رایگان یا مخاطره آمیز ؟ مقایسه نگرش کارگران در قراردادهای اشتغال انعطاف پذیر و سنتی

This study contributes to research on the impact of different kinds of employment contract on worker attitudes in the context of debates about the changing nature of employment in the 21st century and in particular the emergence of what have been described as either free or precarious workers. Work experiences and attitudes associated with part-time, temporary and multiple contracts are compared with those of workers in single, permanent, full-time jobs. Among a sample of UK pharmacists, few significant differences are found between workers on any of these contracts, either singly or in combination, and those in traditional employment contracts. The role of contract of choice and work orientations as potential mediators was explored. Although both were associated with attitudes, they had only a very minor role as either mediators or moderators. It is concluded that among this sample of professional workers, those on atypical employment contracts report experiences and attitudes that are at least as positive as those of workers in traditional employment contracts.

Employment flexibility has been widely advocated as a means of ensuring the full and efficient use of human resources (Capelli, 1999, Handy, 1989 and Lepak and Snell, 1999). While this might bring benefits to the organization, the consequences for the workforce, and more particularly those who are not seen as valued core employees, may be less positive. Indeed, the pace of change in organizations and their markets has led some observers (see, for example, Capelli, 1999, Smith, 2001a and Smith, 2001b) to suggest that no workers, however valued by their present employer, can feel secure in their employment. In the European Union, the assumed costs to workers of employment flexibility has resulted in legislation to restrict working hours with further legislation proposed to protect the rights of those in temporary employment. A number of writers and researchers have presented a different perspective on flexible employment. Barley and Kunda (2004), based on their research on technicians and engineers have identified what Knell (2000) among others has termed the “free worker” who is able to choose his or her position in the labour market. Building on research in Silicon Valley, Bridges (1995) has highlighted the concept of employability allied to new freedoms permitted by the growth of knowledge work (Leadbetter, 1999). This focus overlaps with the burgeoning body of work concerned with the boundaryless career (Arthur & Rousseau, 1996) and the importance of career self-management (King, 2004). This stream of writing suggests that an increasing number of workers are asserting control over their working lives and choosing where, for whom and on what kind of employment contract they work. As Storey, Salaman, and Platman (2005) note, there is an implicit positive discourse whereby the concepts of enterprise, freedom and independence are applied to self-employment and career self-management. Allied to this development, there has been a growing interest in work–life balance and a view that younger workers, those who have sometimes been classified as Generation X (Conger, 1998 and Crainer and Dearlove, 2000), are less willing to accept what they would perceive as unreasonable demands from their employer. There is also evidence in the USA that the greater demands on life both at work and outside work are causing growing numbers of workers to reconsider their work–life balance (Friedman & Greenhaus, 2000). Pink (2001) has portrayed 21st century America as a nation in which flexible working has freed large numbers of workers to choose the working arrangements that suit them. However not everyone is in a position to become a “free” agent. The assumption in the emerging literature is that employment freedom depends upon being in an area of work that requires distinctive knowledge and expertise and for which there is a high level of demand. It is therefore more likely to apply to those with key professional or technical skills who have some scope to choose a career pattern that suits them and who have the resources and know-how to maintain their employability and their employment-related networks (Marler, Barringer, & Milkowich, 2002). Paradoxically, these are the same workers that organizations are likely to look upon as core workers whom they would like to retain in permanent employment so they may be willing to negotiate reduced hours or other more flexible arrangements to hold on to them (Lee, Hourquet, & MacDermid, 2002). They can be contrasted with low skill workers who may become marginalized as organizations cut back their workforces and whose position in the labour market is likely to be more precarious as a result. Despite the assumption that flexible employment is growing and is here to stay, the standard form of employment, which remains the benchmark against which forms of flexible employment can be judged, is the permanent, full-time job. Among all types of worker, including the potentially “free” knowledge workers, this remains the norm. Indeed, as Auer and Cazes (2003) have highlighted, across OECD countries permanent long-term employment in the same organization remains perhaps surprisingly resilient. Casey and Alach (2004) note that if the permanent, full-time job is the norm, then jobs that depart from this are often viewed as problematic and defined as part of a “secondary” labour market. There may therefore be disadvantages and costs for those workers who are engaged in forms of atypical, flexible employment, even among the potentially “free” professional and knowledge workers. By implication, the more flexible the employment, the greater the potential costs. An aim of the present study is to test this assumption. Employment flexibility can take a variety of forms. An organization may reasonably seek flexibility with respect to the employment, allocation and reward of its workers (Atkinson, 1984). These are sometimes described, from an organizational perspective, as numerical, functional and reward flexibility (Beatson, 1995). Our focus is on the forms of employment contract that facilitate numerical flexibility. From a worker's perspective, the key forms of employment flexibility are likely to concern the hours of employment and the contract of employment. Reflecting this concern, and reinforcing the traditional stereotype of the desirability of “normal”, “primary” employment, it is these atypical arrangements that have been the focus of legislation in the European Union. Feldman (1990), with respect to part-time