Relationship between mental health and job competence of college teachers: Application of sentiment analysis algorithm

Abstract

In systematically collecting and investigating data related to the teaching field, the association between teachers’ Mental Health (MH) and Job Competence (JC) is an important field that has not been systematically studied. Understanding this association is vital due to its direct impact on the Quality of Education (QoE) and its standard educational system. Due to problems in evaluating all qualitative data, such as Open-Ended Survey Feedback (OESF), traditional models frequently find it challenging to epitomize intricate relations accurately. This article reports on these tasks by introducing a Mixed-Method Approach (MMA). The research study was directed through online learning across numerous higher education institutions in three provinces of China, and a combination analysis of quantitative study data using qualitative Sentiment Analysis (SA) was recommended. The new thing about the method is that it suggests an algorithm based on Latent Dirichlet Allocation (LDA) to SA that lets all of the qualitative data from the OESF questions be studied. This algorithm suggests a more philosophical knowledge of teachers’ MH and its association with their specialized skills overall OESF. The study’s results represent a significant insight into the dynamics between MH and JC related to College Teachers (CT). It highlights how every feature impacts others and the prerequisites for educational strategies supporting teacher well-being. By overcoming boundaries in existing models, the proposed work contributes to a broader and better knowledge of teacher well-being, its impact on educational quality, and the potential for SA in educational research. We compared how well three classifiers—Naïve Bayes (NB), Support Vector Machine (SVM), and Linear Regression (LR)—performed on six topics that were chosen for this SA research. The performance analysis for evaluation is accuracy, precision, recall, and F1-score. The reliability of the OESF measures was confirmed with Cronbach’s alpha values signifying high internal consistency: 0.85 for JC, 0.88 for MH status, and 0.82 for innovative teaching ability.