Scheduled for Poster Session: Research Across the Disciplines I, Thursday, April 2, 2009, 12:45 PM - 2:15 PM, Tampa Convention Center: Exhibit Hall RC Poster Sessions


Validity of Physical Education Steps/Minute Guidelines

Philip W. Scruggs, University of Idaho, Moscow, ID

Purpose: The validity of pedometer steps/min guidelines for physical education (PE) physical activity (PA) recommendations was investigated. Objectives were: (a) examine the influence of lesson duration (min), stature (cm) and BMI (kg/m2) on the relationship between steps/min and percent of a PE lesson physically active (%PA); (b) assess the validity of a common steps/min guideline for 33.33 and 50 %PA; and (c) examine the common steps/min guidelines across genders and grade level groups (1st–6th and 7th–12th). Methods: Data sets from previous research, where PE PA was quantified via pedometry, were combined. A total of 1192 first- through twelfth-grade participants with concurrent steps/min and observed %PA scores in PE were included in the analysis. Analysis/Results: Data were analyzed using correlation, regression, and diagnostic test statistics. Alpha was set at .01. Overall, PA outcome measures were strongly correlated and significant (r = .914; r12•345 = .906). Correlations between steps/min and %PA were .910 and .913, for females and males, respectively; and .833 and .958, for grade level groups, respectively. Linear regression equations were created where steps/min and %PA were entered as both the dependent and independent variables (i.e., laboratory and field equations). Steps/min, %PA, and stature were consistently significant predictors. Steps/min and %PA accounted for 83.5% of the shared variance; however, stature only accounted for .6 to 1.4% of the shared variance. Steps/min was the only covariate in the logistic regression model that was a consistent and significant predictor of the %PA guidelines. Overall, and by gender and grade groups, ROC statistics indicated that steps/min was an excellent discriminator (AUC ≥ .90) of those who met and did not meet the %PA guidelines. Steps/min values of 60.84 (c = 88.6%, sensitivity = 88.51%, specificity = 88.74%) and 79.98 (c = 92.7%, sensitivity = 92.54%, specificity = 92.74%) were the most accurate cut points for the 33.33% and 50% PA guidelines, respectively. The steps/min guidelines of 60.84 and 79.98 applied to genders (c = 87.93–93.70%, sensitivity = 87.53–93.29%, specificity = 88.63–93.90%) and grade level groups (c = 84.91–96.29%, sensitivity = 83.61–9 6.69%, specificity = 86.67–95.24%) demonstrated acceptable validity outcomes. Conclusions: In summary, the factors of lesson duration, stature, BMI, gender, and grade level were not found to limit the validity of a common steps/min guideline for either PE PA guideline. These findings support the contention that a common steps/min guideline can be applied in the surveillance of PE PA.
Keyword(s): measurement/evaluation, physical education PK-12, research

Back to the 2009 AAHPERD National Convention and Exposition (March 31 - April 4, 2009)