NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED448210
Record Type: Non-Journal
Publication Date: 2000-Apr
Pages: 19
Abstractor: N/A
Reference Count: N/A
ISBN: N/A
ISSN: N/A
The Susceptibility of Item Parameters to Instructions for Completion.
McDonald, Jo-Anne
The purpose of this study was to examine the effect of completion instructions on item parameters and category use as students completed a self-report personality survey. Instructions allowed free-choice allotment of ratings (nonforced distribution) or requested the subject to assign a certain number of ratings to either the highest or lowest rating-scale categories (forced instructions). A total of 126 participant responses using nonforced instructions was obtained from a self-report survey completed at 1 company. The comparison sample of 346 forced distributions of ratings was collected from 5 companies. The hypothesis that alternatively worded instructions for survey completion would not impact item parameters was unsupported. The two sets of instructions produced nonequivalent patterns of response and item statistics. In addition, person separation and reliability were lower in the forced distribution condition. Data collected under the forced conditions did not fit the Rasch model as well as did the nonforced distribution data. Five appendixes contain the instructions for both conditions, two figures, and three tables of data. (SLD)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A