Open AccessSECULAR TRENDS IN AN ADJECTIVE CHECKLIST 1, 2
Author(s) -
Hofstee Willem K. B.
Publication year - 1964
Publication title -
ets research bulletin series
Language(s) - English
Resource type - Journals
eISSN - 2333-8504
pISSN - 0424-6144
DOI - 10.1002/j.2333-8504.1964.tb00511.x
Subject(s) - adjective , test (biology) , trait , psychology , computer science , econometrics , statistics , cognitive psychology , mathematics , artificial intelligence , noun , paleontology , biology , programming language
Secular trends in tests were redefined as effects of systematic changes in response to test items resulting from experience with similar items, rather than from learning outside the testing situation. In the present study, two such trends were investigated in a list of 180 trait‐descriptive adjectives: a trend for items to be responded to more desirably as the list continues, and a trend for items to be responded to more or less desirably depending upon the desirability of the preceding item. The method consisted of comparing item popularities, as found in the original format, with the popularities of the same items presented in forced‐choice format: if item popularities had been biased through the operation of secular trends in the original list, such bias should to some extent be detectable in the responses to the forced‐choice format. The results showed no evidence for the first trend; with respect to the second, it was clearly demonstrated that items in the original list were responded to more desirably as the preceding item was more desirable. The importance of controlling and studying secular trends was discussed.