This presentation summarizes the results of the analysis of the impact of changes in questionnaire format in o large-scale undergraduate survey program - the National Survey of Student Engagement (NSSE). Since 1999, NSSE has been administered at about 1,100 four-year Institutions in the U.S. and Canada with more thon 1,000,000 students responding (www.nsse.iub.edu). The NSSE has both a paper and a web component. Two changes were implemented in 2005 that effected the web version of the survey. First, the informed consent information was placed on the first page of the survey and not sent in the recruiting messages. The informal consent page also had a graphic that was specific to each school. Second, prior to 2005, the students had only 4 submits throughout the questionnaire. In 2005, the number of submits increased to 17. As a result, many questions were located in different places in grids. Placing the informed consent statement on the first page along with a school-specific graphic appears to have increased the proportion of students who visited the website and started the survey. However, the abandonment rate increased and the survey was abandoned sooner than previous years when there were only four submits. The changes in question placements in screen grids appear to have affected the responses. Many items that changed location in the web version had different distributions between 2004 and 2005 but there were no substantial differences in the same questions for the paper questionnaire which did not change formats. ln the presentation, we will describe these changes more fully to include an analysis of the characteristics of students who abandon the survey and the possible differental impact of changing question locations for different types of students. We will also present 2006 data which will help understand the longer-term implications of these changes.