Rater Errors in Teacher Trainee Performance Assessments in a Nigerian University

Abstract

This study investigated the prevalence of rating errors in teacher trainee performance assessments within a Nigerian university, focusing on how evaluators applied the Teaching Practice Performance Assessment Scale (TPPAS). Teaching practice is a compulsory component of teacher education, yet its reliability is often undermined by subjectivity in rater judgments. Five raters who supervised postgraduate diploma trainees during their practicum were purposively selected, and their completed rating sheets were analyzed. The TPPAS, a validated instrument with a Cronbach’s alpha of 0.87, was used to capture multiple dimensions of teaching competence. Data analysis employed descriptive statistics alongside error-detection metrics using the R software, which allowed raters to be classified according to severity, leniency, central tendency, and halo effect biases. The results revealed that all raters demonstrated at least one form of error, with the halo effect emerging as the most widespread. Leniency bias was observed among three raters, while severity bias appeared in two, and central tendency was particularly evident in those who clustered scores around the midpoint of the scale. These findings confirm that subjectivity is systemic in teacher trainee evaluations and raise concerns about the fairness and validity of current practices. The study concludes that such biases compromise the credibility of teaching practice assessments and recommends the adoption of structured rater training, use of standardized rubrics, multiple raters, and regular monitoring systems to strengthen the reliability of teacher education assessments