Survey analysts are often challenged by open-ended survey comments, which interject textual data into an otherwise quantitative analysis task. In this presentation, we describe our team’s mixed-methods approach to sampling and analyzing lengthy open-ended comments for a large-scale military survey. We hope to show participants that qualitative survey data, even in large quantities, can be analyzed rigorously and painlessly. In the summer of 2010, the U.S. Department of Defense engaged Service members about the perceived impacts on the military if Don’t Ask, Don’t Tell (the law banning openly gay individuals from military service) were repealed. Westat supported this effort with numerous data collection activities, including an online survey with a sample of 400,000 Service members. Responses to the open-ended item on this 103-item instrument were critical to understanding respondents’ perceptions on repeal, but also presented an enormous challenge: Of the approximately 115,000 respondents, over 46,000 wrote comments, many of which were more than a paragraph long. How could we analyze so much data in a cost-effective and analytically rigorous way, and that met our client’s quick turn-around timeframe? In the presentation, we first describe our sample selection process, an important facet of which was classifying respondents as having “positive,” “negative,” or “mixed” attitudes towards repeal based on their responses to 21 critical survey items. We then discuss our comment analysis, including the development of a coding scheme, our use of a qualitative software package to manage the data, and, ultimately, how the key issues differed by respondent classification.