A Comparison Between Multiple Choices and Short Answer Questions in Reading Comprehension Tests

A Comparison between Multiple Choices and Short Answer Questions in Reading Comprehension Tests

1. Introduction

Reading is an important way for people to obtain information and communicate with each other. Reading is one of the most important sources of input in the language acquisition process and in EFL (English as a Foreign Language) learning. Reading comprehension has long been the most important part of English test and usually occupies a large proportion in various tests. Reading assessments can be found in most large-scale tests both at home and abroad because of the significant role reading plays in the process of foreign language learning. Specifically speaking, the reading comprehension part always takes up a large proportion of the whole test. Thus, to some extent, scores that the test-takers receive in the reading comprehension part of an examination may play a decisive role on their performance in the whole test.

Reading comprehension is a complicated task that receives much attention both in teaching and testing structure. In College Entrance Examination, reading comprehension accounts for about 40 marks out of the total marks of 150. With the complicated process of reading and its central role in English learning, whether various reading test formats can check out the students’ real reading ability is rather questionable. A study conducted by Andre (2006) indicated that different test formats result in various response processes in test takers and then lead to different performances. To put it simply, test takers were affected by the different item types in the study.

In order to find out which test format is better for testing students’ reading ability, it is important to have some knowledge of what is reading comprehension. Reading is a receptive language process. It is a psycholinguistic process in that it starts with a linguistic surface representative encoded by a writer and ends with meaning which the reader constructs.There is thus an essential interaction between language and thought in reading. The writer encodes thought as language and the reader decodes language to thought (Carrel, etc, 1988:12). Reading comprehension, accordingly, is complex interactions between text, setting, reader, reader background, reading strategies, first and second language, and reader decision-making (Erler and Finkbeiner, 2007).

As Heaton (2000) puts it, “certain of the item types will be more suited to testing reading comprehension of a particular test than other item types”. Indeed, there are numerous ways of testing reading comprehension, ranging from selected-response questions to open-ended questions. The commonly used selective answer test questions including multiple-choice questions, the true or false questions, matching questions and other kinds of questions. Common open-ended questions include short-answer questions, translation and composition. Among these formats, the multiple-choice (MCQ) is the most frequently used one. Accordingly, we adopt this format in our test. To compare with this type of test format, another test format “Short answer question (SAQ)” is used.So in this study; we mainly focus on the comparison between multiple choices and the short answer questions.

Therefore, to make a comparison between the two test formats, the author designed this study. To speak specifically, the purposes of this study are as follows: 1) to investigate the influences of different item types exerting on performance of test takers; 2) and to find out the relationship between item types and the test-takers’ performances; 3) to study the validity of different item types in reading comprehension, that is, which item type has the more validity.

2. Design of the Research

2.1 Research Questions and Hypotheses

    The aim of the present study is to explore the influence of different task fomats on reading comprehension performance and the relations between task formats and reading abilities among middle school students. To fulfill the purposes of this thesis, and based on the above questions designed for the research, four hypotheses were made as follows:

Hypothesis 1: Different item formats have significant influences on test-takers’ reading test performance.

Hypothesis 2: Students of different English levels perform equally well on different item types with all other conditions unchanged.

Hypothesis 3: The scores students get in different items have a positive relation to their language proficiency.

Hypothesis 4: Multiple choices have more advantages than short answer questions in reading test.

2.2 Participants

In this study, 60 Grade Nine students from No.2 Middle School in Nanchang are taken as subjects. They are from the same class, so they are taught by the same teacher. The assessment will be made by the English teacher and the author of this paper together.

For the sake of convenience, the students are grouped into two teams according to their English proficiency: the higher and lower group. The standard of grouping is their scores of mid-term examination.The details of grouping are as follows:

Group1, the higher group, consists 30 students (18 females, 12 males), with the mean scores of mid-term exam above 95(the total points is 120); Group2, the lower group, consits of 30 students (11 female, 19 males), with the mean score of 77.2. The purpose of grouping according to their scores is to test whether the different levels of test takers may be affected by the different test formats in the hypothesis 2.

2.3 Instruments

The present study employed a formal experimental reading test as an instrument, which was conducted to probe into the relationship between task types of reading and test takers’ performance. Altogether four passages were utilized in the present study in order to measure test-takers’ reading comprehension. The students were asked to finish the reading exercises in 30 minutes. And it is very important to ensure the content validity of the test. So all the four passages of the present study were carefully selected from the authentic English papers of entrance examinations for high school of the past few years.

Table 1 Structure of Test Papers

2.4 Research Procedures

Before conducting research, I choose one class because the students in this class are top students in No.2 Middle School. The English level of the whole class is high. Besides, my friend is the English teacher of this school, so it is convenient for him to choose these students as my participants.

Before the test, the students were informed that the reading test was for research purpose and they have known the importance of their contributions to the results of the study. And for the sake of research, the test should be taken seriously and honestly. Generally speaking, most students took the tests seriously. They themselves showed enthusiasm toward the tests.Therefore, the scores those students got from their test papers were considered valid.