Scheduled for Poster Session: Teacher Education and Curricula in Physical Education—Measurement and Research Methods, Wednesday, March 17, 2010, 1:45 PM - 3:15 PM, Convention Center: Exhibit Hall RC Poster Area


Preliminary Results of a Multigrade Physical Education Attitudes Scale

Richard E. Cain, Southern Connecticut State University, New Haven, CT and Debra Ballinger, Towson University, Towson, MD

Background/Purpose

This study presents the preliminary analysis regarding the development of an attitudes toward physical education scale that teachers can use to measure attitudes across different grade levels and frequency of physical education programs.

Method

The survey was adapted from a youth sport satisfaction scale. Physical education was substituted for sport as the environment, and items related to health-related fitness and perceived perceptions of physical education desirability were added. The initial set of questions were reviewed for face validity by teachers (N=20) who had been trained in FITNESSGRAM and Physical Best, and again following a focus group discussion of teachers (N=10). Following procedures approved by the university institutional review board, teachers of students enrolled in grades 4-12 physical education classes administered the revised survey to students (N=1353) in their regular physical education classes to determine their students' attitudes toward physical education.

Analysis/Results

An Exploratory Factor Analysis using principal axis factor analysis was conducted using SPSS (Version 16). These factors were rotated using Promax rotation with Kaiser normalization and yielded two interpretable factors: 1) physical relevance and 2) social relevance. The eigenvalues for these two factors were 6.09 and 1.15, accounting for 46.8% and 8.9% of the variance, respectively. Cronbach's alphas for the entire survey, Factor 1 and Factor 2 were .87, .87 and .92 respectively. A multiple linear regression was also performed to determine if differences were discernable by students' grade level, and the number of days the students had P.E. As grade level and the number of days per week the students had physical education increased, students' attitudes toward Factor 1 "physical relevance" decreased. However, the regression model indicated no relationship for the social relevance factor. Furthermore, the two factors were found to be significantly correlated (r =.77; p<.001), suggesting that the factors were not totally independent.

Conclusions

The strength of the study is that the survey was conducted on a large sample comprised of elementary, middle, and high school students, as well as the fact that the instrument can easily be distributed and completed in a short amount of time to reliably assess student attitudes toward physical education class. A weakness is that the results are based upon student self-report and may be subject to bias, as well as the relatively small percentage of variance from Factor 2. Future studies utilizing confirmatory factor analyses to replicate the findings with large sample sizes in diverse cultures are warranted.


Keyword(s): measurement/evaluation, physical education PK-12, research

Back to the 2010 AAHPERD National Convention and Exposition (March 16-20, 2010)