Reliability and Validity of a Novel Reactive Agility Test with Soccer Goalkeeper-Specific Movements

The purpose of this study was to develop a reactive agility test with soccer goalkeeper (GK)-specific movements (G-RAT) and to examine the reliability and validity of college-aged GKs. We designed a five-branch star-shaped course with diving and ball-catching movements under reactive conditions. In the setup, a start-goal line was set on the top of a branch and 3.5 m away from the center of the star-shaped setting. Content validity was assessed by six experts, and the item-content validity index (I-CVI) was calculated. Thirty-three male GKs performed the test trial twice. One test trial of G-RAT consists of three shuttles from the start-goal line to diving and ball-catching. For the reactive condition, GKs were instructed on which ball directions should dive when their body trunk reached 1.5 m away from the start-goal line. GKs were classified into regular (R) or non-regular (NR) groups. The intraclass correlation coefficient (ICC) and the receiver operating characteristic (ROC) curve analyses were used to assess the reliability and predictive power as convergent validity. The I-CVI was 0.83, which was greater than the acceptable level of 0.78. The ICC value was 0.94 (p < 0.01; 95% confidence interval (95%CI), 0.88-0.97). The GKs completed the test 14.3 +/- 0.7 and 15.3 +/- 1.0 s in the R and NR group (p < 0.01; Cohen's d = 0.89), respectively. The area under the curves of G-RAT was 0.80 (95%CI, 0.64-0.96). These results show that a GK-specific agility test under reactive conditions would have sufficient reliability and both content and convergent validity in college-aged GKs.