Contextual and Individual Determinants of Sustainable Careers: A Serial Indirect Effect Model through Career Crafting and Person-Career Fit

By inergency On Mar 29, 2024

[ad_1]

4.1. Construct Validity and Reliability

We evaluated the construct validity by using confirmatory factor analysis (CFA) via IBM SPSS Amos 23.0. The CFA results revealed that the hypothesized second-order 10-factor measurement model fits adequately (χ² = 2076.43, df = 1324, χ²/df = 1.57, CFI = 0.94, NNFI = 0.93, SRMR = 0.047, RMSEA = 0.037). The hypothesized model included a four-dimensional career crafting construct, a two-dimensional person-career fit construct, a two-dimensional perceived employability construct, and seven unidimensional constructs. Additionally, we compared the hypothesized model with alternative models. The results indicated that the first-order 15-factor measurement model demonstrated a good fit (χ² = 1965.39, df = 1272, χ²/df = 1.54, CFI = 0.95, NNFI = 0.94, SRMR = 0.051, RMSEA = 0.062), while one-factor model indicated an overall poor fit (χ² = 8754.62, df = 1377, χ²/df = 6.36, CFI = 0.40, NNFI = 0.38, SRMR = 0.120, RMSEA = 0.114) (see Table 1). Considering that both the hypothesized second-order model and the first-order model fit the data well and the statistical difference between these models is rather minor, we continued the analyses with the initial hypothesized second-order model.

To ensure convergent validity, we examined the factor loadings, AVE (average variance extracted) scores, and construct reliability scores [56]. Standardized factor loadings of items were generally above 0.70 (see Appendix A for items and factor loadings). Six items had factor loadings lower than 0.70, but they were significant and loaded higher than 0.50, and thus, were retained. The AVE scores of all variables were above the cutoff value of 0.50. Construct reliability scores for each variable were well beyond 0.70. The results indicate that the constructs in the model have convergent validity [56]. Furthermore, the results revealed that there are moderate correlations between the constructs, and the squared root of AVE scores for each construct was greater than the correlations with other constructs, which indicates that the constructs in the model have discriminant validity [57]. Table 2 demonstrates the descriptive statistics, CR and AVE scores, and correlations.

4.2. Hypothesis Tests

To test the predicted hypotheses, we performed hierarchical regression analyses via IBM SPSS. In addition, to test the indirect effects, we carried out analyses through SPSS Process Macro, which produces more robust and reliable results through a bootstrapping method [58]. We performed Process Macro’s measurement model-4 and model-6 to test the simple and serial indirect effect paths. The estimates were obtained by running 5000 bootstraps and resampling with a 95% CI. We first tested the relationships between contextual demands and resources, career crafting, and person-career fit. Results revealed that perceived work demands (β = 0.163, p p p p p p p Table 3).

Moreover, we tested whether an indirect relationship existed between contextual demands/resources and person-career fit through career crafting. When career crafting was included in the regression model, it was found to be positively related to person-career fit (β = 0.633, p < 0.01), and the effect sizes of contextual demands and resources decreased substantially. Results also showed that perceived work demands (b = 0.0852, CI: [0.0312, 0.1446]) and family demands (b = 0.0821, CI: [0.0370, 0.1249]) were indirectly related to person-career fit via career crafting. Therefore, H1a and H1b were supported. Furthermore, family-supportive supervision (b = 0.1530, CI: [0.1067, 0.2025]) and work-supportive family (b = 0.0991, CI: [0.0544, 0.1464]) had significant indirect effects on person-career fit via career crafting. Thus, H2a and H2b were supported.

Furthermore, we tested the relationship between career crafting and career sustainability indicators. Findings from hierarchical regression analyses revealed significant relationships between career crafting, person-career fit, and career sustainability indicators (see Table 4). Initially, we found that career crafting was positively related to person-career fit (β = 0.688, p p p p p

Additionally, by performing three-step hierarchical regression analyses, we tested the relationship between contextual demand/resources and career sustainability indicators (i.e., work-family balance, work engagement, employability, and work strain). We also included career crafting and person-career fit sequentially in each step, thereby examining any potential change in the model’s explained variance. First, hierarchical regression analysis results regarding the work-family balance model demonstrated that the contextual resources of family-supportive supervision (β = 0.332, p p p p ² = 0.129). In the third step of the analysis, the person-career fit was included in the model. Results did not yield a significant relationship between work-family balance and person-career fit (see Table 5).

Second, findings from the hierarchical regression analysis pertaining to the work engagement model indicated that the contextual resources of family-supportive supervision (β = 0.362, p p p p ² = 0.242). In the third step of the analysis, the person-career fit was involved in the model. Although person-career fit was found to be positively related to work engagement (β = 0.142, p ² = 0.010) (see Table 6).

Third, hierarchical regression analysis results regarding the employability model displayed that the contextual resources of family-supportive supervision (β = 0.300, p p p ² = 0.120). When person-career fit was included in the model in the third step of the analysis, it was found to be positively related to employability (β = 0.247, p ² = 0.031) (see Table 7).

Fourth, findings from the hierarchical regression analysis pertaining to the work strain model demonstrated that the contextual resources of family-supportive supervision (β = −0.241, p p p p ² = 0.038). In the third step of the analysis, the person-career fit was included in the model. Results indicated that person-career fit was negatively related to work strain (β = −0.120, p ² = 0.007) (see Table 8).

Moreover, we performed Process Macro analyses to test the serial indirect effect model between contextual demands and career sustainability indicators through career crafting and person-career fit. Results revealed that work/family demands and work-family balance were not serially linked through career crafting and person-career fit. Hence, H4a was not supported. Rather, we found that both work demands (b = 0.0705, CI: [0.0254, 0.1185]) and family demands (b = 0.0679, CI: [0.0313, 0.1069]) were indirectly related to work-family balance via career crafting. Findings also showed that work/family demands and work engagement were serially linked through career crafting and person-career fit. We found that work demands were both directly (β = 0.111, p < 0.05) and indirectly (b = 0.0131, CI: [0.0025, 0.0278]) related to work engagement. Although family demand was not directly related to work engagement, it was indirectly related to work engagement (b = 0.0126, CI: [0.0029, 0.0254]). Hence, H4b was supported. We also found that work/family demands and employability were serially linked via career crafting and person-career fit. Although work and family demands were not directly related to employability, both work demands (b = 0.0222, CI: [0.0064, 0.0442]) and family demands (b = 0.0214, CI: [0.0079, 0.0375]) were indirectly related to employability. Therefore, H4c was supported. In addition, although the results demonstrated that work/family demands had marginally significant indirect effects on work strain, the confidence intervals were too close to the acceptable threshold, and the t-values did not reach a p < 0.05 significance level. Therefore, H4d was not supported. However, results revealed significant indirect effects of work demands (b = −0.0467, CI: [−0.0910, −0.0138]) and family demands (b = −0.0450, CI: [−0.0823, −0.0169]) on work strain through career crafting.

Furthermore, we tested the serial indirect effect model between contextual resources and career sustainability indicators through career crafting and person-career fit. Findings displayed that although the contextual resources of family-supportive supervision (β = 0.332, p p p p p p p Table 9 shows the results regarding the indirect effect paths.

[ad_2]