The present protocol analysis supplements Sasaki's(1993) study, which used structural equation modeling to investigate the relationships among measures of second language proficiency (SLP), foreign language aptitude, and two types of intelligence (verbal intelligence and reasoning). In the present study, 6 participants randomly selected from the same subject pool took different types of SLP tests, a foreign language aptitude battery, and an intelligence test. The researcher attempted (a) to test Bachman and Palmer's (1982) hypothesis that a general SLP factor is related to the amount of information processing required for solving a given item, (b) to compare cognitive processes and strategies employed for different types of aptitude/intelligence tests, and (c) to compare the quality and quantity of test‐taking processes employed by students with different levels of SLP. The results provided important information not captured in the previous product‐oriented psychometric study (Sasaki, 1993). First, Bachman and Palmer's (1982) hypothesis only partly supported the data. The size of the general second language proficiency (SLP) factor effect on test scores did not completely correspond to the amount of information processing required by these tests. Second, the protocol analysis helped to check the validity of the tests. Most of the tests appeared to measure what they were supposed to measure. Lastly, the results revealed differences in information processing between the high and low SLP groups. The high SLP group differed from the low SLP group in assessment, planning, and execution of problem solving processes.
|出版ステータス||Published - 1993 12月|
ASJC Scopus subject areas