working, and Connelly and Gallagher (2004) with respect to temporary contracts, have highlighted the need to recognise the heterogeneity of forms that these can take. There is particular scope for variety with respect to hours, since those working part-time may be employed on anything up to 30 h a week in the UK and even up to 35 h a week in Australia (Walsh & Deery, 1999). While it may be assumed that part-timers are employed for shorter hours than those in full-time work, this does not take account of the possibility that they hold second jobs. The possibility that some workers may have multiple jobs therefore needs to be built in to the analysis. Feldman (1990) and Connelly and Gallagher (2004) note that forms of employment flexibility may combine so that workers may be both part-time and temporary. Walsh and Deery (1999) have compared part-time and temporary service workers and showed that they had different backgrounds and reasons for seeking a particular type of contract and that workers' reactions depended partly on how well it fitted their circumstances and values. However there has been little research that has explored the experience of combining forms of employment flexibility including multiple job-holding. The aim of this paper is to explore the reactions of a sample of potentially “free” workers to flexible employment. It seeks to test the competing broad hypotheses that flexible employment brings benefits or costs to such workers compared with those in traditional employment. It does so by focusing on three sets of potentially positive and three potentially negative aspects of the employment experience among workers who are in full-time and part-time jobs, employed on temporary or permanent contracts and working in a single or in multiple jobs. To explore the boundaries of flexibility, we will consider each of these forms of employment flexibility in turn but also explore the impact of combining several forms of flexibility together on employee work experiences and attitudes. We will build in to the analysis the influence of volition and work orientation, both of which have been identified as potential mediating variables. Given the assumed growth of flexible employment as a 21st century phenomenon, this will provide some initial insights into the costs and benefits of such employment. In the next section, we briefly review the relevant literature on these different forms of employment flexibility.

The means, standard deviations and zero-order correlations of the main study variables are presented in Table 1. The mean score is at or above the mid-point of 3.00 on the 5-point scales for all variables. Therefore, overall, the respondents tend to report positive satisfaction, commitment and control but also to experience a degree of stress, work–life conflict and, more particularly, work overload. Table 1 also reveals that the correlations between the different types of flexible employment are relatively low.Hypothesis 1 suggested that multiple-job holding would be associated with higher levels of stress and work–life conflict and lower levels of satisfaction and commitment than those in single permanent jobs. The results in Table 4 show no support for any of these propositions. In other words, contrary to expectations, there is nothing to indicate that in this sample, the experience of multiple job-holding results in attitudes to work that are any different from those in single, full-time permanent jobs. There is one unexpected exception. They report higher levels of control over their work. The reasons for this are unclear unless it is an extension of their control over their employment. Hypothesis 2 suggested that part-time workers would report lower stress and work–life conflict but also lower levels of control and higher levels of work overload. There is some support for this hypothesis in so far as part-time workers do report lower levels of stress and work–life conflict. They also report higher levels of organizational commitment. However, contrary to the hypothesis, they do not report lower levels of control over their work. In short, all the indications are that part-time work is a positive experience among this sample. Hypothesis 3 proposed that workers on temporary contracts will report higher levels of stress and lower levels of control, satisfaction and commitment. The results in Table 4 show that none of these propositions are supported. In addition, temporary workers report significantly less work overload than those in full-time permanent jobs. The three elements in Hypothesis 4a, Hypothesis 4b and Hypothesis 4c all propose that combining forms of employment flexibility will lead to more negative consequences. To test the fourth set of hypotheses, a series of ANCOVAs were conducted. Possible interactions between the different forms of flexibility were examined and findings are shown in Table 5. No significant interactions were found between any of the forms of flexibility, therefore these hypotheses were not supported. Combining forms of employment flexibility does not appear to lead to more negative consequences either than working in a single, full-time permanent job or in a single form of flexible employment.The final hypotheses suggest that being on contract of choice and having different career orientations may either mediate or moderate the relationships between forms of contract flexibility and work attitudes. Mediation effects were examined through extensions of the regressions presented in Table 4, using the method suggested by Baron and Kenny (1986). The hypothesised mediator is entered in a third step of each regression. A mediation effect is indicated if the relationship between the independent and dependent variable is removed or substantially reduced once the mediator is entered into the regression model. Mediation requires two other relationships to be demonstrated. Firstly, the independent variable must be related to the mediator and secondly, the mediator must be related to the dependent variable. The correlations between the independent variables and mediators and between the mediators and the dependent variables are presented in Table 1. Although a number of non-significant results are found in the correlations and in the regressions reported in Table 4, the full test for mediation is still applied as this gives further information about the relationship between the hypothesised mediators and the dependent variables. The findings of the mediation tests are presented in Table 6. For each form of flexibility the beta weight found in Table 4 is presented for comparison with the beta weights found in the models in