According to the self-determination theory, when people choose to work voluntarily and have intrinsic motivations they show more interest and enthusiasm that manifest in their results toward work and general life. With a sample of 2320 temporary agency workers and through structural equations models, our main goal was to analyze the relationship between motivations for being temporary agency workers and work engagement. Additionally, we studied the extent to which more work engagement is associated with higher levels of context free well-being and we also analyzed the role of engagement as mediator in the relationship between motivations and this context free well-being. Generally, the results support our assumptions and suggest the important role of temporary workers’ motivations in their well-being.
Motivation has been a construct largely studied and valued in the field of Psychology mainly due to its consequences. Motivation leads the individuals to action, giving both direction and intention to this (Ryan & Deci, 2000). According to the self-determination theory (SDT), when individuals are involved in a task or action they can have many different reasons for doing so and these reasons have impact in the quality of their behavior and their mental health (Deci & Ryan, 2000).
Among different types of employment relationships that we can find in contemporary organizations, the present study will focus on temporary agency workers (TAW) that are characterized by having a threefold employment relationship, i.e., they are hired by a temporary work agency – the juro employer – to perform work at a client company – the de facto employer (De Cuyper et al., 2008).
Although permanent employment relationships (i.e. an employment relationship where it is expected that work was done in full-time, would continue in a long-term, and was performed at the employer's place of business under the employer's direction) are the most predominant, the growth in temporary employment is a striking aspect in Western working life (De Cuyper et al., 2008). In 2012, almost 36 million people around the world have been working in agency work (CIETT, 2014). More specifically, in Portugal – the country where the current investigation is performed – the number of workers with temporary employment contract increased 2.00% between 2005 and 2013 (PORDATA, 2014). This temporary employment relationship can be seen as supporting people in important transitions in the labor market. For example, transitions from short term employment to a more permanent position and also transitions from unemployment to work (CIETT, 2014). In Portugal, according CIETT report (2014), the amount of agency workers who remain employed after temporary work is significantly higher (71%) than those who do not have this experience of temporary employment (18%). This suggests that agency work may not only function as a short term solution but also increases the opportunity to overcome the unemployment situation.
The agency work was developed to meet an increased volatility in labor demand and to support organizations in adapting to the impact that each economic cycle has on their employment levels (CIETT, 2014; EUROCIETT, 2011). It is important to note, that in the last years, organizations suffered from a constant pressure due to economic (e.g. increased international competition, market fluctuations) and social changes (e.g. workforce demographic characteristics, technological development). Thus, a more complex environment emerges resulting in structural organizational changes including downsizing and outsourcing (Chen, Tsui, & Zhong, 2008; Hiltrop, 1996; Zhao, Wayne, Glibkowski, & Bravo, 2007).
Considering the specificity of the agency employment it is important to observe how motivations for being TAW are related to their well-being. Along with the previous overview, other studies underline the vulnerability of temporary workers. This point of view comes from the instability and insecurity associated with the work arrangement, deficient benefits and on-the-job-training, lack of promotion prospects and exposure to hazardous work conditions (e.g., Benach, Benavides, Platt, Diez-Roux, & Muntaner, 2000; Kompier, Ybema, Janssen, & Taris, 2009; Sverke, Hellgren, & Naswall, 2002; Underhill & Quinlan, 2011). Instead of this negative view it is also interesting to analyze if the workers that have a positive work experience (i.e. cognitive and affective positive mindset) spillover this experience to non-work domains. As such, in this study we analyze: (1) the relationship between the motivations and the work engagement, (2) the relationship between the work engagement and the context free well-being and (3) the work engagement as a mediator between the motivations and the context free well-being.
