It is noted that while language proficiency tests are not an ideal method for predicting student performance, they are commonly used as admission criteria. A methodology for comparing college tests of English as a Second Language (ESL) that are used for program admission is offered. The procedure imposes certain constraints on data collection, including minimization of the time lag between administration of the tests being considered, replication of testing conditions, and assuring approximately equal familiarity with the different tests. When scores on one test are not available for comparison, score distribution charts can be used as a substitute. Statistical techniques for linking tests can be techniques of either projection or moderation. Suggestions are made for estimating decision consistency, decision agreement, and construct relevant variance. The methodology is based on a comparison of results on the Test of English as a Foreign Language and the Canadian Test of English for Scholars and Trainees, both administered to four groups of examinees (n=250). (Author/MSE)