Title

In Search of Patterns at the Desk: An Analysis of Reference Question Tracking Logs

Document Type

Conference Presentation

Publication Date

May 2012

Abstract

This study is a content analysis of reference questions and answers asked at an academic library Information Desk during the Fall 2010/Spring 2011 academic year. An Excel file was created consisting of Reference question statistics recorded using the Gimlet (http://gimlet.us) question tracking system. A data dump was performed, resulting in separate Excel columns for service provider, duration of transaction, question format, Reference Effort Assessment Data (READ) Scale difficulty level*, month, time of day, day of the week, and free-text question and answer. The authors took the free-text question and answer fields and recoded them into the following fields: “Reference Tags,” “Subject/College,” “Exact Source,” and “Quality.” The new categories for each field were finalized after several rounds of 50-question sample calibrations by three coders. Then, a representative cross section of the 3,422 total questions was achieved by sampling from questions categorized at all six levels of the READ Scale. A random number table was used to take a 20% sample from all questions at levels 1-3 of the READ Scale; and all questions were included from levels 4, 5, and 6 for a sample size of 934 questions. Coders added coding information for the four new fields. The sample was then analyzed using the Statistical Package for Social Sciences (SPSS) to look for relationships among variables.