جستجوی کار و تئوری رفتار برنامه ریزی شده: تفاوت گروه اقلیت اکثریت در هلند

The labor market in many Western countries increasingly diversifies. However, little is known about job search behavior of “non-traditional” applicants such as ethnic minorities. This study investigated minority–majority group differences in the predictors of job search behavior, using the theory of planned behavior (Ajzen, 1985). Data were collected in a two-wave longitudinal design among 697 temporary employees in The Netherlands. Results showed that the ethnic minorities' perceptions of social pressure predicted intentions to search for a (new) job more strongly than their personal attitudes did. The opposite was found in the native-Dutch group. Self-efficacy did not contribute to the prediction of job search intention. Job search behavior related significantly to job search outcomes, such as job attainment.

The process of pursuing (new) employment, or job search behavior, is an important aspect of people's work lives. That is, job search behavior determines the opportunity set of potential jobs from which job seekers may choose (Barber, Daly, Giannantonio, & Phillips, 1994), and influences outcomes such as employment status and employment quality (Schwab, Rynes, & Aldag, 1987). Job search behavior can be defined as “the specific behaviors through which effort and time are expended to acquire information about labor market alternatives” (Bretz, Boudreau, & Judge, 1994, p. 278). It includes activities such as preparing a resume, reading personnel advertisements, making inquiries to prospective employers, and going to job interviews. Previous research has investigated the antecedents of job search behavior and employment outcomes among both individuals entering the workforce, and unemployed and employed individuals. In a recent meta-analysis, Kanfer, Wanberg, and Kantrowitz (2001) quantitatively reviewed this literature. Kanfer et al. (2001) concluded that the vast majority of studies investigated job search behavior among job losers and college graduates, whereas only a few studies reported data from employed samples. Moreover, hardly any of the studies focused on job search behavior of “non-traditional” applicants such as ethnic minorities. In the present time of culturally diversifying workforces (Chemers, Oskamp, & Costanzo, 1995; Triandis, Kurowski, & Gelfand, 1994), this is a serious omission in the literature. In the current longitudinal study we, therefore, investigated and compared job search behavior and its predictors among ethnic minorities and the majority group in The Netherlands. We used a sample of individuals who worked or recently had worked for a temporary employment agency. Job search behavior is especially salient in this group because their employment position is relatively unstable and uncertain compared to the position of permanent employees. In addition, the workforce increasingly consists of temporary workers (e.g., Berchem, 2002; Feldman, Doerpinghaus, & Turnley, 1994; Pot, Koene, & Paauwe, 2001), whilst little research has been conducted in this group (cf. Ellingson, Gruys, & Sackett, 1998). Also, the proportion of ethnic minorities is substantially larger among temporary workers than in the total workforce (Van der Ende, Donker van Heel, Koene, & Nauta, 2002). We investigated the predictors of job search behavior among ethnic minorities and the Dutch majority using the theory of planned behavior (TPB; Ajzen, 1985 and Ajzen, 1991). Although two previous studies already used the TPB to predict job search behavior (Caska, 1998; Van Ryn & Vinokur, 1992), the current study extends the existing literature in three ways. First, previous research examined the applicability of the TPB in US samples of unemployed individuals (Van Ryn & Vinokur, 1992) and graduating students (Caska, 1998). The current study extended the generalizability of the TPB by using the theory to predict job search behavior in a non-US sample of temporary employees. Second, the current study used a longitudinal design and assessed both the predictors and the outcomes of job search behavior. Third, although some studies investigated job search behavior of ethnic minorities (Green, Tigges, & Diaz, 1999; Nesdale & Pinter, 2000), no research specifically examined and directly compared the predictors of job search behavior between ethnic minorities and the majority group. Before discussing the hypothesized cultural differences, we first present the general research model used in this study. 