Measuring the mediating role of quality education for ensuring employability skills: An analysis of higher education student perception in Bangladesh

VIF = 2.075). The relationship between Pol and QE is not supported in H2 (β = 0.092, T Value = 1.195, P Value = 0.232, F2 = 0.016, VIF = 1.538). The study of the relationship between QE and ES is supported in H4 (β = 0.612, T Value = 9.557, P Value = 0.000, F2 = 0.597, VIF = 1.000). Testing the mediating effect. An investigation was undertaken to determine whether QE mediated the relationships between PA, CS, Pol and ES among students in higher education. QE (H6: β = -0.238, t = 4.289, p = 0.000, H9: β = 0.235, t = 3.716, p = 0.000) mediated the relationship between PA, ES and ES in higher education. QE (H6, H7) did not mediate the relationship ES between CS and Pol. Thus, these results are shown in Table 5. Predictive relevance analysis (R2 & Q2). The R2 value indicates the accuracy with which the independent variables of a model predict the dependent variable [61]. The R2 cut values of 0.75, 0.50, and 0.25 were categorised by [51] as significant, moderate, and feeble, respectively. The findings of this research demonstrate that the endogenous construct is explicable (QE: 0.650 and ES: 0.374), suggesting that the exogenous variables (PA, CS and Pol) account for around 37% and 65% of the variance, respectively. The suitability of this research model is cer- tified by the R-square criterion. Exogenous constructs are observed to predict endogenous constructs when the Q2 value is more significant than zero [44]. QE: 0.609 and ES: 0.415 indicate that the model is statistically highly predictive. The results presented in Table 6 demonstrate that the model possesses pre-

dictive validity. Discussion Direct effects of constructs on quality education This research contributes to the vital understanding of HEIs in Bangladesh and the student perception of how administrators and policymakers shape the curricula and other features. Prior literature has described and measured a solid direct effect of Course Structure, Physical Aspects, Institutional Policy, Quality Education and Employability Skills of Students. The current study has underlined similar findings as per the survey conducted on 154 students. Physical Aspects (PA) and Quality Education (QE) have all supported a direct effect with Quality Education (QE) and Employability Skills of Students (ES). [18] opined that the ‘tan- gibles’ (physical aspects) of any service determined the direction of the gap between per- ceived and experienced service. Some other items to which students agreed were "My university offers the facilities of my choice." (44.9%), "My university offers various support services." (46.2%), and "My university campus is attractive." (48.7%), and strongly agreed "I Table 5. The result of the mediating effect (Source: Authors). Hypothesis Path Std. Beta t-value p-value Decision H5 PA -> QE -> ES 0.238 4.289 0.000 Supported H6 CS -> QE -> ES 0.035 0.624 0.532 Not Supported H7 Pol -> QE -> ES 0.056 1.167 0.243 Not Supported https://doi.org/10.1371/journal.pone.0310815.t005 Table 6. Predictive relevance of the exogenous constructs (Source: Authors). Construct R-square Q-square ES 0.374 0.415 QE 0.650 0.609

https://doi.org/10.1371/journal.pone.0310815.t006 PLOS ONE Measuring the mediating role of quality education for ensuring employability skills PLOS ONE | https://doi.org/10.1371/journal.pone.0310815 October 7, 2024 11 / 18