Reading Between The Lines: A Text Analytics Exploration Of Social Emotional Learning

Research has shown that social emotional learning (SEL) is being increasingly valued by schools due to its associations with academic achievement and student engagement. Unfortunately, the current state of SEL, with its lack of time or resources allocated by practitioners and prevalence of self-report assessment, restricts scalability and does not meet the forecasted demand. Additionally, difficulties in measuring SEL persist. This paper will present that SEL assessment needs to be innovated to help address the limitations of self-report measurements and necessity for a scalable solution. Text analytics and natural language processing (NLP) serve as a flexible, low-lift assessment method that can analyze contextual differences addressing these limitations of existing SEL assessment. Therefore, this study presents a text analytics and NLP evaluation of a proposed text-based SEL assessment of growth mindset assessing if analysis of text message conversations between agents and students can be used to assess a student’s level of SEL. Conducting a review of the relevant literature, to the best of our knowledge this is the first study to assess a text-based SEL assessment using a text analytics and NLP approach. Ultimately, this study created five prediction models for growth mindset scales with predictive validities between .37 and .43.