1.1. Research model The theory of planned behavior, an extension of the theory of reasoned action (TRA; Fishbein & Ajzen, 1975), has proved its value in the prediction of a whole range of behaviors, as is demonstrated in several reviews (e.g., Ajzen, 1991; Armitage & Conner, 2001; Sutton, 1998). Applied to job search behavior, the TPB states that the immediate antecedent of job search behavior is the intention to look for a job. Job search intention in turn, is predicted by the extent to which a person has a positive or negative evaluation of job search behavior (i.e., job search attitude), the perception of social pressure to look for a (new) job (i.e., subjective norm), and people's confidence in their ability to perform various job search activities (i.e., perceived behavioral control; Ajzen, 1991). According to the TPB, job search intention completely mediates the effects of job search attitude and subjective norm on job search behavior. Thus, there is no direct link between attitude and subjective norm on the one hand, and behavior on the other. Perceived behavior control, however, is supposed to influence behavior both directly and indirectly through intention. That is, people will be more likely to perform their intended job search activities, such as writing an application letter, if they feel confident in their ability to write a proper application letter (cf. Ajzen, 1991). Previous research has found support for the use of the TPB to predict job search behavior ( Caska, 1998; Van Ryn & Vinokur, 1992). Therefore, we expect the following: Hypothesis 1. (a) Job search attitude, (b) subjective norm, and (c) perceived behavioral control positively predict job search intention. Hypothesis 2. (a) Job search intention and (b) perceived behavioral control positively predict job search behavior. Hypothesis 3. Job search intention: (a) completely mediates the relation of job search attitude and subjective norm with job search behavior, and (b) partially mediates the relation of perceived behavioral control with job search behavior. The most obvious purpose and consequence of job search behavior is successful attainment of (new) employment. Schwab et al. (1987) noted that success in generating job alternatives is a function of the intensity of the individual's job search behavior. Indeed, research has indicated that individuals who spend more time on job seeking are more likely to find a (new) job than others (Kanfer et al., 2001). Based upon this research we expected that: Hypothesis 4. Job search behavior relates positively to job attainment. In addition to finding employment the quality of the obtained employment is an important employment outcome (Schwab et al., 1987; Wanberg, Kanfer, & Banas, 2000). Job search behavior is thought to relate positively to employment quality, because a more intense job search is likely to result in more job opportunities, allowing the job seeker to choose the best alternative. In the current study, we assessed employment quality with two measures: job satisfaction and agreement between the obtained and wanted job. We expected that: Hypothesis 5. Among individuals who obtained (new) employment, job search behavior relates positively to: (a) job satisfaction in the new job and (b) the agreement between the obtained and wanted job. It should be noted that job attainment and employment quality do not just depend on job search behavior. Other variables such as the labor market demand (Wanberg, Hough, & Song, 2002), interviewing skills (Caldwell & Burger, 1998; Maurer, Solamon, Andrews, & Troxtel, 2001), and discrimination in personnel selection (Evers & Van der Flier, 1998; Stewart & Perlow, 2001) influence job attainment. Because our focus was on the predictors and outcomes of job search behavior, we did not include these variables in the current study. Regarding the potential effects of these variables on job search behavior, it should be mentioned that the TPB is a complete theory of behavior, in that other variables are thought to influence behavior only indirectly ( Conner & Armitage, 1998; Fishbein, 1980). Specifically, external variables such as demographics, personality traits, perceived labor market demand, and culture will affect the attitudinal and normative considerations, and those considerations will ultimately predict intentions and behavior. In addition, external variables may have an impact on the relative importance of attitudes and subjective norms ( Fishbein, 1980). In the next section, we will discuss the anticipated effects of culture on the relative importance of job search attitudes and subjective norms in the prediction of job search intentions. 