This study has contributions to the literature and future investigations. First, although some studies suggest the significant impact of motivations in TAW outcomes, these studies used merely a dichotomous perspective (e.g., Chambel, 2011), or consider only three types of motivations (De Cuyper & De Witte, 2008). The present research, using a measure already validated by Lopes and Chambel (2014), aims to provide a more complex analysis of TAW motivations, conceptualized in SDT (Deci & Ryan, 2000). With the recognition of the differentiated impact that each type of motivation can have on the workers’ results, it is expected to contribute with proposals or actions to be taken by temporary work agencies and their client companies. Secondly, as previously noted, the present research aims to give some clues about the importance of the experiences in the professional context to the experiences outside the work, through the analysis of well-being reported by individuals. By focusing on the work engagement (i.e. a positive dimension of work well-being) we expect help to clarify the repercussions of one domain in other life domains. Third, work engagement has been suggested as a possible mediating variable (e.g. Salanova & Schaufeli, 2008; Yalabik, Popaitoon, Chowne, & Rayton, 2013). However, to our knowledge, no studies have yet been done to examine the role of work engagement as a mediator variable of the relationship between the motivations and the context free well-being of individuals.
2Literature review and hypotheses2.1Motivations and work engagementBased on the SDT, in the present research we considered four types of motivations, attending to the characteristics of temporary work. The types of motivations vary within a continuum of motivation: from motivations with a more intrinsic and autonomous/self-determined nature to motivations with increasingly extrinsic and less autonomous/self-determined nature (Ryan & Deci, 2000). We distinguished: (1) intrinsic motivation that represents individuals that choose this temporary employment arrangement for its inherent satisfaction, which is highly autonomous and does not depend of external contingencies; (2) integrated motivation that represents individuals that accept the current temporary arrangement because it is actually the type of employment that fits better their personal needs or commitments (e.g. school, family) or that better provides the flexibility they need, as well as the freedom to balance it with other personal goals; (3) identified motivation that reflects a conscious valuing of a behavioral goal or regulation, in such a fashion that the action is accepted or owned as personally important. When TAW motivation is regulated through identification, they recognize the value of that job to their personal objectives, since they see it as a mean to develop skills that will be helpful for them or as a mean to gain a permanent employment – i.e., stepping stone motive. Finally, we also considered a less autonomous type of motivation, the (4) external motivation, which encloses individuals that opt for this temporary arrangement in order to obtain a desired consequence (e.g. tangible rewards) and to avoid punishments (e.g. unemployment).
Some theoretical studies suggest a relationship between motivations and work well-being, namely positive relationships between intrinsic and integrated motivation and engagement (e.g. Gagné & Deci, 2005; Stone, Deci, and Ryan, 2009). Work engagement can be defined as a cognitive and affective positive mindset related to work, and characterized by vigor, dedication and absorption (Schaufeli, Salanova, González-Romá, & Bakker, 2002). Vigor is translated in high levels of energy and mental resistance as well as in desire and capacity to invest effort in work. Dedication is a feeling of relevance, enthusiasm, inspiration, pride and challenge directed toward work. Absorption is similar to a state of persistent flow, where the person is totally concentrated, loses the notion of time and goes on involved with the job. Thus, it is expected that more positive levels of work engagement are resulting from more autonomous and voluntary types of motivations.
As such, hypothesis 1 has been established:H1 The motivations for being TAW are related to their work engagement, in a way that TAW that have more autonomous and voluntary types of motivations (i.e. intrinsic, integrated and identified motivations) are those that show higher work engagement, when compared with TAW with more involuntary types of motivations (i.e. external motivations).
There are several studies suggesting that the experiences at work are particularly important for the individual's overall level of well-being and mental health (e.g. Ahola et al., 2005; Hakanen & Schaufeli, 2012; Schaufeli, Bakker, & Van Rhenen, 2009).
What happens at work matters because working individuals spend so much of their time at work, but also because it is salient in the way they think about themselves and the value they attach to their lives (Böckerman, Bryson, & Ilmakunnas, 2012). In that sense, work is considered a fundamental dimension of the individuals’ life and when workers have more positive psychological state about their work (e.g. engagement) they also have more well-being in their general life (Hakanen & Schaufeli, 2012). In fact, work engagement has the potential to promote the quality of life, a state of overall satisfaction derived from the realization of one's potential, that includes satisfaction with life (Diener, Emmons, Larsen, & Griffin, 1985) and positive perception of health (Ware, Davies-Avery, & Donald, 1978). Generally, we expect that work engagement spills over and generalizes to other life domains of the individuals resulting in more positive levels of context free well-being, which encloses more positive levels of satisfaction with life and health perceptions.
