ABSTRACT
This chapter focuses on methods for exploring the empirical functioning of rating scales that provide evidence to inform the interpretation and use of ratings. It describes the degree to which individual rating scale categories are actually functioning across various facets of a rater-mediated assessment system. The chapter illustrates methods for evaluating the empirical functioning of rating scales based on polytomous Rasch models. Building upon these guidelines, G. Engelhard and S. A. Wind described a systematic approach to evaluating rating scale functioning using numerical and graphical evidence based on polytomous Rasch models. The chapter discusses guidelines and sources of evidence in terms of four major categories: category structure; directional orientation with the latent variable; category precision; and model-data fit. Focusing on rating scale categories, information functions for rating scale categories are illustrated using the example dataset. The chapter presents diagnostic evidence that provides researchers with tools for exploring raters' operational use of rating scale categories.
