Test Method Effect On Reading Comprehension Test

The purpose of the study is to investigate the effect of test methods on test-takers' reading performance and whether test formats exert similar influence on reading performance of test-takers at high-level and low-level language proficiency in reading comprehension tests.Besides,the test-taking strategies used by participants during reading comprehension tests are also analyzed as to provide possible reasons for the effect and explore difference in multiple-choice format(MC format)and short-answer format(SA format).Two reading comprehension tests with same passages and equivalent-stem items and think-aloud method were instruments in this study.Totally,68 juniors majoring in English took part in the major study.They were divided into high proficiency-level and low proficiency-level group based on their average scores in TEM-4 and in final exam of comprehensive English class last semester.60 students from two parallel classes completed reading comprehension test in MC format and in SA format respectively.Another eight students were divided into two parallel groups.Each group was composed of two high-level students and two low-level students.They were required to answer multiple-choice questions(MCQ)and short answer questions(SAQ)respectively while reporting their ongoing thoughts using think-aloud method.An immediate retrospective interview was conducted after students completed reading each reading passage.After the experiments,both quantitative and qualitative analysis were conducted.Test results of reading comprehension tests were analyzed using Independent Sample T test and two-way ANOVA.Besides,think-aloud protocols were transcribed and coded.The quantitative analysis showed that test format had significant effects on reading performance of test-takers.They performed better when answering MCQs than SAQs.High-level and low-level test-takers were both greatly affected by test formats while low-level test-takers were more likely to be affected by test formats.In terms of qualitative analysis,it was found that test-takers employed a higher proportion of bottom-up strategies than top-down strategies in both MC and SA test.In addition,it was shown that test-takers used more top-down strategies when they responded to SAQs than MCQs.Test-takers with high language proficiency employed more types of strategies and use reading strategies as well as top-down strategies more frequently than those with low language proficiency in the SA test.The difference in test-taking strategy use is one of the possible reasons for the difference in test performance and reflects the difference between the two test formats.