Improvements in Predicting PACER Performance From BMI: Pretest Versus Posttest

• Published in: Research quarterly for exercise and sport 2016-06, Vol.87 (S2), p.A10
• Main Authors: Liu, Wenhao, Hull, Ethan E, Istvan, Kovacs
• Format: Article
• Language: English
• Subjects: Body Composition; Body mass index; College Students; Females; Males; Physical fitness; Regression (Statistics); Scientific Concepts; University students
• ISSN: 0270-1367; 2168-3824

When cardiorespiratory fitness is improved after interventions, the correlation/regression strength between body composition and cardiorespiratory fitness might change accordingly. This study examined how prediction of Progressive Aerobic Cardiovascular Endurance Run (PACER) performance from body mass index (BMI) would change after a 1-semester physical activity (PA) class among university students. At the beginning (pretest) and end (posttest) of a semester, BMI (weight [kg]/height [m2]) and the 20-m PACER were assessed for 120 university students (58 men, Mage = 21.69 ± 1.74 years; and 62 women, Mage = 21.34 ± 2.05 years) who enrolled in PA classes in the United States, and linear regression analysis was conducted to evaluate the prediction of PACER laps from BMI. The prediction/correlation strengths at the 2 assessment points were compared, and BMI and PACER performance between the 2 assessment points were also compared with a paired-samples t test. During the 1-semester interval, BMI remained unchanged for men (26.20 ± 3.07 vs. 26.18 ± 3.15, p = .77) and women (23.73 ± 3.55 vs 23.70 ± 3.20, p = .68), whereas PACER laps improved significantly ( p , .001) for men (57.95 ± 18.10 vs. 68.57 ± 17.40) and women (39.97 ± 14.21 vs. 51.21 ± 15.60). Accordingly, compared with pretest, prediction/correlation strength between BMI and the PACER increased at posttest with larger absolute values of r = 2.31 versus2.37 for men and r = 2.39 versus2.44 for women, larger r 2 values of .10 versus .14 for men and .15 versus .19 for women, and larger slopes and F values, and more significant p values associated with regression equations. Specifically, for men, y = 106.40 2 1.85 BMI, 95% CI [23.35,2.35], F = 6.1, p , .05 versus y = 121.88 2 2.04 BMI, 95% CI [23.41, 20.66], F = 8.79, p , .005; and for women, y = 76.87 2 1.56 BMI, 95% CI [22.51, 20.60], F = 10.65, p , .005 versus y = 102.05 2 2.15 BMI, 95% CI [23.28, 21.02], F = 14.42, p , .001. With the significant improvement in PACER performance among university students, the proportion of PACER performance variance accounted for by BMI increases from 10% (r 2 = .10) to 14% (r 2 = .14) for men and from 15% (r 2 = .15) to 19% (r 2 = .19) for women. Further, with each unit increase in BMI, there is a greater decrease in PACER laps for men (from 1.85 laps to 2.04 laps) and for women (from 1.56 laps to 2.15 laps) at the posttest as opposed to the pretest, indicating a stronger relationship between BMI and PACER performance after