The purposes of the present study were to ascertain whether or not two particular formatting characteristics influenced seventh grade student performance on math tests. One research question focused on the effects of decimal positioning in multiple choice options for computational items. In a second question, labeling format was compared with identification format for geometry items. The influences of prior achievement and test composition were also studied. Results indicate that prior achievement had a strong, consistent influence on student performance for both computational and geometry items. Test composition (single versus mixed format) had little influence. Results for formatting are not clearly prescriptive, although additional considerations for geometry formatting are identified. (Author)