Anthropometric analysis and performance characteristics to predict selection in young male and female handball players

Abstract The aim of this study was two-fold. The first aim was to determine if there were any anthropometric and physical performance differences (controlling for maturation) between male and female handball players selected in training categories as well asthe relation of these differences with the performance level achieved. The second aim was to identify the discriminatory variables between the performance levels achieved. A total of 216 young handball players (125 men and 91 women) participated in the study. The data were classified by selection level (regional n=154; national n=62), gender (men; women) and age category (under-15; under-17). The use of MANCOVA analyses, controllingfor maturation, identified how gender could determine variables related to handball players' future competitive levels. The results revealed that anthropometric variables such as height, arm span, trochanter height, thigh girth, and leg girth were more influential in men than in women. In addition, the physical performance tests of vertical jump (squat jump and counter movement jump with/without arm) and 10x5m shuttle run were determinants in both sexes. Discriminatory analysis predicted that a combination of five variables (counter movement jump with arm, body mass, 10x5m shuttle run, dominant hand length and trochanter height) would successfully distinguish between regional and national players, with a predictive accuracy of 81.9% for all players.

摘要 本研究旨在达成双重研究目标。其一，在控制成熟度变量的前提下，探究不同训练梯队选拔出的男女手球运动员在人体测量学指标与身体运动表现方面是否存在差异，以及此类差异与运动员所达成的竞技水平之间的关联；其二，识别出可用于区分不同竞技水平运动员的判别变量。本研究共纳入216名青年手球运动员，其中男性125名、女性91名。研究依据选拔层级（区域级n=154；国家级n=62）、性别（男、女）以及年龄组别（U15、U17）对数据进行分类。通过采用控制成熟度的多变量协方差分析（Multivariate Analysis of Covariance，MANCOVA），本研究明确了性别对与手球运动员未来竞技水平相关变量的影响模式。研究结果显示，身高、臂展、大转子高度、大腿围度以及小腿围度等人体测量学变量对男性运动员的影响程度显著高于女性。此外，垂直跳跃测试（深蹲跳与无臂/带臂摆动的反向跳跃）以及10×5米折返跑测试的成绩，在两性中均为关键判别指标。判别分析结果表明，结合五项变量（带臂摆动的反向跳跃、体质量、10×5米折返跑、利手长度以及大转子高度）可有效区分区域级与国家级运动员，全体样本的预测准确率达81.9%。