Publisher
University of Tennessee at Chattanooga
Place of Publication
Chattanooga (Tenn.)
Abstract
Emotional Intelligence (EI) is a rapidly expanding field in Psychology, but the term has often been interpreted too broadly, frequently being equated with good character or social skills, such as personality traits. However, EI should be understood as the ability of individuals to react effectively to others based on the emotional information. The purpose of this study is to explore the plausibility of indirectly measuring emotional intelligence as an ability through an Artificial Intelligence chatbot. This chatbot extracts various textual features from users’ free-text responses collected during online conversations and employs machine-learning algorithms to infer emotional intelligence scores that will be resulted in four facets, (1) ‘Identifying Emotions’, (2) ‘Facilitating Emotions’, (3) ‘Understanding Emotions’, and (4) ‘Regulating Emotions’. The psychometric properties of machine-inferred scores will be examined, and those are internal consistency split-half reliability, factorial validity, convergent validity and discriminant validity, and criterion-related and incremental validity. For the method, five hundred participants will be recruited through SONA, a research management system. Participants in the training sample (n=400) will complete a list of Situational Judgment Test (SJT) items to measure their emotional intelligence on Qualtrics, and they will interact with an AI chatbot for approximately 45 to 60 minutes. Participants in the test sample (n=100) will undergo the same study procedure as the training sample and go through additional steps such as providing their college GPA, ACT and/or SAT scores, and completing a couple of surveys: an IQ (g) test and a life satisfaction scale to assess the criterion variables on academic performance, life satisfaction, and cognitive ability.
Subject
Industrial and organizational psychology
Document Type
posters
Language
English
Rights
http://rightsstatements.org/vocab/InC/1.0/
License
http://creativecommons.org/licenses/by-nc-sa/4.0/
Included in
Measuring an ability-based emotional intelligence through an AI chatbot
Emotional Intelligence (EI) is a rapidly expanding field in Psychology, but the term has often been interpreted too broadly, frequently being equated with good character or social skills, such as personality traits. However, EI should be understood as the ability of individuals to react effectively to others based on the emotional information. The purpose of this study is to explore the plausibility of indirectly measuring emotional intelligence as an ability through an Artificial Intelligence chatbot. This chatbot extracts various textual features from users’ free-text responses collected during online conversations and employs machine-learning algorithms to infer emotional intelligence scores that will be resulted in four facets, (1) ‘Identifying Emotions’, (2) ‘Facilitating Emotions’, (3) ‘Understanding Emotions’, and (4) ‘Regulating Emotions’. The psychometric properties of machine-inferred scores will be examined, and those are internal consistency split-half reliability, factorial validity, convergent validity and discriminant validity, and criterion-related and incremental validity. For the method, five hundred participants will be recruited through SONA, a research management system. Participants in the training sample (n=400) will complete a list of Situational Judgment Test (SJT) items to measure their emotional intelligence on Qualtrics, and they will interact with an AI chatbot for approximately 45 to 60 minutes. Participants in the test sample (n=100) will undergo the same study procedure as the training sample and go through additional steps such as providing their college GPA, ACT and/or SAT scores, and completing a couple of surveys: an IQ (g) test and a life satisfaction scale to assess the criterion variables on academic performance, life satisfaction, and cognitive ability.
Department
University of Tennessee at Chattanooga. Dept. of Psychology