High-quality vs low-quality teaching

Abstract

Purpose The purpose of this study is to examine student sentiments regarding high-quality vs low-quality teaching. Design/methodology/approach This study uses a text mining technique to identify the positive and negative patterns of student sentiments from student evaluations of teaching (SET) provided on Ratemyprofessors.com. After identifying the key positive and negative sentiments, this study performs generalized linear regressions and calculates cumulative logits to analyze the impact of key sentiments on high- and low-quality teaching. Findings Results from 6,705 SET provided on Ratemyprofessors.com indicated that students express different sets of sentiments regarding high- vs low-quality teaching. In particular, the authors found positive sentiments such as passionate, straightforward, accessible, hilarious, sweet, inspiring and clear to be predictive of high-quality teaching. Additionally, negative sentiments such as disorganized, rude, difficult, confusing and boring were significantly related to low-quality teaching. Originality/value This study is one of the first few studies confirming that high- and low-quality teaching are not completely opposite to each other from the student’s perspective. That is, the presence of high-quality teaching does not necessarily mean the absence of low-quality teaching. As such, this study provides an important theoretical base for future researchers who wish to explore approaches for improving faculty teaching in the higher education setting. Additionally, this study offers educators some recommendations that may help students experience positive sentiments while minimizing negative sentiments.

The Student Evaluation of Teaching Premium for Clinical Faculty in Economics

“…RMP provides the largest publicly available online SET data source, and it has been widely employed in the business education literature (e.g., [45,46]). Many studies discuss straightforward analyses of the determinants of RMP ratings.…”

Contextualizing gender disparities in online teaching evaluations for professors

“…We performed topic analysis using BERTopic at the sentence level for reviews, recognizing that each review may encompass various aspects of teaching and differing emotional valences within different sentences [26,69]. Segmenting comments into sentences is also recommended for using BERTopic to identify multiple topics [70].…”

Exploring the influence of student emotions and professor behaviour on course ratings: a quantitative analysis

“…There are no technical barriers to reforming the SET as advances in automated text analysis in the past decade allowed researchers to study the corpora of student evaluations of teaching. These studies applied finding keywords and phrases (Subtirelu, 2015(Subtirelu, , 2017Park, 2019;Murray et al, 2020), performed sentiment analysis (Chou et al, 2020;Okoye et al, 2022) or topic modelling (Azab et al, 2016). Recurrent neural networks have been used to analyse 154,000 instructor reviews (Onan, 2019), and they achieved higher accuracy than the conventional machine learning classifiers on the sentiment analysis task.…”

“…Several works discuss the correlation between the sentiment of the student reviews and the course ratings (Chou et al , 2020; Okoye et al , 2022) or the perceived course difficulty (Felton et al , 2008). However, for the first time in the literature, this paper applies an advanced natural language processing (NLP) method called zero-shot learning to extract more complex student emotions rather than sentiment from the large corpus of one million student reviews.…”