The Hierarchical Rater Thresholds Model for Multiple Raters and Multiple Items

Abstract In educational measurement, various methods have been proposed to infer student proficiency from the ratings of multiple items (e.g., essays) by raters. However, suitable models quickly become numerically demanding or even unfeasible as separate latent variables are needed account for local dependencies between same response. Therefore, in present paper we derive a flexible approach based on Thurstone’s law categorical judgment. The advantage this is that it can be fit using weighted least squares estimation which computationally less compared most previous approaches case an increasing number variables. addition, new applied existing variable modeling software. We illustrate model real dataset Trends International Mathematics and Science Study (TIMMSS) comprising 10 4 raters 150 subjects. compare including facet model, hierarchical rater class model.