In our educational system, the problems relevant to language testing need to be reconsidered. One issue is developing and validating a Nationwide TEFL MA Entrance Examination. The current version of this test, a knowledge test consisting of 80-100 items, suffers from deficiencies related to reliability, validity, and efficiency. Being too long a test, it basically consists of memory-based and/or trivial items. The very feature which makes the test comprehensive, i.e., its length, renders it as impractical. In this research, an attempt was made to develop a test of 30 "conceptual" items and verify if it correlates with its criterion measure which is itself, according to the experts' opinion, deficient. The results of the validation stage of the newly-developed test showed that the reliability and validity indices of the test were .35 and .95, respectively indicating that the new test cannot be used interchangeably with its criterion measure due to its low reliability. However, given the implications offered at the end of the study, some steps can be taken to lessen the problems.