4.3 The Test Format of TEM 8 Mini-Lecture Comprehension.......
Test measures must vary in order to construct authentic target-language use con-
texts. In order to be integrative and communicative in nature, the mini-lecture part
followed by a gap-filling task is well acknowledged to be highly interactive (Zou
2004). More specifically, TEM 8 Mini-lecture and Gap-filling task is a structured
gap-filling task on the summary of a mini-lecture, for the response format is an
outline with bullet points that summarizes main points as well as their supporting
details in a hierarchical order (see Appendices 4 and 5). The testing procedure goes
like this: students listen to a mini-lecture within 10 min (900–1000 words) for once
only, and while listening, they can take notes on the important points. The notes
will not be marked, but they are needed to complete a gap-filling task after the
mini-lecture. Students will be given 2 min to check their notes after the mini-lecture
and then, another 10 min to complete the gap-filling task on an answer sheet
delivered to them after the mini-lecture. They are required to write no more than
three words for each gap with reference to their notes. The response format of the
gap-filling task is a listen-to-summarize cloze task in an outlined form.
The TEM 8 gap-filling task on a structured summary is different from other
summary forms, such as short-answer questions or writing summary passages.
However, summarization forms vary in different language tests, because“the dif-
ferent conceptualizations and the multiple purposes and types of summaries indicate
that the concepts of summary are far more complex than a single, uniform entity”
(Yu 2013: 104). According to Buck (2001), the processes involved in gap-filling
tests could vary and hence the construct to be measured. The two variables
determining the processes and the relevant construct are first, “the type of
gap-filling task”and second,“the nature of the information that has to be inserted
into the gap”(Buck 2001: 73). If only the word recognition skill is to be tested, the
construct does not target a higher level of listening comprehension, discussed in
Chap. 3. Buck (2001: 73) further argued that the gap-filling task on summaries
could test:
general linguistic knowledge, discourse knowledge or even inferencing, depending on the
nature of the information to be inserted. They might also need some metacognitive pro-
cessing in order to determine what information the test-developer intends for the blank.
(p. 73)
In another word, each blank intended in the gap-filling task should be aimed at a
certain part of the relevant construct and hence needs prudent consideration for its
appropriateness. A number of studies ensured high reliability of the gap-filling task
(Lewkowics 1991; Buck 1992), but there are also pitfalls hard to avoid: gaps can be
filled without listening comprehension and certainty of the information to befilled
in the gaps might not be guaranteed. Besides a pre-test, Buck (2001: 72) also
proposed a group of people’s pre-doing of the task without listening to the acoustic
input to make sure that the gaps do need candidates’listening comprehension.
36 4 Task Characteristics of TEM 8 Mini-Lecture...