The participants were selected randomly from among undergraduate medical and
paramedical students (n=137) who had enrolled in ESP I in Isfahan University of
Medical Sciences. A standardized language proficiency test (Intermediate TOEFL
Test Practices by Keith S. Folse 1994) was given to them in each field (i.e. medicine,
pharmacy, dentistry, midwifery, nursing, physiotherapy, health services management
and medical records) and those whose scores fell beyond one standard deviation
above or below the mean were excluded since they as the most successful and the
least successful learners respectively may have certain unique characteristics which
may lead to biased findings and interpretations and divert the orientation of the study.
Then, 103 learners with mean ± 1SD were identified. 14% of the participants were
male and 86% were female. The participants’ mean of age was 20 ± 1.12.
This test was selected because of its standard format, the current level of language
proficiency of the participants derived from a survey of their prior educational
experience, ease of administration and scoring, and its availability. It was also
selected after consulting with several experts in language testing. In addition, the test
was critically read by some experts in applied linguistics to check for its validity.
Although it was a standard test, it was piloted in conditions similar to our main study
to ensure its reliability for the context of our study, its test- retest reliability was
calculated. It turned out to be 0.75.
Three areas of language proficiency were tested using the multiple-choice format of
the selected test: grammar (20 items), vocabulary (20 items), and reading
comprehension (10 items). This test was performed as part of classroom evaluation
activities with the help of the instructors.
Since it was a qualitative study and random sampling was not practically possible,
from among the existing ESP I classes for each of the above-mentioned fields, one of
them was selected randomly to form the participants of the study in a kind of stratified
sampling way. Since there was a small number of participants in each group,
normality test was performed using Eviews software. Jarque_Bera statistics for all
fields was less than 5.99 with confidence interval 95% and degrees of freedom equal
to two which ensured the normality of the population in each field.
The participants had all studied the same textbooks in compulsory English courses
at junior and senior high schools mainly based on grammar-translation method. They
had all participated in a nation-wide university entrance exam which included an
English test on high school English