Examining the Factor Structure Underlying the TAP System for Teacher and Student Advancement

• Main Authors: Edward Sloat, Audrey Amrein-Beardsley, Kent E. Sabo
• Format: Article
• Language: English
• Published: SAGE Publishing 2017-10-01
• Series: AERA Open
• Online Access: https://doi.org/10.1177/2332858417735526

In this study, we investigated the factor structure underlying the TAP System for Teacher and Student Advancement using confirmatory and exploratory factor–analytic methods and under conditions of multilevel (nested) data structures and ordinal measurement scales. We found evidence of generally poor fit with the system’s posited first-order, three-factor structure with relatively large correlations among measured dimensions. Exploratory analysis suggests one to two interpretable factors, one of which accounts for the majority of explained variance (i.e., a general or common underlying factor). Higher-order modeling confirms the presence of a bifactor structure composed of a single general trait supported by one or two subscales. We use this evidence to question the validity of the inferences drawn from TAP subscale scores. We accordingly discuss implications for low- and high-stakes applications of TAP output, especially when consequential decisions are attached to subscale-level estimates (i.e., teacher compensation based on latent performance as rated through weighted subscales).