The effect on childrens' scores of different item formats used in standardized mathematics achievement tests was investigated. Second grade students were given a mathematics computation test using formats derived from five standardized achievement tests. Identical content was tested with each format . Differences in test scores between types r0f formats were statistically significant at p(.001 (F = 45.25). These results indicate that what a student appears to know is substantially influenced by the format of the particualar test used in measuring achievement. These differences are not accounted for by the rno~mative scaling of the different tests. Greater attention should be ~i,en to the effect of test item format in selecting and administering cac1 i evemen t tests.