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

variable; the absence of bias from a single data source is confirmed when the VIF falls below 3.3. As the analysis produced a VIF below 3.3, single-source bias did not appear to be a signifi- cant concern with these data. A significant concern for the multilevel conceptual model was the potential for Omitted Variable Bias at the individual level [55], especially since the “Student” level had no other con- trol variables. However, as justified by [56], it is impractical to collect all requisite data since the present study is strictly observational. Measurement model Based on the recommendations put forth by [57], the constructed model underwent two-step testing. Before proceeding, the instrument’s validity and reliability were assessed on the mea- surement model per the protocols outlined by [51, 58]. Then, the developed hypothesis was evaluated by executing the structural model. The loadings, average variance extracted (AVE), and composite reliability (CR) for the mea- surement model were evaluated. The CR should exceed 0.7, the loading values should exceed 0.5, and the AVE value should exceed 0.5. AVEs and CRs exceeded 0.5 and 0.7, respectively, as shown in Table 2. Convergent validity. The Composite Reliability (CR) values for Employability Skills of Students (ES), Quality Education (QE), Course Structure (CS), Institutional Policy (Pol), and Physical Aspects (PA) were 0.836, 0.857, 0.823, 0.891, 0.808 and 0.798, respectively indicated that all items constantly measure the same construct. Furthermore, contemporary research

Table 2. Measurement of construct validity* (Source: Authors). Construct Items Loadings CR AVE Course Structure CS1 0.717 0.823 0.609 CS2 0.783 CS3 0.836 Employability Skills of Students ES1 0.716 0.836 0.507 ES2 0.797 ES3 0.64 ES5 0.654 ES7 0.743 Physical Aspects (PA) PA1 0.752 0.798 0.500 PA2 0.659 PA4 0.776 PA5 0.629 Institutional Policy Pol1 0.691 0.808 0.585 Pol2 0.783 Pol3 0.816 Quality Education QE1 0.742 0.857 0.500 QE2 0.75 QE3 0.777 QE4 0.667 QE5 0.643 QE7 0.652 * Note: Two items were dropped from employability skills of students (ES4, ES6), one item was dropped from physical aspects (PA3), and one item was dropped from quality education (QE6) because of their low loadings (below 0.5). https://doi.org/10.1371/journal.pone.0310815.t002 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 9 / 18