A study of the usefulness of prior knowledge, prior experience, and spatial ability for predicting physics achievement of male and female high school students

The purpose of this study was to determine if the physics performance of females is more like the physics performance of males after accounting for performance on measures of prior knowledge, prior experience, and/or spatial ability. Male and female students taking a Regents Physics course in New York State completed a survey instrument designed to assess the extent and nature of their prior experiences in spatial activities including the types of sports in which they have participated, the toys with which they have played, and their hobbies and special interests. The spatial ability of each student was assessed with the Mental Rotations Test (mental rotations), the Paper Folding Test (spatial visualization), and the Water Glasses task (space perceptions). Prior knowledge was assessed using a science and mathematics vocabulary test and by reviewing students' transcripts to determine the number and types of mathematics and science courses they had taken and their level of success in each course.;Students completed three physics assessments consisting of multiple-choice and free-response test items, half of which had been previously determined to show gender differences in performance and half of which were determined to be gender-neutral.;The facets being studied were statistically controlled through weighted structural regression to determine the degree to which spatial ability, prior experience, and prior knowledge predict physics achievement and if these facets account for the differences in physics performance of male and female students.;Analysis of the data indicated that the best predictors of performance on physics items identified as gender-dependent were average Regents mathematics and science scores and gender. For physics test items that were not gender-dependent, average Regents mathematics and science scores were the best predictors of achievement. Spatial ability and prior experiences were not found to be predictors of physics achievement in this study. When performance on content subtests was analyzed separately for male and female students, the best predictor of physics achievement for males was prior performance on mathematics Regents examinations, while the best predictor of physics achievement for females was prior performance on science Regents examinations.