TEM 8 Mini-lecture and Gap-filling task. The lowest mean comes from items 22
and 23 of identifying micro discourse markers, which to students, is the least
important. This low perception can be attributed to the reason that“few hesitations,
pauses, and other features characteristic of spontaneous spoken language”(Dang
2004) are found in TEM listening audio input.
5.5.2 Exploratory Factor Analysis........................
In exploratory factor analysis, reliability statistics, KMO and Bartlett’s test, scree
plot and Rotated Component (Factor) Matrix of teachers’and students’question-
naire data are reported respectively.
5.5.2.1 Sample Size for Factor Analysis
Comrey and Lee (1992) advised sample size for doing factor analysis: 50 cases is
very poor, 100 is poor, 200 is fair, 300 is good, 500 is very good and the more the
better. But the sample size in conducting social science research would be restricted
due to quite a number of practical conditions. As regard to the current research, it is
not easy to locate a teacher sample large enough to exceed 300 or 500 and the
annual TEM 8 human rating gathering is a suitable occasion to collect teachers’
opinions because most of the teachers present at the rating section are experienced
teachers and familiar with the TEM 8 test. Even though teachers came from more
than 50 colleges and universities in this annual gathering, the number is normally
fixed around 200. In order to enlarge the teachers’questionnaire sample size, more
college teachers with the relevant experience of teaching listening comprehension
were sought and they were also helpful with the survey.
The student sample size is a bit larger, but in order to enhance the validity of the
questionnaire, all the students participating in the questionnaire survey must
complete a simulation test in the classroomfirst and this precondition has largely
restricted the student sample size, because only junior and senior students need to
prepare themselves for TEM 8 test. So, concerning the current project, a TEM 8
simulation test during the teaching time for juniors and seniors did not interfere
Table 5.4 (continued)
Descriptive statistics
N Minimum Maximum Mean Std. Deviation
Item 33 204 1 5 3.70 0.929
Item 34 204 1 5 3.72 1.039
Item 35 204 1 5 3.82 1.018
Valid N (listwise) 20450 5 Exploring Task Demands of TEM 8 Mini-Lecture...