A Statistical Assessment of Order of Presentation in Multiple Item Likert Scale Survey Responses, or "Who's on First?" [microform] / Chester H. McCall, Jr. and Suzanne Gardner

The Research Services of the National Education Association (NEA) conducted a nationwide teacher opinion poll (TOP) based upon a stratified disproportionate two-state cluster sample of classroom teachers. This research study was conducted to test the hypothesis that the order of presentation of items would make no difference in the conclusions which might be drawn. Five specific issues were assessed on two different forms of the survey. The issues were various aspects of teaching, standardized tests, tuition tax credits, job the President is doing, and influence of special interest groups on governmental policies. The only difference between the forms was the order in which the items were presented to the respondents. Comparisons were made between the opinion responses as well as the demographics of the two groups. Statistically significant differences were found which suggest that position may have an impact on the response distributions. The test forms are reproduced in this document. (Author/DWH)