II.2 - DATA COLLECTION TECHNIQUES
AND INSTRUMENTS
As research instrument, we used written tests. The
identification cards were designed to discover the students. In the cards we
wanted to identify the gender of the testee and check also if he/she is a
newcomer. The second aspect is important for us because we aimed at assessing
the level of students just coming from secondary schools. We were interested in
the gender because we wanted to know the rate of participation of girls in the
test.
To go into the study proper, we followed Katz's (1977:19)
proposition quoted in Levinson (1983) stating that a pragmatic theory is part
of performance because pragmatic theories explicate the reasoning of speakers
and hearers in working out the correlation in a context with a token
proposition. We used the written test as the performance item.
The written test was preferable to the oral test because the
data collection was easier and quicker. In fact, a prior informal observation
of students' spoken data had allowed us to notice that they are almost bare of
modals. In addition, students' understanding of the modals used in some
interviews was more difficult to check. We thought the solution was in written
test. We made a pretest to see if the test would be adapted to the level of
students. The pretestees made some suggestions and we made the final test.
We utilized two modal understanding /production tests of
forty sentences, each test made of three series of exercises. The first test
embodying twenty sentences was designed to assess the students' understanding
of modals. We proposed the context wherein we used the modal and we asked the
students to find the pragmatic meaning in a multiple choice format.
The first series, made of seven sentences, was about the
deontic meaning of modals; the second series, composed of three sentences, was
about the epistemic meaning of modals; and the third series, containing ten
sentences, was a combination of deontic or/ and epistemic meaning. The
contexts were made clear enough to avoid confusion of meanings, except in the
third series where we intentionally left two sentences without clear contexts.
By doing so, we wanted to assess the students' modality tendency. (cf. Appendix
II)
In the second test we used twenty sentences. We utilized a gap
filling test where we asked the students to fill in the blank with the
appropriate modal verb. In this test also divided into three series, the first
series had seven sentences and was about the deontic meaning of modals; the
second series, with three sentences, dealt with the epistemic meaning of
modals; and the third series had ten sentences and was a combination of deontic
and epistemic meanings.
In each series, we proposed a list of modals and gave a
context where a modal should be used. We ask the students to give the right
modal referring to the pragmatic meaning of the modal indicated in front. The
context was so that only one right modal was appropriate. To assess the
students' modality tendency we left two sentences without contexts but with two
meanings and we arranged that only one modal expressed both the deontic and the
epistemic meanings.
In both tests, we used miscellaneous meanings and modals to
display the differences between the deontic and the epistemic meanings, and the
inner differences between modals within each type of meaning. (cf. Appendix
II)
All the sentences were formulated taking into account the
distinctive features of the meanings of modals.
We asked the students to fill in the identification cards
before dealing with the test. The session lasted more than two hours and the
following section describes our sampling.
| 
