This paper discusses an investigation that compared college students' reading gains when using general or specific diagnosis. It reports that significant gains were made on many subskills of the Stanford Diagnostic Reading Test by the group that had the prescriptions developed as a result of a differential (specific) diagnosis and that there were no significant gains noted on the Nelson-Denny Reading Test, a general survey test. An analysis of two other research studies on this topic shows that results of this study support the results of a study by G. L. Cox and J. E. Swalm and contradict the results of a study by B. D. Smith. It is noted, however, that since Smith used the Nelson-Denny Reading Test to test gains from specific diagnosis, her findings are not necessarily valid and not necessarily contradictory to those of the other two studies. The evidence in the literature indicating that the Nelson-Denny Reading Test is not an appropriate test for evaluating the reading ability of college students in developmental programs and that survey tests are often used inappropriately as measurements of achievement in college reading programs is discussed in the light of the findings from these research reports. (MKM)