NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED205595
Record Type: RIE
Publication Date: 1981-Apr-14
Pages: 12
Abstractor: N/A
Reference Count: 0
ISBN: N/A
ISSN: N/A
Methodological Issues Related to the Study of Context Effects in Multisection Tests.
Stewart, E. Elizabeth
Context effects are defined as being influences on test performance associated with the content of successively presented test items or sections. Four types of context effects are identified: (1) direct context effects (practice effects) which occur when performance on items is affected by the examinee having been exposed to similar types of material earlier in the test; (2) indirect context effects occurring when the examinee has been exposed earlier in the test to a different type of material; (3) serial position effects which result from the amount, rather than the type, of material that has preceded a particular section; and (4) motivational effects, when the examinee believes that a particular section is non-operational and does not count toward his score. Various methods for controlling these effects are suggested and a hypothetical test format, containing two operational sections and one of two non-operational sections with twelve possible sequences, is presented to illustrate the ways in which context effects may be statistically assessed. Finally, it is recommended that if there are too many sections in a test to adequately assess context effects in one test administration, a data collection plan allowing comparisons over a number of test administrations should be developed. (Author/AEF)
Publication Type: Speeches/Meeting Papers; Opinion Papers; Reports - Evaluative
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Identifiers: Operational Questions; Standard Deviation
Note: Paper presented at the Annual Meeting of the National Council on Measurement in Education (Los Angeles, CA, April 14, 1981).