The 'Task Type' Effect On English Reading Test Performance For The Senior Middle School Students

Over the past three decades,considerable research has been conducted in the field of language testing,indicating that performance on language tests is affected by a wide variety of factors,and any attempt to overweight or ignore some factors would have a harmful backwash,either to language testing itself or,in turn,to language learning and teaching.Thus, a clear understanding of what and how these factors may affect test performance is fundamental to the development and use of language testing(Bachman,1990),which would facilitate language teaching and learning as well.。Based on Bachman's(1990) general model with regard to the test performance explanation,the present study,by means of a quantitative design,aims to empirically investigate the 'task type' effect on predicting and explaining the senior middle school students' reading performance,hoping that the results of this study can help better understand the general model for explaining the test performance and in turn,may have some beneficial backwash to language teaching and learning.The study focuses on the following research questions:1.How do the participants perform on the two task types?2.To what extent does the 'task type' factor affect the participants' reading performance?3.How well do the two task types correlate with each other?4.Among the three major sources based on Bachman's(1990) model in terms of explaining the language test performance,which factor may affect the reading comprehension test performance in participants' eyes?The sample papers of PETS-2 reading section was administered to 90 senior middle school students,containing two task types:cloze and multiple-choice.The performance data from the two task types were analyzed using the SPSS 16.0 statistical package for Windows. Data analysis and the consequent results in this study indicate that the 'task type' factor does exist and have a main effect on the reading comprehension test performance while not greater than the 'proficiency' or 'language ability' effect,which supports Bachman's(1990) general model for explaining performance on language tests.What is more,the results reveal that there seems to be a threshold level of language proficiency,beyond or below which the 'task type' factor does not seem to have a significant effect on test-takers.The data collected from the questionnaire also reinforces the above statistical results, indicating that a particular participant's test performance is a function of task characteristics, his or her personal characteristics,and in most cases,interactions between the two; furthermore,only taking the notion—'language ability threshold' into account seriously could both task and personal characteristics predict participant's performance reasonably.The thesis is composed of five chapters.Chapter One briefly declares the research background,the purpose and the overall structure of this thesis.Regarding the controversial issue concerning the 'task type' effect on the language test-takers' performance,this study aims to empirically evaluate the extent to which the 'task type' may affect the reading comprehension test performance for the senior middle school students.Based on Bachman's(1990) model with regard to explaining the test performance, Chapter Two reviews the relevant existing literature from theoretical and empirical perspectives.In addition,the two task types employed in this study are presented in details.Chapter Three describes the in-depth design of the proposed study,including research questions,participants,instruments,variables under investigation,data-collection and data-analysis.Chapter Four reports the results to the four research questions,as well as the discussion, explaining the possible reasons for the findings of this study.Chapter Five makes a summary of this thesis,including major findings,limitations, implications of the findings and recommendations for future research.