As such, hypothesis 2 has been established:H2 Work engagement relates positively with the two dimensions of context free well-being – i.e. satisfaction with life and health perceptions.
As aforementioned, previous studies suggest that individuals’ motivations are related with their engagement toward work, as well, that work engagement may spill over to other life domains resulting in more positive levels of context free well-being, which encloses both positive levels of satisfaction with life and health perceptions. It is also largely supported, by SDT, a close relationship between motivation and context free well-being. For example, Ryan and Deci (2000) referred that “comparisons between people whose motivation is authentic (literally, self-authored or endorsed) and those who are merely externally controlled for an action typically reveal that the former, relative to the latter, have more interest, excitement, and confidence, which in turn is manifest (…) [in their levels of] general well-being” (pp. 69).
Work engagement as a positive, fulfilling, and work-related state of mind may function as a mechanism that allows explain how motivations lead to workers’ results (Beek, Taris, Schaufeli, & Brenninkmeijer, 2014). The SDT (Ryan & Deci, 2000) that conceptualizes the existence of three basic psychological needs – autonomy, competence and relatedness – suggests that individuals feel up motivated and will enhance your well-being depending on the extent to which they experience the satisfaction of basic psychological needs. If environment provides support to psychological needs, we should expect that individuals internalize and integrate better the norms and regulations of the context, presenting motivations to work with a more voluntary and self-determined nature. In face of these motivations with more voluntary and self-determined nature, we may expect that workers present more work engagement, since it connects to energy (Vigor), enthusiasm (Dedication) and involvement (Absorption) and enjoy their current employment arrangement. Moreover, as aforementioned, this positive state of mind at work spills over to other life domains resulting in more positive levels of context free well-being. In summary, we suggest that motivations for being TAW are related to the levels of context free well-being, since that, at first, these motivations contribute to work engagement, which in turn should reflect an improvement in other life domains, resulting in more positive levels of context free well-being.
As such, hypothesis 3 has been established:H3 The relationship between motivations for being TAW and their context free well-being is mediated by work engagement.
The study sample consists of TAW (n=2320) from different temporary work agencies in Portugal that work in manufacturing or service sectors.
The sample has a relatively balanced distribution in terms of gender (54.00% of the participants are women and 46.00% of the participants are men), with the median age of 31 years (SD=8.29) – varying between 18 and 64 years – and a skill level of secondary education – 12th Grade (51.10%) – or higher (44.20%). Only a reduced number of participants are students (12.60%).
Relatively to the situation of employment, mostly the participants considered that the present job is adequate to their skill level (61.40%) and have only a single employment relationship (91.50%) in a full time schedule (78.80%). Prior to being in the present contract, the largest percentage was unemployed (51.60%).
The research procedure involved several steps. A researcher met the CEO of the each respective agency of temporary work to explain the purpose and requirements of the study. The CEO was asked to disseminate the study through the company's intranet with the link that allowed workers to access the questionnaire, which was answered on-line. The participants participated voluntarily and the results were reported directly to them as well as to their respective CEO (survey feedback method). All the respondents completed the survey anonymously and were assured by the researcher that their answers would remain confidential. There was no incentive (cash or otherwise) for participating in this study.
