Is it possible to translate an existing test from the source language to a target language and then to empirically link the translated test to the source language standard? Some tests are amenable to translation of their content and some are not. Within tests, some items may be translatable and others may not. Some items may measure meaningful content when translated, and others may not. Within items, some options may be translatable, and others may not. This paper focuses on the rigorous method used to translate and adapt from English to Spanish three equated forms of the Tests of General Educational Development (GED tests) to ensure that all items are valid and that the translated instrument measures a comparable construct. Most specifically, the paper addresses issues related to translating items on the language subtest. The Spanish-language GED tests were normed using only graduated high school seniors in Puerto Rico. In addition, the paper outlines a linking design that introduces a procedure for screening biliterate students for linguistic equivalency across languages before including them in the linking sample. This design will be tested in May 1997 when the test is administered to a large sample of biliterate Hispanic high school seniors. (Contains 2 tables and 12 references.) (Author/SLD)