1.2. Minority–majority group differences Populations and workforces in many Western countries increasingly diversify (Alders, 2001; Chemers et al., 1995; Hall, 1997). In the United States, black, Hispanic, and Asian Americans constitute about a quarter of the population, and this proportion is projected to rise substantially (Triandis et al., 1994). In The Netherlands, about 15–20% of the population has a cultural background other than Dutch (Statistics Netherlands, 2001). The major ethnic minority groups in The Netherlands are from Indonesian, Surinamese, Antillean, Turkish, and Moroccan descent. These minority groups have different statuses (Pettigrew, 1998). That is, whereas the Indonesian and Surinamese/Antillean immigrants are from (former) Dutch colonies, the Turkish and Moroccan immigrants originally arrived in The Netherlands as “guest workers” in the 1960s and 1970s. Many of them, however, did not return to their countries of origin, but rather had their families come over to The Netherlands (Van Oudenhoven, Prins, & Buunk, 1998). In general, ethnic minorities have lower levels of education as compared to the native-Dutch (Statistics Netherlands, 2002). Furthermore, their position at the labor market is relatively weak, as is indicated by high unemployment rates and overrepresentation in lower skilled jobs (Dagevos, 2001; Statistics Netherlands, 2002). There is some evidence, however, that these differences are diminishing over time (Te Nijenhuis, De Jong, Evers, & Van der Flier, 2003). Discrimination based on race, ethnicity or nationality is forbidden by law in The Netherlands, and employers are required to make an effort to achieve a proportional representation of ethnic minorities within their organization. The Netherlands have adopted the ideal of multiculturalism, meaning that respect for cultural differences and egalitarian goals are promoted (Arends-Toth & Van de Vijver, 2003). There exists a stern norm against blatant prejudice, illustrated by Pettigrew's (1998) findings that blatant prejudice is relatively low in The Netherlands. Nevertheless, research has shown some evidence for subtle prejudice (Pettigrew, 1998) and indirect discrimination (De Vries & Pettigrew, 1998; Van der Werf, 1992) in The Netherlands. Individuals with different ethnic backgrounds are likely to differ in their attitudes, values, and norms, because of their different cultural roots. Hofstede, 1980 and Hofstede, 1991 extensively investigated value differences between 53 countries and regions. Individualism versus collectivism appeared to be an important dimension on which country cultures differ. Hofstede (1991) described individualism as pertaining to cultures in which the ties between individuals are loose, and people are expected to look after themselves. In these cultures people tend to perceive themselves as autonomous individuals who are independent of the group, and they tend to give priority to personal goals over collective goals (Markus & Kitayama, 1991). Collectivism has been described as pertaining to cultures in which people are integrated in cohesive ingroups that protect them in exchange for unquestioning loyalty (Hofstede, 1991). In these cultures people tend to perceive themselves as interdependent with their group, and they tend to give priority to the goals of the group over their personal goals (Markus & Kitayama, 1991). Previous research has characterized the Dutch national culture as highly individualistic. Hofstede, 1980 and Hofstede, 1991 classified The Netherlands among the five most individualistic countries in his study. In the GLOBE project, The Netherlands belonged to the five countries ranked lowest on group and family collectivism (Javidan & House, 2001). In contrast to the highly individualistic Dutch culture, the cultures of the major ethnic minority groups in The Netherlands have been described as more collectivistic (GLOBE-study, 2001; Hofstede, 1991; Mesquita, 2001). Based on the cultural differences in individualism and collectivism, we formulated the following hypotheses: Hypothesis 6. Whereas (a) subjective norm is a stronger predictor of job search intention than job search attitude in the ethnic minority group, (b) job search attitude is a stronger predictor of job search intention than subjective norm in the Dutch group. Hypothesis 7. Whereas (a) subjective norm is a stronger predictor of job search intention in the ethnic minority group than in the Dutch majority group, (b) job search attitude is a stronger predictor of job search intention in the Dutch majority group than in the ethnic minority group. To summarize, we expect that the TPB accurately predicts job search behavior of temporary employees in The Netherlands. Moreover, we expect job search behavior to be a significant predictor of job search outcomes, such as successful attainment of (new) employment, job satisfaction in the new job, and agreement of the obtained job with the type of job wanted. Furthermore, we expect several differences in the importance of the various predictors of job search behavior between ethnic minorities and the majority group. Specifically, we propose that personal attitudes regarding job seeking are more important predictors of job search intention in the majority group, whereas perceptions of social pressure to seek (new) employment are more important predictors of job search intention in the ethnic minority group.

