| Reading comprehension has long been the most important part of English test and usuallyoccupies a large proportion in various tests. Meanwhile, being one of the basic skills of Englishlanguage, reading comprehension is also the most difficult part for the test takers. Thus it is safeto say that the scores test takers get in this part directly determine the sores of the whole paper.Since language testing begins, reading comprehension has been the favorite test form to themost of testers. It is long been the predominant form in the testing system. Both the frequency itappears and the rate it accounts in a single test paper are far higher than other test forms such ascloze or translation. It is true that reading comprehension enjoys the unique advantages in testingthe test takers’ reading competence. However, in the light of the results of the development oftheories in the fields of linguistics and language testing, more and more scholars have expressedtheir doubts and advocated their different voices on reading comprehension which takes MultipleChoice Question, the traditional testing method as its only method for so long period. Scholarsraised their suggestions based on their studies and researches on reading comprehension fromdifferent aspects. Some contributed to the validity of different item types such as MCQ, TR andSAQ; some hold diverse views on whether different item types have significant influences ontest-takers’ reading test performance; some believe that different item types have differentadvantages in different aspects and the test takers should make a choice when they are testing thetest takers language competences in different level, some hold students of different Englishproficiency levels will be affected by different item types. However, these studies are mostlyconcerned about English majors and leave the non English majors untouched or only take ashallow research on this realm. So far still insufficient evidences are not gathered for a firmconclusion, that’s why this study is necessary and important.In this study84first-year non-English major students from two classes of the same Englishteacher are taken as the research subjects. The subjects of this study will be put into higher,medium and lower groups according to the scores they got in the National Entrance Exam andthe scores in the pilot study. In this experiment, each group will have two tests of the readingcomprehension with the passage unchanged while the item types altered—the first test takesMultiple Choice Question and T/F as the item types, the second test uses Translation and ShortAnswer Question as testing item types. Finally a questionnaire was handed out to the test-takers.The statistics will be analyzed by spss13.0in this study. This experiment mainly deals with thefollowing questions:1) Do different item types have significant influences on test-takers’ reading testperformance?2) Do Students of different English proficiency levels perform equally well on different item types? If yes, which proficiency level performs the best?3) Do the scores students get in different items have a positive relation to their languageproficiency?4) Which item type enjoys the more validity, the subjective items or the objective ones?The results of the study show that:1) Different item types have significant influences ontest-takers’ reading test performance. Test takers score highest in T/F questions and lowest inShort Answer Questions;2) The lower group are greatly influenced by the item types, and on theother hand the high group are least influenced;3) Scores students got have a positive relationwith their true English level regardless of the item type;4) The questionnaire shows that moststudents agree the point that though they score easily in objective items, actually Short AnswerQuestion can truly test their language competence，which means short answer question enjoysthe more validity.At the end of the study, the thesis also offers some advisable suggestions to foreign languagetesting and teaching, as well as to the future study. |
PDF Full Text Request