This study compared 4,594 student responses from three different surveys of incoming students at the University of South Florida (USF) with data from Florida's State University System (SUS) admissions files to determine what proportion of error occurs in the survey responses. Specifically, the study investigated the amount of measurement error in student responses to questions about application and admission activities to universities other than the one at which they were enrolling. A literature review is included that examines the problem of measurement error in other studies that used survey or self-report measures. The study found that considerable measurement error can exist in self-report measures even when a subject is reporting simple factual information; in this case the level of unbiased error was about 4 percent, and biased error was about 20 percent. The amount of error was directly proportional to the severity of the memory demands as well as the characteristics of the survey population. It was noted that with all but the smallest sample sizes, the measurement error will likely be larger than the sampling error, and that psychological forces, such as social desirability and cognitive dissonance, can create substantial measurement bias. Recommendations for future research are provided. (Contains 15 references.)