An investigation of the effects of psychological contract and organization-based self-esteem on organizational commitment in a sample of permanent and contingent workers.

For the mediator effect to be established, four criteria must be fulfilled (Baron & Kenny, 1986). First, the independent and dependent variables must be correlated; second, the independent variable must correlate with the mediating variable; third, the mediating variable must affect the dependent variable; and last, the effect of the independent variable on the dependent variable should be significantly less, in the case of partial mediation, than the direct effect between the two variables.

In both hypothesized relationships, the respective independent-dependent variable relationships were established. However, only Hypothesis 3a was supported. The effect of perceptions of relational obligations on value commitment was reduced considerably when OBSE was introduced to the model. Hypothesis 3b was not supported because there was no significant relationship between perceptions of transactional obligations and OBSE. Therefore, there was not potential for the hypothesized intervening effect.

Qualitative Findings

Participants were also afforded the opportunity to respond to an open-ended question to elaborate on why they believed that the organization did, or did not, fulfill its obligations to them. These data were collected for the firm's management and not coded or analyzed in this study. Twenty-six respondents provided comment. Primarily the comments were affirmative. However, there was a common concern expressed in many of the 26 comments. The primary concern can be summarized in the words of one of the respondents: "As this company grows, it's less like the family that it was a couple of years ago when we were smaller."

Discussion

In this article, we explored the relationships between three important constructs--psychological contract, OBSE, and organizational commitment--as related to contingent workers within the context of a high-tech organization. The findings from this study were mixed, in that many of the hypotheses were supported by the data, but others were not. In this section, we will discuss the various outcomes, followed by implications for management and study limitations.

We first hypothesized that there would be differential ratings of the two obligations of the psychological contract, as stratified by employment status; however, the data did not support these hypotheses. Permanent employees, who were hypothesized to demonstrate greater perceptions of relational contract obligations by the employer, did not indicate a significantly different perception between the two types of obligations. Similarly, contingent workers, who were hypothesized to rate transactional obligations higher than relational, also indicated no difference between the two psychological contract obligations.

One reason for this may be that most of the contingent workers had been assigned to this firm for as long as the majority of the permanent employees had been employed. Therefore, the permanent workers may not have had the time necessary to develop differential perceptions of long-term, relational obligations of the employer. In this way, the contingent and permanent workers may not have seen their employment situations as being different. Additionally, the physical environment of the firm was such that little distinction was made between the permanent and contingent workers. All workers appeared to have been treated the same within the same job categories. Although security badges and name tags were color-coded to indicate security clearance, there was no distinction with regard to employment status.

Although our findings did not lend support to our hypotheses about differential relationships between psychological contract obligations and employment status, they did confirm Sloboda's (1999) assertion that contingent workers were unlikely to distinguish between the relational and transactional obligations of the psychological contract.

We also hypothesized that regardless of employment status, workers perceptions of the relational contract obligations would be more highly related to value commitment than to continuance commitment. We found evidence to support the relational obligations-value commitment hypothesis in absolute terms. Although we did find support for this relationship, the strength of the relationship may be lessened in that relational obligations shared a statistically significant relationship with both commitment factors.

Again, the fact that many of the contingent workers had been assigned to this firm for as long as, or longer than, many of the permanent employees, there may not have been adequate time for the workers to develop differential levels of continuance commitment or expectations of employer obligations. However, although both relationships were found to be significant, we discovered that permanent workers rated their value commitment somewhat higher than did the contingent workers.

We suggested that transactional obligations would be more highly related to continuance than value commitment. This was also found to be true in absolute terms but only marginally. We offer the same rationale for this phenomenon as for the relational obligations and value commitment relationship: Permanent employees were employed no longer in this fledgling firm than contingent workers. Therefore, the differences in perceptions as related to commitment were not manifest in this sample.

Differential relationships, by employment status, were hypothesized between the contract obligations and the dimensions of organizational commitment. We found evidence that the relationship between relational psychological contract obligations and value commitment was greater for permanent than for contingent workers. This suggests that across all workers, the relational obligations, in relation to value commitment, were rated higher by permanent than by contingent workers.

This notion did not hold for the hypothesis that for contingent workers, transactional obligations would be more related to continuance than to value commitment. In fact, the opposite appeared to be true, which indicated that permanent employees saw connections between the obligations and dimensions of commitment more clearly than the contingent workers. A reason for this, stemming from the length-of-service discussion offered earlier, is that contingent workers are not actually employed by the host firm. When the job is finished, they leave for the next organization that has contracted their services. It is possible that contingent workers viewed the psychological contract obligations and commitment dimensions differently because of this. The permanent employees did not have a multiagency relationship (Gallagher & McLean Parks, 2001) as did the contingent workers, which might affect perceptions of obligations as well as levels of commitment to the host organization.

Finally, we suggested that OBSE would mediate the relationships between the psychological contract obligations and the commitment dimensions. Although we found clear evidence that OBSE partially mediated the relationship between relational obligations and value commitment, we did not find a mediating effect for transactional obligations and continuance commitment. In fact, the only relationship in this model that was neither substantive nor significant was that between transactional contract obligations and continuance commitment. It appeared that the workers, regardless of employment status, developed perceptions of relational obligations as well as a level of value commitment to the organization, whereas the more economic, short-term obligations did not seem to have become manifest. This is not totally surprising when we consider the environmental antecedents of OBSE, most of which are affected by management. Contingent workers' value commitment, which is affect-based, is influenced by environmental phenomena (Gallagher & McLean Parks, 2001).

Last, Feldman (2006) suggested that individual difference variables have a moderating influence on the relationship between contingent employment and work-related outcomes. However, we did not find evidence of this phenomenon in our sample. Gender, age, and education were not correlated with any of the variables measured in this study. One reason for this may have been the small sample size.

Implications for Management

Managers who desire to maintain a competitive advantage in their respective industries realize that knowledge rapidly becomes obsolete. One way to mitigate this obsolescence is to employ contingent workers (Szabo & Negyesi, 2005). However, as evidenced in the literature review, as well as in this study, the decision to employ a contingent workforce affects important attitudinal aspects useful in managing workers.

These findings are important for working managers. Although performance outcomes were not considered in this study, there have been linkages established in prior research between all three variables and performance. For example, Kraimer et al. (2005) discovered relationships between psychological contract perceptions and performance; positive relationships between organizational commitment, specifically, affective commitment, and performance were found by Meyer et al. (1989); and OBSE has been shown to moderate performance outcomes (Pierce et al., 1993).