The effect of a proportional reasoning-based test preparation instructional treatment on the mathematics achievement of eighth grade students

This quasi-experimental, pretest-posttest, control-group design study examined the effect of a proportional reasoning-based test preparation instructional treatment on the mathematics achievement of eighth grade students enrolled in regular eighth grade mathematics at a Title I junior school in southeast Texas. Mathematics achievement was measured using the Texas Assessment of Knowledge and Skills. The treatment for the experimental group (n = 40) was ten ninety-minute lessons consisting of forty-five minutes of test preparation and forty-five minutes of proportional reasoning lessons while the comparison group (n = 47) had ten ninety-minute lessons of test preparation only. The differences between the pretest and posttest aggregate raw scores, raw sub scores for the proportional reasoning items, and raw sub scores for the non-proportional reasoning items for the two groups were calculated using ANCOVAs.;Further research is needed to empirically investigate the kinds of test preparation practices that can most effectively improve the achievement of eighth grade students. More research is also needed to further investigate the finding that home language was a predictor of mathematics achievement on proportional reasoning items, but was not a predictor of mathematics achievement on non-proportional reasoning items.;No statistically significant differences were found between achievement scores of the treatment and comparison groups. The results of a linear regression indicated that of previous mathematics achievement, gender, socioeconomic status, and home language, previous mathematics achievement (total score) statistically significantly contributed to the posttest total score, previous mathematic achievement in proportional reasoning and home language statistically significantly contributed to the proportional reasoning posttest scores, and previous mathematics achievement in non-proportional reasoning contributed statistically significantly to non-proportional reasoning posttest scores.