3.2Measures3.2.1MotivationsThe four types of motivations examined in this study – (1) intrinsic motivation (e.g., “Because I like to be temporary”; α=.88), (2) integrated motivation (e.g., “Because it is the job that best fits to my needs (example: studies, family)”; α=.80), (3) identified motivation (e.g., “I chose to be temporary because it will allow me to get a permanent job”; α=. 80) and (4) external motivation (e.g., “Because I need to have a salary to live”; α=.81) – were measured, each, by the participants’ responses to four items scored on a seven-point Likert scale (1=not at all to 7=totally). The set of four items, for each type of motivation, was built by Lopes and Chambel (2014) based on the previous scale – “The Motivation at Work Scale” – developed by Gagné et al. (2008). Highest scores in a particular type of motivation analyzed indicate that workers’ reasons for being TAW are better reflected by this type of motivation.
3.2.2Work engagementWork Engagement, which included the dimension of vigor (three items e.g., “When I wake up in the morning, I feel good about going work”; α=.90), dedication (three items e.g., “My work inspires me”; α=.90) and absorption (three items e.g., “I am immersed in my work”; α=.76), were measured by the participants’ responses to items scored on a seven-point Likert scale (1=never to 7=every day). The set of nine items was built by Schaufeli, Bakker, and Salanova (2006), which had already been used in Portugal (e.g. Lopes & Chambel, 2014; Moura, Orgambídez-Ramos, & Gonçalves, 2014). Higher values indicate that workers have higher levels of work engagement.
3.2.3Context free well-beingWe measured TAW context free well-being with the assessment of Satisfaction with Life and Perceptions of Health. Satisfaction with Life was measured by the participants’ responses to five items (e.g., “If I could live my life again I would hardly change anything at all”; α=.89) scored on a seven-point Likert scale (1=totally disagree to 7=totally agree). The set of five items was built by Diener et al. (1985), which had already been used in Portugal (e.g. Carvalho & Chambel, 2013; Neto, 1992). Health Perceptions were measured by the participants’ responses to four items (e.g., “I am as healthy as others”; α=.90) scored on a five-point Likert scale (1=definitely false to 5=definitely true). The set of four items was built by Ware et al. (1978), which had already been used in Portugal (e.g. Carvalho & Chambel, 2013). Higher values on both dimensions of context free well-being indicate that workers have a favorable evaluation of these variables.
3.2.4Control variablesIn order to eliminate potential alternative explanations for the results, gender (0=Men, 1=Women) and educational level (1=9th Grade, 2=12th Grade, 3=Graduate degree) were used as control variables.
4Results4.1Confirmatory factor analysisSince this study used self-reported measures, we considered the recommendation of Podsakoff, MacKenzie, Lee, and Podsakoff (2003) to test the error of common method variance, performing a Confirmatory Factor Analysis of our Theoretical Model. To confirm the dimensionally, two possible models were examined for comparison purposes.
The models were compared based on chi-square difference tests and on other fit indices: the Standardized Root Mean Square (SRMR), the Bentler Comparative Fit Index (CFI), the Tuckler Lewis Index (TLI), and the Root Mean Square Error of Approximation (RMSEA). Levels of .90 or higher for CFI and TLI and levels of .06 or lower for RMSEA, combined with levels of .08 or lower for SRMR, indicated that the models fitted the data reasonably well (Arbuckle, 2003).
The model with seven latent factors (i.e. intrinsic motivation, integrated motivation, identified motivation, external motivation, work engagement, satisfaction with life, and perceptions of health) that corresponds to our Theoretical Model showed a good fit [χ2 (326)=2371.24, p<.01, SRMR=.05, CFI=.95, TLI=.94, RMSEA=.05]. Comparing with other alternative models tested, we verified a significantly lower fit on the single factor model (Alternative Model I) – where all items loaded on a single latent variable [χ2 (347)=22,971.66, p<.01; SRMR=.15, CFI=.42, TLI=.37, RMSEA=.17] – as well as on the four latent factor model [Alternative Model II – χ2 (341)=8018.25, p<.01, SRMR=.10, CFI=.80, TLI=.78, RMSEA=.10], where the four types of motivations considered were grouped only as one latent variable. Furthermore, the difference between our Theoretical Model and other alternative models was significant [with Alternative Model I – Δχ2 (21)=20,600.42, p<.01; with Alternative Model II – Δχ2 (15)=5647.01, p<.01]. Thus, our theoretical model was the one that represented the best fit (see Table 1).
