As part of the Content Analysis Project, a panel of six experts in mathematics education examined the congruence between the Maryland grade-8 test and the National Assessment of Educational Progress (NAEP) test. Two panelists were very familiar with the Maryland School Performance Assessment Program (MSPAP), two with the NAEP, and the other two were familiar with neither test. Maryland was one of two states chosen for a similar analysis; the other, North Carolina uses an assessment composed entirely of multiple-choice questions, while the MSPAP is entirely constructed-response questions. The panel concluded that the differences found between the MSPAP and the NAEP at grade 8 along the content dimension are not sufficient to account for the magnitude of difference between proficient performance on the Maryland test and proficient performance on the NAEP. Neither are the slight differences between the Maryland test and the NAEP test at grade 8 along the cognitive dimension sufficient to account for the magnitude of differences between proficient performance on the Maryland test and proficient performance on the NAEP. The panelists offered three possible reasons for the discrepancy: (1) differences in stakes or consequences between the two tests; (2) the connections between the test and instruction; and (3) the standards-setting processes used. (Contains 12 references.) (SLD)