Optimal Scale Points for Reliable Measurements: Exploring the Impact of Scale Point Variation

Abstract

Ensuring reliable measurements is crucial for minimising errors in assessments. The assessmentcommunity commonly employs the evaluation of reliability coefficients to estimate the dependability oftest scores. Despite its significance, limited research has explored the relationship between the estimated reliability coefficient and the number of scale points utilised. This study aims to provide valuable insights to practitioners by investigating the optimal number of scale points required for the most accurate reliability coefficient estimation. Using simulated data, the research scrutinises scales with varying points, ranging from 2 to 11. The results reveal a substantial impact of the number of scale points on reliability estimation. The most accurate estimate of reliability is obtained for scales with 8 points. This study helps us understand the optimal number of scale points for reliable measurements and guides future assessment improvements.