Results of confirmatory factor analysis (n=2320).
Models | χ2 | Δχ2 | SRMR | CFI | TLI | RMSEA |
---|---|---|---|---|---|---|
Theoretical model | ||||||
Seven latent factors model = Intrinsic Motivation, Integrated Motivation, Identified Motivation, External Motivation, Work Engagement, Satisfaction with Life and Health Perceptions, each one as a latent factor | χ2 (326)=2371.24** | – | .05 | .95 | .94 | .05 |
Alternative model I | ||||||
One latent factor model= with all items, of each variable studied, loading into one latent factor | χ2 (347)=22,971.66** | Compare to Theoretical Model Δχ2 (21)=20,600.42** | .15 | .42 | .37 | .17 |
Alternative model II | ||||||
The four types of motivations grouped only as one another latent factor and Work Engagement, Satisfaction with Life and Health Perceptions, each one as a latent factor | χ2 (341)=8018.25** | Compare to Theoretical Model Δχ2 (15)=5647.01** | .10 | .80 | .78 | .10 |
The means, standard deviations and correlation matrix are presented in Table 2. Looking at the mean rates of the four types of motivation, it is observed that the workers attribute on average a lower score regarding the intrinsic motivations for being in their current employment arrangement (M=1.83, SD=1.14; considering a Likert scale of seven-points) followed by the other types of motivations studied, where the workers rate progressively more highly each type of motivation (integrated motivation – M=2.84, SD=1.45; identified motivation – M=3.49, SD=1.51; external motivation – M=5.79, SD=1.19). The mean values obtained for the three dimensions of work engagement (i.e., vigor – M=5.37, SD=1.44; dedication – M=5.25, SD=1.61; and absorption – M=5.32, SD=1.37; considering a Likert scale of seven-points) suggest also that the workers have positive levels of this cognitive and affective mindset related to work, observing that the mean values obtained are relatively similar across the three dimensions of work engagement analyzed. The mean value registered for the satisfaction with life (M=3.79, SD=1.50; considering a Likert scale of seven-points) indicates a slightly neutral perception of this variable, whereas for the health perceptions it is observed that workers classify these in a more positive way (M=3.68; SD=.86; considering a Likert scale of five-points).
Means, standard deviations and correlations between variables (n=2320).
Mean | SD | r Sample | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
1. | 2. | 3. | 4. | 5. | 6. | 7. | 8. | 9. | .10 | |||
1. Gender | .54a | .50a | ||||||||||
2. Educational Level | 2.40a | .58a | .18** | |||||||||
3. Intrinsic Motivation | 1.83 | 1.14 | −.02 | −.16** | ||||||||
4. Integrated Motivation | 2.84 | 1.45 | −.03 | −.13** | .64** | |||||||
5. Identified Motivation | 3.49 | 1.51 | −.09** | −.17** | .45** | .55** | ||||||
6. External Motivation | 5.79 | 1.19 | .05* | −.10** | −.06** | .03 | .16** | |||||
7. Vigor | 5.37 | 1.44 | −.05* | −.22** | .25** | .27** | .44** | .07** | ||||
8. Dedication | 5.25 | 1.61 | −.04 | −.24** | .26** | .31** | .49** | .10** | .87** | |||
9. Absorption | 5.32 | 1.37 | −.03 | −.17** | .21** | .23** | .40** | .11** | .80** | .80** | ||
10. Satisfaction with Life | 3.79 | 1.50 | .02 | −.12** | .31** | .36** | .38** | −.06* | .40** | .44** | .34** | |
11. Perceptions of Health | 3.68 | .86 | −.14** | −.12** | .17** | .22** | .23** | .04 | .39** | .35** | .34** | .40** |
Analyzing the correlations between the variables studied (see Table 2), we found that all the studied motivations relate significantly to the three dimensions of work engagement, registering a weaker relationship between the external motivation and the three dimensions of work engagement (with vigor – r=.07, p>.01; with dedication – r=.10, p>.01; with absorption – r=.11, p>.01), comparing with the other types of motivations studied (i.e. intrinsic motivation, integrated motivation and identified motivation), which are conceptualized as having a more intrinsic nature. As predicted, vigor, dedication and absorption were all work engagement dimensions significantly related to the two dimensions of context free well-being, namely satisfaction with life and health perceptions – observing that the correlations between the variables vary between .34 and .44 (p>.01). Similarly, we found also significant relationships between motivations and satisfaction with life and health perceptions with the exception of external motivation, which is non-significantly, related to the health perceptions (r=.04, n.s.).
