Social media discussions on educators: Selecting and appraisal of recent research using TF-IDF

Abstract

This systematic literature review, analyzing 36 peer-reviewed publications from 2019 to February of 2025, addresses a critical gap by examining the use of social media analytics (SMA) for faculty evaluation. Employing a novel methodological approach that combines machine learning-assisted screening (ASReview) with TF-IDF, the study finds that platforms like Twitter and Facebook are increasingly analyzed using sentiment analysis, machine learning, and text mining. These techniques provide real-time, unfiltered student feedback on teaching effectiveness, complementing traditional evaluation instruments and helping to monitor institutional reputation. While SMA offers valuable insights, the review highlights significant challenges, including data quality and credibility, algorithmic bias, ethical concerns, and generalizability. Effectively leveraging SMA's potential requires addressing these issues through robust theoretical frameworks, balanced institutional policies, and enhanced digital literacy to improve teaching practices while safeguarding academic integrity.