The results provided partial support for the theory of planned behavior in predicting job search behavior in a sample of temporary employees in The Netherlands. Job search attitude and subjective norm were significant predictors of job search intention, and job search intention significantly predicted job search behavior. Intention fully mediated the effects of attitude and subjective norm on behavior. Self-efficacy, however, did not add to the prediction of intention and behavior. Also the zero-order correlations of self-efficacy with job search intention and behavior were not significant. A possible explanation for these null findings relates to the measure used to assess self-efficacy. The fact that the self-efficacy items did not exactly match the items used to measure job search intention and job search behavior might have deflated the relations of self-efficacy with intention and behavior. The job search attitude and subjective norm measures, however, did not match the job search intention and behavior items either, and these variables did show significant relations with intention and behavior. Although ideally the measures for attitude, subjective norm, and self-efficacy should correspond exactly with the measures for intention and behavior, we chose to use the more global measures to reduce the length of the questionnaire, and to avoid asking too many seemingly repetitious questions. Another possible explanation for the lack of significant findings with respect to self-efficacy relates to the composition of our sample. That is, some individuals in our sample may not (intend to) engage in job search behavior, although they do have high levels of job search self-efficacy. Employed respondents who do not intend to leave their current jobs might be an example of such respondents. The presence of such respondents might have deflated the relation of self-efficacy with job search intention and behavior. This idea, however, was not supported by our data. We repeated our analyses using a subsample of respondents who intended to spend time on at least one job search activity (n=615). The zero-order correlations of self-efficacy with intention and behavior were still non-significant. Also the regression results were highly similar to the results presented before. In the regression analysis of intention, the effect of self-efficacy on job search intention remained non-significant. Previous research studying the relation of self-efficacy with job search intention and behavior found mixed results. While some studies reported moderate or even strong relations (e.g., Blau, 1994; Caska, 1998; Saks & Ashforth, 1999), others found weak or no relations (Van Ryn & Vinokur, 1992; Wanberg et al., 1996). A moderating effect of sample type seems plausible (i.e., self-efficacy is a more important predictor in samples of inexperienced job seekers). Closer examination of our data revealed some support for this notion. For example, among job seeking students (n=268), self-efficacy correlated marginally significant with job search intention, r=.10, p<.10. In contrast, among job seeking non-students (n=346), self-efficacy was not related to job search intention, r=−.02, p=.71. In their meta-analysis, Kanfer et al. (2001) reported a mean corrected sample-weighted correlation of .27 between self-efficacy and job search behavior, which was not found to be moderated by sample type, however. But due to the limited number of studies comprising employed job seekers, this moderator analysis concerned job losers versus new entrants only. Future research should investigate the relation of self-efficacy with job search in different sample types more closely. Following Ajzen and Driver (1992), we distinguished between an instrumental and an affective component of attitude. This theoretical distinction was clearly supported by the data. Both components were only weakly correlated, and showed substantially different correlation patterns with the other study variables. Instrumental and affective job search attitude both added to prediction of job search intention. With this, the current study extends previous research examining job search behavior in the context of the TPB (Caska, 1998; Van Ryn & Vinokur, 1992) or its predecessor, the TRA (Vinokur & Caplan, 1987), suggesting that an individual's job search behavior is stimulated not only by positive instrumental attitudes such as regarding job seeking as wise and useful, but also by affective attitudes such as regarding job seeking as interesting and enjoyable. 4.1. Job search outcomes Consistent with previous research (Griffeth, Hom, & Gaertner, 2000; Kanfer et al., 2001), we found a significant relation between job search behavior and job attainment. However, as noted by Wanberg et al. (1999), in times of a healthy economy most people who are looking for a (new) job, will be able to find it (see also Taris, Heesink, & Feij, 1995). Therefore, it is important to see whether satisfactory employment is found. Our results did not support the expected positive relation between job search behavior and satisfaction with the new job. Previous research reported mixed results concerning this relation. Some studies found a positive association between job search behavior and job satisfaction (e.g., Leana & Feldman, 1995; Steffy, Shaw, & Noe, 1989), whereas others found little or no support (e.g., Saks & Ashforth, 2002; Wanberg et al., 2000; Wanberg et al., 1999; Werbel, 2000). A possible explanation for this null finding might be that those low in job satisfaction may have already started a new job search, or in fact may never have stopped their job search. Indeed, other research has identified job dissatisfaction as an important antecedent of job search behavior among employed individuals (e.g., Blau, 1994; Boudreau, Boswell, Judge, & Bretz, 2001; Bretz et al., 1994; Hom, Caranikas-Walker, Prussia, & Griffeth, 1992). This negative relation between job satisfaction and subsequent job search behavior might have overshadowed the positive relation between job search behavior and subsequent job satisfaction