in fact, this study focused on the following questions: 1. is there any difference between the effect of functional/notional approach and the structural approaches to language teaching on the proficiency test of efl learners? 2. can a rather innovative language test referred to as "functional test" ge devised so so to measure the proficiency test of efl learners, and thus be as much reliable and valid a test instrument as an already standardized language test? to test the first null hypothesis, a number of procedures were taken: first. 82 male and female english major students at allameh tabatabaee university, faculty of foreign languages were selected from total population of 100 students. the subjects were randomly assigned to two groups in both experiments. the incoming students who took the fall 94 were regarded as the experimental group, and the incoming students who took the 95 winter term were considered the control group. it was noteworthy to mention that in order to hold the teacher variable constant in this study, the investigator was also the two groups teacher in language lab 1 for two consecutive terms. second, to ensure the homogeneity of the two groups in terms of their overall proficiency, the melab (michigan english language assessment battery) test was administered to both groups. computing the means and variances and taking an f-test, the two groups were equated. third, to measure the effect of instructional materials on the overall proficiency of the subjects, the same melab test along with the newly designed functional test were administered to both groups as a posttest. the statistical procedures emplyed was the t-test formula. the results revealed that the subjects who were taught on the basis of the functional syllabus were better language performers, thus rejecting the first null hypothesis. as for the second null hypothesis which was to determine whether the newly developed functional test was as much reliable and valid as was melab test, the following steps were taken: first, to determine the reliability coefficients of the subtests along with the functional test, an internal consistency measure, namely kr-21 was used. then, these coefficients were adjusted by means of the spearman-brown priphecy formula. as shown in chapter 4, the adjusted results are reliable. next, to determine the concurrent validity of the functional test, intercorrelation coefficients were used. the adopted formula was the pearson product moment correlation. (see chapter4) considering the statistical properties of the functional test, it can be safely concluded that the functional test was as much reliable and valid as the melab subtests. as such, the second null hypothesis was also rejected.