6533b81ffe1ef96bd1276ea8

RESEARCH PRODUCT

Contributions and limitations of self assessment of competences by higher education graduates

subject

description

This work presents a continuum of competence. On one extreme of which there is a subtlety of concept of competence. On the other, there are rigorous econometric analyses to find the answer to a critical question: ―to what extent is the assessment of competence reliable? ‖ Quest into the etymology helped us providing a rationale in favour of the preferred use of competence over competency. Moreover, we come to trace its roots in French literature where it was in use long before its emergence in English literature. We found that competences were common among various geopolitical locations and inferred that interest in competence has been internationalised realistically. Both English and French literature helped us in this regard. Before carrying out econometric analyses it is apt to probe into the literature on the assessment of competence. Rigorous analyses revealed that self assessment of competence is reliable to a modest level. We probed into the reliability of competence (self) assessment by the higher education graduates and by the young knowledge workers. We come to affirm reliability for both of the cases; however, to a modest extent. We simplified the research question by bifurcating it as in the following. 1. To what extent is the self assessment of acquired levels of competence by the higher education graduates reliable? 2. To what extent is the assessment of required levels of competence by the young knowledge workers reliable? We tested for both of these questions and found affirmative response each time. Karl Popper's criterion of falsifiability facilitated us to accept the theses that 1. firstly, the self assessment of acquired levels of competence by the higher education graduates is reliable, however, to a modest extent. 2. secondly, the assessment of required levels of competence by the young knowledge workers' is reliable, however, to a modest extent. The data set used in this study was provided by the Reflex team. We made use of SPSS and Stata for the analyses. A number of statistical techniques have been manipulated including ordered probit, OLS regression, and parametric and non parametric analyses of variances. We observed startling similarity in the levels of significance of coefficient estimates of ordered probit and OLS regression. Large data sets may exhibit such behaviour. This might invoke some deeper reflection for statisticians. The fact that we remained confined to one data set is main limitation to this study. We suggest its replication with more variables and with other data sets of comparable size.