The structural validity of the Ohio Ninth Grade Proficiency Test was investigated with the test scores of 713 eighth graders who had their first opportunity to take the test in March 1999 in a large urban school district in Ohio. The study examined a hypothetical structure of the test to determine its fit to the data. Confirmatory factor analysis was performed, and several fit indices including X[superscript 2] statistics and the root-mean-square error of approximation were examined. The theoretical structure of the test was first found to be valid, and then the invariance of the structure was examined across males and females. Both the tests of structural fidelity and invariance were cross-validated using a sample of 735 eighth graders who took the parallel test in March 2002. The results provided evidence of the invariance of factor loadings and factor variances in both years. Factor covariances were found to be invariant in 1999, but this was not confirmed by 2002. Results consistently showed the lack of invariance of error variances. (Contains 1 figure and 7 tables.)