 4.808 P-T fit 0.015 0.035 0.017 0.436 Intrinsic motivation 0.438 0.041 0.481*** 10.693
Col. social support 0.002 0.039 0.002 0.051
Sup. social support -0.035 0.040 -0.039 -0.872
PJ fit x In. mot. -0.024 0.041 -0.149 -0.589
PJ fit x Co. so. sup. 0.118 0.028 0.186*** 4.218
PJ fit x Su. So. Sup. -0.056 0.033 -0.084 -1.671
PO fit x In. mot. 0.072 0.033 0.096* 2.159
PO fit x Co. so. Sup. -0.039 0.036 -0.045 -1.097
PO fit x Su. So. Sup. -0.076 0.035 -0.101* -2.195
PT fit x In. mot. -0.018 0.037 -0.025 -0.475
PT fit x Co. so. Sup. 0.077 0.028 0.132** 2.742
PT fit x Su. So. Sup. -0.077 0.037 -0.093* -2.077
Statistical significance: *p <0.05; **p <0.01;***p <0.001.
Furthermore, the results of the regression show a negative but insignificant regression coefficient for the interaction variable PJ fit x Supervisor social support (ß = 0.084, t(320) = -1.671, p > 0,05). Hypothesis 5a is therefore partially supported: it can be concluded that the
effect of PJ fit on job satisfaction among nurses depends on colleague social support. More specifically, the relationship between PJ fit and job satisfaction among nurses is positively moderated by colleague social support, so that this relationship is stronger for higher values of social colleague support.
Figure 2: Moderating effect of colleague social support on the PJ fit – job satisfaction relationship
Hypothesis 5b predicted that the positive relationship between PO fit and job satisfaction would be moderated by social support, so that this relationship would be stronger for higher values of social support. Results (see Table 5) showed that the regression coefficient for the interaction variable PO fit x Colleague social support is 0.045 (t(320) = -1.097, p > 0.05). However, this coefficient was not significant.
Furthermore, the results of the regression analysis show a negative but significant regression coefficient for PO fit x Supervisor social support (ß = -0,101, t(320) = -2.195, p < 0.05). Figure 3 provides a graphical presentation of the moderating effect of supervisor social support on the relationship between PO fit and job satisfaction. In this figure it can be seen
that the lines do intersect and that the slopes of both lines are positive. However, the line for low values of supervisor social support steeper than the line for high values of supervisor social support, which implies that the effect of PO fit on job satisfaction among nurses is stronger for lower values of supervisor social support. Thus, although the moderating effect of supervisor social support is significant, the results of the regression analysis regarding the moderating effect of social support in the relationship between PO fit and job satisfaction are not in line with the expectations. More specifically, the direction of the effect was different than expected. In other words: the positive relationship between PO fit and job satisfaction among nurses is moderated by supervisor social support, so that this relationship is weaker instead of stronger for higher values of supervisor social support. Therefore, hypothesis 5b was only partially accepted.
Figure 3: moderating effect of supervisor social support on the PO fit – job satisfaction relationship
Hypothesis 5c predicted that the positive relationship between PT fit and job satisfaction would be moderated by social support, so that this relationship would be stronger