4.3Structural equation modelsTesting mediation analysis through structural equation models (SEM) has some similarities with the Baron and Kenny (1986) approach but provides several advantages (James, Mulaik, & Brett, 2006). As a confirmatory approach, SEM simultaneously tests the relationships among an initial variable, a mediator, and an outcome variable. In addition, SEM analyses derive from nested model comparisons, allowing us to hone in on the specific parameters of interest and to contrast a given pattern of effects against viable alternatives.
To test the mediation relationship, we computed two sets of SEM models (see Table 3). First, we analyzed a mediation model that includes a full mediation of the relationship between the four types of motivations and the two dimensions of context free well-being by work engagement [χ2 (335)=2667.30, p<.01, SRMR=.08, CFI=.94, TLI=.93, RMSEA=.06], which showed an acceptable fit. After, beyond the relationships established at the mediation model above, we draw a partial mediation model, which includes direct relationships between the four types of motivations and the two dimensions of context free well-being. This partial mediation model showed an acceptable fit [χ2 (327)=2509.93, p<.01, SRMR=.06, CFI=.94, TLI=.94, RMSEA=.05], significantly better than the full mediation model [Δχ2 (8)=157.37, p<.01].
Regarding Fig. 1, the results obtained with this final model suggest that, contrary to what was expected, in the particular case of TAW, the more intrinsic nature of motivations does not mean more work engagement, refuting our hypothesis 1. Therefore, we found that integrated motivation, related to the vision of temporary work as allowing more flexibility and balance of work with other responsibilities/needs of the individuals, is negatively and significantly related to work engagement (β=−.14, p<.01). Interestingly, among the four types of motivations analyzed – i.e. intrinsic motivation, integrated motivation, identified motivation and external motivation – only identified motivation was found to have a positive and significant relationship with work engagement (β=.54, p<.01). In that sense, when the motivation for being TAW relates to the fact that the temporary work can be seen as allowing access to a permanent employment or as enabling the development of skills that will be useful in the future, more work engagement these workers present. According to what was expected, we observed positive and significant relationships between work engagement and the two dimensions of context free well-being analyzed: satisfaction with life (β=.38, p<.01) and health perceptions (β=.36, p<.01), which support hypothesis 2. Regarding the role of engagement as a mediator of the relationship between motivations and the two dimensions of context free well-being: engagement, it is a mediator who helps to partially explain the relationship between integrated motivation and the two dimensions of context free well-being – life satisfaction (Z=2.70, p<.01) and health perceptions (Z=2.73, p<.01) – as well as a mediator that seems totally explain the relationship between identified motivation and the two dimensions of context free well-being – satisfaction with life (Z=9.66, p<.01) and health perceptions (Z=10.92, p<.01). Given these results, our hypothesis 3 was partially supported.
5DiscussionIn the current study, we investigated (a) whether the motivations for being TAW – namely, intrinsic motivation, integrated motivation, identified motivation and external motivation – are related to work engagement, (b) the relationship between work engagement and the two dimensions of context free well-being (i.e. satisfaction with life and health perceptions) and (c) the mediate role of work engagement to explain the relationship between motivations and the context free well-being.
