Chat Reference Analysis to Determine Accuracy and Staffing Needs at One Academic Library

Abstract

This study focuses on the evaluation of one year of chat reference transcripts at the University of Connecticut (UConn) Libraries. Over 3,000 session transcripts were examined and coded to determine type, accuracy, instruction, and escalation of questions received through our chat reference service to identify potential staffing and training needs. The findings reveal that the librarians providing virtual reference services are extremely accurate in their answers (93%), but the type of questions received and the rate of referral to other staff and services indicate a potential need for more and varied staffing and training solutions.