Exploring Teachers' Test-constructing Processes and Students' Test-taking Processes

• Main Authors: Fan-ping Tseng, 曾繁萍
• Other Authors: Yuh-show Cheng
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/15803603293681557056

博士 === 國立臺灣師範大學 === 英語學系 === 102 === This study aims to investigate three research questions. First, how did experienced and novice teachers construct mock tests for the Scholastic Ability English Test (SAET)? Second, how did higher- and lower-proficiency students take those mock tests? Third, were students’ considerations for answering the tests consistent with teachers’ test-constructing considerations? Four senior high school teachers and forty-eight senior high school students participated in this study. All participants were asked to do think-aloud while performing their tasks. The teachers were asked to construct twenty-eight items of multiple-choice questions on vocabulary, cloze, and reading comprehension. The students were asked to answer the questions constructed by the teachers. Major findings of this study are summarized as follows. First, the experienced teachers and novice teachers seemed to make different types of considerations in constructing their tests. The experienced teachers took more student-oriented factors into account while the novice teachers took more test-construction principles into consideration. Despite their different considerations in test-constructing processes, the two experienced teachers did not seem to produce better test items than the two novice teachers. All four teachers had constructed some items that were deemed poor, problematic, or inappropriate from the authority’s perspective. Second, students generally used different strategies when answering different types of questions. However, they seemed to use the strategy of “elimination” very frequently on three types of tests. In terms of the proficiency levels, higher-proficiency students tended to use their vocabulary knowledge, grammar knowledge, and deductive reasoning more frequently than lower-proficiency students in answering the items. On the other hand, lower-proficiency students tended to use the strategy of “guessing” more frequently than higher-proficiency students across three types of questions. Third, students’ considerations for answering test items clashed with teachers’ test-constructing considerations to a great extent; the overall consistency rate between them was only about 33% in this study. Furthermore, students generally thought in a way more congruent with novice teachers than with experienced teachers. In addition, higher-proficiency students’ considerations clashed more with teachers’ considerations on cloze items while lower-proficiency students’ considerations clashed more with teachers’ considerations on reading comprehension questions.