This experiment was designed to assess the ability of item writers to construct truly parallel tests based on a "duplicate-construction experiment" in which Cronbach argues that if the universe description and sampling are ideally refined, the two independently constructed tests will be entirely equivalent, and that within the limits of item sampling error any person would receive the same score on both tests. Two item writing committees developed forty item driver's license examinations based solely on the material in the driver's manual. The two independently developed tests were administered to 117 high school students who took both forms three to five days apart. The two forms were not equivalent according to Cronbach's criterion. As Cronbach suggests, inspection of individual items designed to measure the same general area, but worded differently, revealed some marked differences in item difficulty. His suggestion that the standard error of measurement be estimated from split-half reliabilities seemed unwarranted. The author states that perhaps on tests of very heterogeneous content domains test-retest coefficients would be more appropriate. (RC)