NotesFAQContact Us
Search Tips
ERIC Number: ED307313
Record Type: Non-Journal
Publication Date: 1989-Mar
Pages: 29
Abstractor: N/A
Reference Count: N/A
Test Anxiety and Item Order: New Parameters for Item Response Theory.
Gershon, Richard C.
Examinees (N=1,233) at the Johnson O'Connor Research Foundation (JOCRF) were administered one of three test forms in which only item order differed. The study was undertaken to determine the validity of the assumption underlying item response theory (IRT) that there are fixed item parameters that can predict performance. The Rasch IRT model was chosen. The three experimental tests were constructed from 950 items found in the JOCRF's item bank. The population seen at the JOCRF ranges in age from 14 to 60 years. Personality tests administered to subjects included Mandler and Sarason's Test Anxiety Scale and a short series of test-taking strategy items. All subjects took a minimum of 18 aptitude tests. Three primary factors were included in the analyses: (1) level of test anxiety; (2) ability; and (3) difficulty order. Results of the study indicate that item difficulty order, test anxiety, and ability all affect performance, violating assumptions of IRT. It appears that the theory neglects to address the effects of individual differences on test behavior. Adaptive testing techniques are reviewed, and the theories of test anxiety and associated measurement instruments are discussed. New testing strategies are proposed in which personality variables and test characteristics can be incorporated as parameters, allowing tests to be manipulated in a manner that maximizes performance. Three tables and nine graphs present study data. A 55-item list of references is provided. (TJH)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A