1. Introduction

The Auraria Library is an academic library serving three institutions in downtown Denver: the Community College of Denver, Metropolitan State University of Denver, and University of Colorado Denver. I work as a graduate assistant in Education and Outreach Services (or EOS). Our job is to tell students about the library in addition to providing reference help and information literacy instruction. Our department strives to consistently assess our instruction, reference, and outreach on campus. Summer is the perfect time to work on projects that we can’t prioritize during the year--one of these projects inspired Project Spring Fling.

 

One of EOS’s major projects for the summer is overhauling all of our online tutorials. In order to create tutorials that are relevant and useful to our first-year students, we as a department needed to consider what topics need that extra bit of explanation. To supplement our suggestions, we surveyed students at an annual event called Spring Fling. Each student that visited our table was asked to write their answer to a question: what should a first-year student know about the library? My boss had brought up text analysis as a way of analyzing the students’ responses, and I jokingly asked if I could head that assessment. When he [seriously] agreed, I jumped at the opportunity to experiment a little and still provide a valuable contribution. 

​

On the foundation of my boss’s suggestion, I decided to create a digital humanities (or DH) project that would serve as an assessment tool. Although I was planning to do some text analysis, I primarily wanted to topic model the responses. Since we needed to find topics for tutorials, I thought topic modeling would be the most useful method. I used both topic modeling and text analysis tools to analyze the 300+ student responses we collected.

 

My research questions for this project were simple and pointed. 1. What topics would be well-suited to the tutorial format? 2. What formats would fit the other topics? These questions would be easy to answer and would provide the foundation for the assessment I wanted to conduct. My overall goal was to find and analyze topics in students’ responses.

 

I was expecting this process to be fairly straightforward--topic modeling, text analysis, answered questions. The end. The reality, though, turned out to be a little messier.