Contrary to what was expected, among the four types of motivation analyzed, only the integrated motivation and the identified motivation appear to be related with work engagement, refuting our expectations. While integrated motivation is negative and significantly related to work engagement, identified motivation showed a positive and significant relationship with work engagement. A possible explanation could be due to the extent to which TAW desire a direct contract with the client organization. In fact, when TAW prefer this temporary status and do not desire a direct contract with the client organization this employment is simply an opportunity to have a work and they invest little emotional attachment or commitment to their work (Chambel & Castanheira, 2007). In that sense, the relationship between integrated motivation – that represents the type of extrinsic motivation with a higher degree of voluntarism and probably with lower desire to obtain a direct contract with client organization – and work engagement will be negative, which our results showed. On the other hand, when TAW show a lower degree of voluntarism and probably wanting to acquire a permanent contract in the client company – that represents TAW with identified motivation – these workers may think they are increasing the likelihood of being hired by the client company, as it is their desire, by showing more positive results toward their work (Lopes & Chambel, 2012). This may explain the significant and positive relationship between identified motivation and work engagement observed at our study. This result contrary to our expectations supports Gallagher and Connelly (2008) claim that “researchers should not simply assume that all behavioral theories will apply equally well to nonstandard workers (…)” (pp. 631). Other studies have been shown that TAW desire to obtain a direct contract with client organization is an important variable to explain their employment relationship with this organization (De Cuyper & De Witte, 2008) because they react differently to its practices (Chambel, Sobral, Espada, & Curral, 2013; De Jong & Schalk, 2010; Espada & Chambel, 2013). In this study we considered that probably the TAW desire to obtain a direct contract with the client organization also is an important variable to explain the relationship between their motivation for being temporary and their well-being. In that sense, similar to previous studies, our study seems to confirm the fragility of temporary workers. However, while previous studies demonstrated that TAW present more positive results independent of organizational actions, we observed that this fragility also operates at motivation level. In this way, when TAW recognize the agency work as an important stepping stone to a permanent position – which reflect identified motivations for being TAW – they show high well-being levels. Future studies should examine the extent to which these explanations are valid to justify the results obtained.
According to what was suggested by other studies (Ahola et al., 2005; Hakanen & Schaufeli, 2012; Schaufeli et al., 2009), we observed positive and significant relationships between work engagement and the two dimensions of context free well-being analyzed: satisfaction with life and health perceptions. These results seem to highlight the importance of work domain to the general life of TAW and supported the assumption that the positive feelings derived from well-being at work can spill over to outside-work domains (Édouard & Duhaime, 2013). According conservation of resources theory (COR; Hobfoll, 1998) exist a possibility of gain spirals, i.e., those who possess resources, are likely to gain more resources over time (Hobfoll, 2001). Being engaged at work may further increase personal resources and may thereby also spill-over to context free well-being. Given the spillover effect suggested, further studies are necessary to better analyze this relationship.
Regarding the mediation hypotheses formulated in our study, it is interesting to observe that for TAW with integrated motivations, work engagement is a mechanism that helps to partially explain their levels of context free well-being. For these TAW with integrated motivations, since they recognize the importance of that work to other personal goals beyond work, work engagement appears as a mechanism that contributes to explain the relationship between motivations and context free well-being, but is not the only variable that explains these relationships. However for TAW with identified motivations, work engagement seems to explain in full the levels of context free well-being observed. These results seem to highlight the importance of work engagement for TAW with identified motives, for which a work that meets their professional objectives will strongly contribute to their work engagement that, in turn, will contribute to their context free well-being. This result reinforces the important role of TAW desire to obtain a direct contract with the client organization. Probably, among TAW hoping that this temporary job might serve as a vehicle to achieve this direct contract, their well-being at work is a vehicle to explain their context free well-being. Showing this positive psychological state at work TAW may consider they enhance their desire fulfillment, namely a direct contract by client organization. Future studies should replicate these analyses to find if this pattern of results would be similar to the obtained with our research.
