An attitude-interest questionnaire was especially designed for administration to entering female college freshmen. It has been administered to the members of two successive freshman classes at an eastern women's liberal arts college; in one case shortly after the students had been admitted to the college, and in the other case when the students were applying for admission. Although questionnaires were identified, in both cases the respondents were assured that their answers would not be made known to the college authorities. Of 145 items in the questionnaire, the answers given to 90 were found to discriminate significantly (CR of 2.00 or higher) between the preadmission and post- admission groups. A qualitative analysis of these differences appeared to indicate that the preadmission respondents were more inclined to rationalize their responses than were the postadmission respondents. It was also found that the preadmission group gave more extreme responses, and this was shown to be significant at the 1 per cent level. From a subsidiary finding the question is raised as to whether or not differences found in magnitude of DR's may be a function of the different lengths of response continua provided for different sections of items.