The Effect Of Item Weighting On Reliability And Validity

Abstract

The purpose of this study is to examine the effect of the item weighting method developed by researchers on the construct validity of the test. For this purpose, a Monte Carlo simulation study was carried out. Test length, average factor loadings, and sample size were considered as simulation conditions. Item weighting method was defined as follows: If average score of the individuals (calculated as individual's test score/the number of items) plus item difficulty index is 1 and over then item reliability index added to individual's item score (1 or 0); if not, then the item score of the individual (1 if 1, 0 if 0) is preserved. As a result of the research, it was observed that the weighting method contributes to the construct validity. According to the results of confirmatory factor analysis, the comparative fit index (CFI) and the root mean square error of approximation (RMSEA) values were improved. According to the research findings, the weighting method used in this research can be recommended.