among individuals who found a (new) job. Besides job satisfaction, we used two additional measures related to employment quality, that is, the agreement between the job obtained and the job sought with regard to the amount of hours and the type of contract. Our results indicated that individuals who invested more time in their job search found a job that matched their desires with regard to the type of contract. This is an important outcome in that it offers some support for the contention that investing time in job seeking does not only pay off in a higher probability to obtain a job, but also in higher levels of agreement between the type of employment job seekers were looking for and the type of employment they obtained. Future research should further investigate the effects of job search behavior on different aspects of the type of employment found, such as agreement with respect to type of business, level of the job, and job characteristics (e.g., level of autonomy, responsibility, or skill variety). 4.2. Cultural differences As indicated by Ajzen (1991) the relative importance of attitudes and subjective norms in the prediction of intentions can vary across situations. More specifically, Fishbein (1980) notes that the relative importance of these variables may be influenced by external variables such as demographics or personality. In the current study, we investigated whether cultural differences could account for differences in the relative importance of job search attitudes and subjective norm in the prediction of job search intentions. When testing the TPB for ethnic minorities and the Dutch majority separately, some differences were found between both groups. In the ethnic minority group we found that subjective norms were stronger related to behavioral intentions than were job search attitudes. In the Dutch majority the opposite was found. Our findings confirm previous research by Abrams, Ando, and Hinkle (1998), in which subjective norms were found to relate stronger to turnover intentions in a Japanese sample than in a British sample. Similar to the differences found in Abrams et al.'s (1998) study, our findings related to minority–majority group differences in the strength of the relation of subjective norms and personal attitudes with intentions can be explained by cultural differences in individualism versus collectivism. That is, in collectivistic cultures behavior is guided more by social norms than by personal attitudes, whereas the opposite is true in individualistic cultures (Markus & Kitayama, 1991). Our findings offer some support for the generalizability of the TPB to other, non-Western cultures. That is, in the ethnic minority group as well as in the Dutch group most relations are consistent with the relations as predicted by the TPB. As discussed above, however, the relative importance of the TPB-variables was different in both groups, that is, the strength of several relations was moderated by cultural background. This is an important finding, because many psychological theories have been developed by European Americans (Hall, 1997), and therefore the relevance of these theories in other cultural contexts has been questioned (Nagayama Hall & Maramba, 2001). In times of increasingly diversifying labor markets in many Western countries, it is important to study cultural differences in vocational behavior. 4.3. Limitations In the current study, we looked at cultural differences in the predictors of job search. A limitation pertains to the composition of the ethnic minority group. This group was not very large and ethnically fairly heterogeneous. Due to the small size of the subsample of ethnic minorities, we were not able to investigate the relation between job search behavior and employment quality for ethnic minorities and the Dutch group separately. Moreover, we were not able to further distinguish between the various ethnic groups within this subsample. Future research must confirm the cultural differences reported in this study for specific ethnic groups. Also, future research should include direct measures of collectivism–individualism. Because the actual collectivistic and individualistic values held by the respondents were not assessed directly in the current study, we cannot rule out the possibility that the differences we found between the two groups were caused by other group characteristics than the assumed differences in collectivistic versus individualistic values. Furthermore, the rather low response rate might have influenced the findings, and therefore may limit the generalizability of the study. We were, however, able to compare the respondents with the overall random sample on gender, age, and level of education. Because females and higher educated individuals were slightly overrepresented among the respondents as compared to the overall sample, gender and level of education (along with age and employment position) were used as control variables. The effects of these variables were mostly small and non-significant. However, because respondents might have differed from the non-respondents on other unmeasured variables, such as language skills in Dutch, some caution is needed with regard to the generalizability of the study findings. In summary, the current study investigated cultural differences in the predictors of job search in a sample of temporary employees in The Netherlands. The results showed a stronger influence of the social environment on job search in the ethnic minority group compared to the Dutch group. This pattern of findings is consistent with the view that people in collectivistic cultures attach more importance to harmonious relationships with others (Markus & Kitayama, 1991). Employment counselors could take these differences into account when assisting people in their job search. Offering social support and exerting social pressure are important means of stimulating job search behavior (Caplan, Vinokur, Price, & Van Ryn, 1989). Our findings suggest this might be of even greater importance for job seekers with a collectivistic cultural background.