Measurement Error in Social and Educational Survey

The Cooperative Institutional Research Program (CIRP) of the American Council on Education is an example of a systematic program of social and educati,mal research in which survey methods using questionnaires constitute a major technique for collecting data. This longitudinal research program is dedicated to the study of higher lducation, particularly its impact on student development. In such a program, measurement error may affect data analysis and interpretation at each state of the research system through which data are transmitted and processed. That is, inaccuracy or imprecision in recorded response and in the statistics based on these data may occur not only during the solicitation of data but also luring information transcription and file development.-The purpose of this report is to present empirical results about certain problems of measurement error in an educational survey system. These results are confined to measurements derived from items in the annual surveys of entering freshmen that are a basic part of the CIRP. After discussion the meaning and consequences of measurement error, the paper presents a review of the pertinent literature, and then presents the results of recent empirical investigations. (Author/HS) No st