Prediction of study success – should selection instruments measure cognitive or non-cognitive factors?

Prediction of study success – should selection instruments measure cognitive or non-cognitive factors?

[featured_image]
  • Version
  • Download 101
  • File Size 98.64 KB
  • File Count 1
  • Create Date August 2, 2018
  • Last Updated August 2, 2018

Prediction of study success - should selection instruments measure cognitive or non-cognitive factors?

This study compares the subtests of a traditional admissions test (the SweSAT) with high school grades in verbal and quantitative subjects in terms of construct and predictive strength as selection instruments to higher education. The purpose is to find out if and how the difference in predictive validity can be related to the properties of these instruments. The data is analysed by factor analysis, structural equation modelling (SEM) and regression analysis. The findings show that there is a common verbal factor in the verbal subtests and the English grade, and a quantitative factor shared by the quantitative subtests and also the maths grade. A third and separate factor is also detected in the grades in Swedish. This is assumed to be a grade related factor incorporating non-cognitive skills. When comparing the predictive strength of these instruments, with credits from the freshman year in a Swedish economics programme, the pattern is not entirely clear, but tests and grades loading on the quantitative factor as well as the x-factor can explain some of the variation on the criterion for academic performance (first year credits). None of the verbal sub-scores in the SweSAT correlates positively with academic achievement.

Attached Files

FileAction
paper_4d73dda.PDFDownload