test Analysis

To compare the differences between participants receiving RJP video and TJP video and test hypotheses 1a: RJPs have stronger effect on increasing individuals’ perception of unmet expectation than TJPs do and 1b: RJPs have stronger effect on increasing individuals’

perception of role clarity than TJPs do, independent-samples t-test was conducted. As the result presented in Table 4.2, individuals receiving RJP video (M= 5.02, SD= 1.14) reported higher score on role clarity than TJP receivers (M= 4.78, SD= 1.13), and the result showed there was no significant difference between two groups t(296)= 1.80, n.s, which rejected hypothesis 1b.

As for job attraction and the discrepancy of job attraction -- unmet expectation, even though participants reviewing TJP video (M= 5.19, SD= 1.04) did have higher score on job attraction than RJP groups (M= 4.86, SD= 1.15) with significant differences, t(296)= -2.64, p<.01, the result of unmet expectation showed that RJP group slightly scored higher than TJP group but the difference between two groups was not significant t(296)= 1.43, n.s.; as a result, hypothesis 1a was not supported.

Table 4.2.

Results of Independent-Samples T-test for Role Clarity, Job Attraction, and Unmet Expectation in RJP and TJP (n=298)

37

Paired-Samples T-test

In order to compare the differences of individuals’ role clarity and job attraction between pre-test and post-test, paired-samples t-test was conducted. RJP group and TJP group both reported higher score on role clarity in post-test, which indicated that either RJP video t(147)=

-10.93, p<.01, or TJP video t(149)= -10.47, p<.05, helped individuals understand the role of marketing specialist in the scenario. However, the result showed that both RJP and TJP groups’

scores on job attraction decreased in posttest, which indicated that both RJP video t(147)= -4.32, p<.01, and TJP video t(149)= -2.91, p<.01, had negative effect on individuals’ perception of job attraction, and RJP video (d= -.37) had stronger negative effect on job attraction than TJP (d= -.26) did.

Table 4.3.

Results of Paired-Samples T-test for Role Clarity and Attraction in Pre-test and Post-test in RJP Group (n=148)

38

Table 4.4.

Results of Paired-Samples T-test for Role Clarity and Attraction in Pre-test and Post-test in TJP Group (n=150)

In order to examine the causal relationship among variables in the current research, path model was constructed and analyzed using AMOS Structural Equation Modeling technique. At first, model fit was checked in order to ensure the goodness of the model and the results are as following: (1) ²/df was 2.763, which fits the criterion (< 5.0), (2) GFI was .848, which was acceptable (>.80), (3) AGFI was .814, which was acceptable (>.80), (4) RMSEA was .077, which was a reasonable fit, and (5) SRMR was .0787, which was an acceptable fit (>.05 but

<.08). The results are shown in Table 4.5. After ensuring the goodness of model fit, regression weight between variables was checked. The results of the standardized estimates show that the independent variable, job preview, significantly affects the dependent variable, individuals’

perception of unmet expectation (β= .13, p< .05); secondly, job preview has a slightly significant effect on the mediator, role clarity (β= .069, .05< p <.1); at last, the result shows the mediator, role clarity, has a strong negative effect on the dependent variable, unmet expectation

39

(β= -.33, p< .01). From the results of path model analysis in SEM, hypotheses 2, 3, and 4 are supported.

Table 4.5.

Model Fit Summary

2 df P 2/df GFI AGFI RMSEA SRMR

569.137 206 .000 2.763 .848 .814 0.77 .0787

Figure 4.1. Path model of RJP, role clarity and unmet expectation

40

Table 4.6.

Standardized Coefficient and T-values of Structural Equation Modeling

Path

Standardized coefficient

T-value

Job Preview → Role Clarity .12† 1.815

Role Clarity → Unmet Expectation -.33*** -3.548

Job Preview → Unmet Expectation .13* 2.332

CDMSE → Unmet Expectation .05 .933

Note. †p < .1, *p < .05, **p < .01, ***p < .001. CDMSE= Career Decision-Making Self-Efficacy

Sobel Test

In order to examine the mediating effect of role clarity between job preview and unmet expectation, Sobel test was conducted to test hypothesis 5: Role clarity mediates the relationship between job preview and unmet expectation. Through conducting Sobel test (Preacher & Leonardelli, 2017), the result indicated that role clarity did not carry the influence of the independent variable to the dependent variable as the critical z-value was 1.62, which did not meet the required threshold, 1.96 and the p-value was .11, which was higher than .05.

(Sobel, 1982) Consequently, hypothesis 5 was not supported.

Summary of Analyses Results

The purpose of the current study was to compare the differences between RJPs and TJPs as well as both job previews’ influence on individuals’ perception of role clarity and unmet expectation during pre-hire phase of recruitment. In order to analyze the proposed hypotheses,

41

three main approaches of test were conducted to examine the research hypotheses, including Pearson’s correlation test, t-test analysis, and path analysis. Through the performance of SPSS and AMOS, the results provide answers to the hypotheses proposed by the research.

Firstly, Pearson’s correlation test revealed the relationship between each variable, and it was found that job preview had significant relationship with post-test of job attraction, and role clarity had significant negative relationship with unmet expectation. In addition, career decision-making self-efficacy (CDMSE) as a control variable had significant correlation with role clarity and post-test of job attraction. Secondly, to examine first hypothesis, t-test analysis was conducted and the results showed that RJPs did not have stronger effect on increasing participants’ score on role clarity nor on increasing unmet expectation than TJPs; consequently, hypothesis 1a and 1b were not supported. At last, SEM analysis and Sobel test were performed to test hypotheses 2 to 5. Even though the result supported hypothesis 2 to 4 from the outcome of SEM analysis, role clarity has insignificant effect as a mediator on the relationship between the independent variable, job preview, and the dependent variable, unmet expectation. As a result, hypothesis 5 was not supported. The summary of the hypothesis testing is shown in table 4.7.

42

Table 4.7.

Hypothesis Testing Results Summary

Hypothesis Results

H1a

RJPs have stronger effect on increasing individuals’ perception of unmet expectation than TJPs do.

Rejected

H1b

RJPs have stronger effect on increasing individuals’ perception of role clarity than TJPs do.

Rejected

H2

Job preview brings positive effect on unmet expectations during pre-hire phase.

Supported

H3

Job preview brings positive effect on role clarity during pre-hire phase.

Supported

H4 Role clarity brings negative effect on unmet expectation. Supported

H5

Role clarity mediates the relationship between job preview and unmet expectation

Rejected

43