Method: Participants were 881 students in the 7th, 8th, and 9th grades enrolled in PE at two schools, from one school district, in the Southwestern U.S. One school taught PE in a coeducational format, while the other school taught PE a same-sex format. At the end of the spring semester all student's completed a three section survey (Treanor, Graber, Housner, & Wiegand, 1998). The demographic section included questions on gender, grade, age, sports participation, and a section for teachers to complete which categorized students motor skill abilities as above, at, or below grade level. The second section consisted of eight questions that students rated on a Likert-type scale from 1 {strongly disagree) to 4 (strongly agree) regarding perceptions of their level of skill, fitness, effort, and enjoyment in PE class. The third section of the questionnaire consisted of a series of 12 preference statements regarding coeducational and same-sex PE.
Analysis/Results: A 3 x 2 x 2 (Grade x Gender x School) ANCOVA with covariates of sports participation and skill level was performed for each of the eight questions regarding student self perceptions of skill, fitness, effort, and enjoyment in PE. Correlations were performed to determine if grade, gender, school, sports participation, skill level, or ratings on the eight student self-perception items were significantly related to student responses on the 12 questions regarding preferences for coeducational or same-sex PE. Due to the use of multiple statistical procedures significance level was established at p < .01. ANCOVA's revealed several significant main effects for each question by gender, grade, school, and grade x school. Pearson's product correlation analyses revealed several significant relationships between gender, grade, school, sports participation, skill level, and four of the eight self-perceptions items.
Conclusions: Males, athletes, and higher skilled students expressed more positive self-perceptions. Positive self-perceptions also decreased as grade level increased. Students in same-sex classes enjoyed and tried harder in PE than students in coeducational classes. However, correlations found that student preferences towards same-sex and coeducational PE were associated with what students were exposed to. Students in coeducational PE preferred coeducational classes, while students in same-sex PE preferred same-sex classes. Grade level data also revealed an affinity to prefer coeducational PE as grade level increased. This could be related to a desire for varied social interactions.