which the mediators are controlled. There was only very limited support for the role of contract of choice, work–life orientation, helping orientation or control orientation as mediators. The significant association between multiple-job holding and control at work becomes non-significant when a helping orientation and control orientation are controlled but the change in the beta weights is very modest (from .07⁎ to .05 and .06 respectively). Nevertheless, even this small change was found to be significantly greater than zero using the Sobel test (Preacher & Hayes, 2004). There was more evidence of a mediation effect of contract of choice with respect to part-time work. Being on contract of choice partially mediated the relationship between part-time work and both control at work (beta reduced from − .15⁎⁎⁎ to − .11⁎⁎) and stress (beta reduced from − .23⁎⁎⁎ to − .16⁎⁎⁎) and fully mediated the relationship with organizational commitment (beta reduced from .09⁎ to .01). More generally, the effects of introducing these potential mediators are small and sometimes inconsistent. Indeed, it was more often the case that the strength of the associations between the forms of flexibility and outcome variables were strengthened when the hypothesised mediators entered the models, which can often be an indication of interactions.Considering the associations between the intervening variables and the outcome variables, it would appear that contract of choice has the strongest association with the outcome variables. It is strongly and positively related to job satisfaction, control and commitment, and strongly and negatively related to stress, overload and work–life conflict after controlling for a number of other variables. Having a work–life orientation would appear to be negatively related to job satisfaction and organisational commitment, while having a helping orientation is positively related to job satisfaction, control, commitment and a higher experience of overload. A control orientation is positively associated with reports of control and commitment but also with reports of work–life conflict. The same variables were tested for moderation using a further series of ANCOVAs and these findings are presented in Table 7. A number of significant interactions were supported and these are pictured in Fig. 1, Fig. 2 and Fig. 3.The specific hypothesis was that those on their contract of choice would report more positive attitudes. Some moderating effects were found, but they were limited in number. Among multiple-job holders, contract of choice moderated the relationship with commitment. Not being on contract of choice had a stronger negative association with the commitment of single job-holders compared with multiple job-holders. Among those in part-time work, being on contract of choice had no moderating effect on any of the outcomes. Among those on temporary contracts, contract of choice moderated the associations with perceptions of job control and commitment. As with multiple-job holding, not being on contract of choice had more negative consequences for the commitment of permanent than temporary workers. The results with respect to control are more complex. Among temporary workers, being on contract of choice is associated with lower control; the implication is that this is what they want. Among those in permanent jobs, being on contract of choice is associated with much higher control. The contrast, highlighted in Fig. 3, does suggest that this is a major distinguishing feature between those in permanent and temporary jobs. The second specific hypothesis was that those with a stronger orientation towards work–life balance would be more positive than those with a low orientation about flexible employment contracts. There are no significant interactions among those in multiple jobs. Among those in part-time work, there is a significant interaction with job satisfaction. This indicates that among those in full-time jobs, a strong orientation towards work–life balance is associated with slightly lower job satisfaction but among those in part-time jobs, it is associated with much lower job satisfaction. By implication, job satisfaction is much less important and much less experienced among part-time workers who give a high priority to work–life balance. Among those in temporary contracts, there is an interaction between a strong orientation to work–life balance and both stress and control. As Fig. 3 shows, among permanent workers, a strong orientation to work–life balance is associated with slightly lower stress whereas among temporary workers, it is associated with significantly more stress. With respect to control, among those in permanent jobs, a strong orientation towards work–life balance is associated with much lower perceptions of job control while among those in temporary jobs, it is associated with slightly higher control. In summary, these results tend to go against the hypothesis. Those with flexible employment contracts and a high orientation to work–life balance are likely to report lower job satisfaction if they are in multiple jobs and more stress, as well as slightly more control over their work if they are in temporary jobs. The trends for those in permanent jobs are in the opposite direction. In addition to the stated hypotheses, there were also some moderating effects of an orientation towards a helping profession and towards control. A strong orientation towards being part of a helping profession is associated with higher job satisfaction among those in multiple jobs and although the trend is in the same direction for those in single jobs, it is less strong. Among those on temporary contracts, a strong orientation towards helping has little impact on job satisfaction while, as we have just noted, it has a strong positive impact on those in single permanent jobs. A strong control orientation was associated with reduced overload and stress for temporary workers, but not permanent workers. Finally, the control orientation has an association with stress and overload among temporary workers. Those with a strong control orientation in permanent jobs report slightly lower work overload while those in temporary jobs report slightly higher overload. A strong control orientation has little impact on the stress of those in permanent jobs but increases it a little among those in temporary jobs. Put another way, a low control orientation is associated with lower overload and stress among temporary workers but not among permanent workers. Since there are no differences in reported levels of control, it appears that it is the control orientation that is more important.