Likert Response Alternative Direction : SA to SD or SD to SA: Does It Make a Difference? / J. Jackson Barnette.

A 20-item survey was designed in 4 forms with response set direction as "strongly disagree" (SD) to "strongly agree" (SA) and SA to SD crossed with the absence or presence of negatively worded item stems. The primary research question related to finding a primacy effect when comparing the two response direction formats. Surveys were administered, randomly by form, to 586 high school students, undergraduate and graduate students, and teachers. There were no differences in internal consistency reliability, total score means, total score variance, item means, item standard deviations, or item-to-total correlations. The presence or absence of negatively worded stems did not affect any of the survey statistics except Cronbach's alpha where there was about a 0.1 lower value when mixed stems were used (half positive and half negative) compared with having all positive or direct stems. However, there was no interaction effect with the response direction variable. It is concluded that for this survey there was no primacy effect. Reasons are proposed for why such an effect was not observed here when others have detected such an effect. These relate to differences in types of surveys, the focus of the survey, and the relationship of the topic and the respondent and how they may influence the likelihood of observing a primacy effect. (Contains 4 tables and 13 references.) (Author/SLD)

Microfiche. [Washington D.C.]: ERIC Clearinghouse microfiches : positive.