6LimitationsAlthough this research has important strengths it is not without limitations. Primarily, this study's cross-sectional design does not allow us to establish any causal relationships between variables, giving no indication of the sequence of events. As such, it is necessary that longitudinal studies are performed in order to overcome this constraint. Second, the exclusive use of self-report measures could lead to problems such as common method effects. However, self-reported data seemed the most appropriate avenue to capture worker's perceptions and appraisals of the variables analyzed (Fox & Spector, 1999) and according to Spector (2006) concerns associated with relying heavily on self-reported data measurements may be overstated. Furthermore, we noted some of the methodological recommendations by Podsakoff et al. (2003): we collect the data from different temporary work agencies; we demonstrated the best fit of our model compared with a single factor model by conducting confirmatory factor analysis; we guaranteed anonymity and confidentiality of responses; we used questions with different response-scale amplitudes; we indicated in every questionnaire that there were no right or wrong answers; we used standardized measures that had already been used in previous studies; and we counterbalance the order of the measurement of the variables conceptualized as independent, mediator and dependent variables. Thirdly, there are limitations due to the nature of the sample, which is composed mainly of people with high skills (51.10% of the sample have a skill level of secondary education and 44.20% of the sample have a graduate degree) and relatively young (most the sample – 54.90% – are between 21 and 31 years old) working only through a temporary work agency. At the level of temporary work, studies are needed that consider other types of alternative work arrangements (e.g. on-call work, outsourcing, independent contractors), beyond the temporary agency work arrangement, since as previous studies suggest (e.g. Cohany, 1996; DiNatale, 2001) there is a wide diversity in terms of demographics, salary and benefits, as well as differences in the reasons that lead the individuals to have a temporary work, according the type of employment arrangement. Finally, this research was conducted in the Portuguese context, with a Western culture, where many temporary agency workers want to be hired by the client company (Chambel & Alcover, 2011). Thus, the results should not be extrapolated to other cultures.
6.1Implications for human resource practicesBased on our results, it is possible to conclude that TAW may experience a positive employment condition that entails their positive well-being. Through actions that signal to TAW that this employment experience is contributing to develop their skills, to increase their employability and to increase the probability to achieve a permanent position they display a positive psychological state at work, e.g. work engagement. Establishing and maintaining this positive psychological state is vital to organizational effectiveness, namely its performance (Halbesleben, 2010; Harter, Schmidt, & Hayes, 2002). Furthermore, this study's results have shown that work engagement is also associated with TAW satisfaction with life and perceptions of health. Thus, the organization also contributes to the fulfillment of its ethical obligation to construct a healthy employment context for its workers. To manage this situation, when temporary work agencies operate in highly competitive markets, in which the hypotheses of relocation of TAW, with fundamental skills and expertise to a number of client companies, are predictably high, the focus should be on establishing an ongoing working relationship with these TAW.
However, the results show that individuals who have voluntary reasons for being temporary workers, who consider this status as the type of employment that fits better with their personal needs or commitments (e.g. school, family), feel poor work engagement. This means that if organizations intend to get well-being of workers through their motivations for being TAW, they must try to understand the reasons to which temporary workers choose and maintain this employment status. This does not mean that organizations should only recruit or reassign TAW with high identified motivation and low integrated motivation. Moreover, both legislation and ethical codes prohibit discriminative selection procedures. We support the use of actions and practices and suggest that HR practitioners pay special attention to those TAW whose well-being cannot be obtained through their motivation, and use other means to obtain it.
In summary, among the four types of motivations (i.e. intrinsic, integrated, identified and external motivation) it was observed that integrated and identified motivations are the ones that contributed to explain workers well-being. In line with this, HR of temporary agencies should develop a close relationship with the client organizations to contract TAW whenever they need to increase their permanent workforce.
This study is part of a larger research project “Temporary agency workers’ transitions: motives, experiences and outcomes – PTDC/MHC-PSO/4399/2012” founded by the Foundation of Science and Technology (FCT) from Portuguese Government. This research was also supported by a doctoral grant from Portuguese Foundation of Science and Technology (FCT) – SFRH/BD/94021/2013. These grants are gratefully acknowledged.