Emotion extraction from Likert-Scale questionnaires: an additional dimension to psychology instruments

Abstract

Sentiment analysis tasks are used in various domains, including education. Likert-scale questionnaires are often used to gain insights into the respondents’ views in various contexts. However, these questionnaires can allow for more information than they are designed for. This research paper explores an emotion classification technique for extracting emotional information from likert-scale questionnaires. A case study is presented in which a tailored questionnaire was employed to gather students’ opinions on school-related matters, such as learning importance, academic performance and family and peer involvement and support. The students (n = 845) answered the questionnaire using a scale from totally disagree to totally agree. Through this questionnaire-based approach, data on students’ emotions was collected.