Modeling Motivational Interviewing Strategies on an Online Peer-to-Peer Counseling Platform

Millions of people participate in online peer-to-peer support sessions, yet there has been little prior research on systematic psychology-based evaluations fine-grained peer-counselor behavior relation to client satisfaction. This paper seeks bridge this gap by mapping chat-messages motivational interviewing (MI) techniques. We annotate 14,797 utterances from 734 chat conversations using 17 MI techniques and introduce four new codes such as ''chit-chat'' ''inappropriate'' account for the unique conversational patterns observed platforms. automate process labeling responses fine-tuning large domain-specific language models then use these automated measures investigate peer counselors via correlational studies. Specifically, we study impact conversation ratings that predict clients' satisfaction with their counseling sessions. When reflection affirmation, clients are more satisfied. Examining volunteer counselors' change usage suggest learn introduction open questions they gain experience. work provides a deeper understanding counselor platforms sheds light how build better training programs