2. Literature Review

This chapter will begin with an introduction of text type and task type that will be investigated in the current study and some other related theoretical items, followed by a review of previous empirical studies in the same or relative fields, and will end with limitations in previous studies.

2.1 Theoretical issues

2.1.1 Listening comprehension

Researchers (e.g., River, 1981) estimated that people listen twice as much as they speak, four times as much as they read, and five times as much as they write. However, listening should be distinguished from listening comprehension. The difference between listening and listening comprehension mainly lies in the motivation and the process. ”Listening#8217; means just listening to the message, it#8217;s unnecessary for listeners to interpret or make any response to the text during the listening process, thus listening can be aimlessly, while ”listening comprehension#8217; is much different, the process of listening comprehension includes meaningful interactivity, in this process listeners obtain useful information from the auditory cues and activate the existing knowledge stored in their memory to better understand and comprehend what they heard(O#8217;Malley Chamot, amp; Kupper, 1990).

A representative definition of listening comprehension was propounded by Clark and Clark(1977). They defined listening comprehension from two perspectives #8211; broad and narrow: In its narrow sense listening comprehension denotes the mental process during which listeners take in the utterance by the speaker and use them to construct an interpretation of what they think the speaker intended to convey. Listening comprehension in its broad sense, however, is not limited thereto, for listeners normally put the interpretations they have built to work.

2.1.2 Task type and listening

Among the many existing variables that are considered to have influence on language test performance, one main variable is task type (e.g., Alderson, 2000; Bachman amp; Palmer, 1996; Brantmeier, 2005; Buck, 2001). There are lots of terms similar to ”task type#8217;, such as ”test format#8217;, ”question type#8217;, ”response type#8217;, ”item format#8217; and so on.

A mass of task types have been employed in language testing, such as gap-filling, matching, multiple-choice, open-ended (or short-answer), ordering, recall, summary and true or false tasks (e.g., Alderson, 2000; Buck, 2001;Kobayashi, 2002). Since there is not a type of task that can function perfectly in every situation, it#8217;s necessary for testers to know well about the characteristics of each type of task and make the best choice according to which ones most appropriately serve the aim of a test in each context. In terms of task types, this study focuses on investigating the following two ones: multiple-choice and short-answer tasks.

Hereinafter is the brief introduction to both merits and weaknesses of multiple-choice and short-answer tasks in a listening test.

Merits and weaknesses of multiple-choice task

Multiple-choice task is a widely used task type in language tests all over the world. It has been the backbone of standardized testing programs in the U. S. (Bennet, 1993). Hughes (1989) proposed that the benefits of multiple-choice task are widely recognized for its objectivity, economical efficiency, being able to incorporate more test items, and higher reliability. But he also stated that the weaknesses of multiple-choice task can be listed as follows: firstly, multiple-choice task merely tests the test takers#8217; recognition knowledge. The performance of multiple-choice task may not be adequate to examine or verify test takers#8217; productive ability. Multiple-choice task can be regarded as a gap to be bridged between knowledge and use. Since communicative tests aim to assess use, that gap means that test scores can at most give incomplete information; secondly, guessing has an unknowable and considerable effect on test scores. When scoring papers, the markers can never know what part of the score has come about simply through guessing; thirdly, it may facilitate cheating behaviors. The options (A, B, C, D) for a multiple-choice task are so simple that it is easy to plagiarize or communicate to other test takers easily.