Reliability and Construct Validity of Yo-Yo Tests in Untrained and Soccer-Trained Schoolgirls Aged 9-16.

Abstract

The reliability and construct validity of three age-adapted-intensity Yo-Yo tests were evaluated in untrained (n = 67) vs. soccer-trained (n = 65) 9- to 16-year-old schoolgirls. Tests were performed 7 days apart for reliability (9- to 11-year-old: Yo-Yo intermittent recovery level 1 children's test; 12- to 13-yearold: Yo-Yo intermittent endurance level 1; and 14- to 16-year-old: Yo-Yo intermittent endurance level 2). Yo-Yo distance covered was 40% (776 ± 324 vs. 556 ± 156 m), 85% (1252 ± 484 vs. 675 ± 252 m) and 138% (674 ± 336 vs. 283 ± 66 m) greater (p ≤ .010) for the soccer-trained than for the untrained girls aged 9-11, 12-13 and 14-16 years, respectively. Typical errors of measurement for Yo-Yo distance covered, expressed as a percentage of the coefficient of variation (confidence limits), were 10.1% (8.1-13.7%), 11.0% (8.6-15.4%) and 11.6% (9.2-16.1%) for soccer players, and 11.5% (9.1-15.8%), 14.1% (11.0-19.8%) and 10.6% (8.5-14.2%) for untrained girls, aged 9-11, 12-13 and 14-16, respectively. Intraclass correlation coefficient values for test-retest were excellent (0.795-0.973) in both groups. No significant differences were observed in relative exercise peak heart rate (%HRpeak) between groups during test and retest. The Yo-Yo tests are reliable for determining intermittent-exercise capacity and %HRpeak for soccer players and untrained 9- to 16-year-old girls. They also possess construct validity with better performances for soccer players compared with untrained age-matched girls, despite similar %HRpeak.