2026-09-21T23:30:07Z https://www.ideals.illinois.edu/oai-pmh

oai:www.ideals.illinois.edu:2142/68781 2023-07-11 col_2142_5131 col_2142_8800 com_2142_5130 com_2142_5132

Ku, Earl Juang-Ell 2014-12-15T18:48:50Z 2014-12-15T18:48:50Z 10000-01-01 1982 1982 Most current foreign language listening comprehension tests require examinees to read and understand multiple choice (MC) items printed in the test booklet in order to make responses. As a result, reading is confounded with listening in such tests. This study proposed a true-false (T-F) listening test as an alternative to reduce the confounding problem in the traditional listening test. The specific hypotheses for this study were: (1) the disattenuated correlation between the reading and the traditional listening examinations is not different from 1.00; (2) the disattenuated correlation between the reading and the T-F listening examinations is less than 1.00; and (3) there is a difference between the two disattenuated correlations. Subjects included 150 Chinese students at the University of Illinois. Data collection instruments included the listening and reading comprehension subtests from TOEFL. The T-F listening test form was constructed by converting the TOEFL listening MC items to T-F items. The disattenuated correlations between the scores on the reading test and the traditional listening test and the T-F listening test were each significantly different from 1.00. There was no significant difference between these two disattenuated correlations. The most plausible explanation for the similarity between the correlations related to the apparently high reading achievement level of the samples which resulted in a low test ceiling. The hypothesized reading confounding was essentially non-existent because the reading was so easy for these particular groups. Two conclusions were drawn from this study. First, the disattenuated correlation between the scores on the reading test and the traditional listening test or the T-F listening test suggested that either listening test is measuring unique skills not shared by the reading test. Because there was no significant difference between these two correlations, both listening test forms seemed to provide equally unique information about listening comprehension. Second, the traditional listening test items were more effective than the T-F listening test items in measuring listening abilities in a foreign language. The traditional listening test had a higher discrimination power than the T-F listening test and its items were more difficult. Made available in DSpace on 2014-12-15T18:48:50Z (GMT). No. of bitstreams: 1 8209599.pdf: 3616286 bytes, checksum: a68d972b69bfdcf522bb4a51fd31d5d7 (MD5) Previous issue date: 1982 Embargo set by: Seth Robbins for item 68947 Lift date: Forever Reason: Restricted to the U of I community idenfinitely during batch ingest of legacy ETDs Restricted to the U of I community idenfinitely during batch ingest of legacy ETDs U of I Only 104 p. Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1982. http://hdl.handle.net/2142/68781 (UMI)AAI8209599 Education, Tests and Measurements An Examination of the Confounding in Measures of Foreign Language Listening Comprehension text Education Education University of Illinois at Urbana-Champaign Dissertation Ph.D.