Predictive validity of predictors in a new talent identification programme by the German Handball Federation

(Prognosevalidität von Prädikatoren eines neues Talentsichtungsprogramms des Deutschen Handballbunds)

The German Handball Federation conducts a talent identification programme (TIP) with female (15-16 years) and male (16-17 years) youth handball players every year. In 2008, the DHB launched a modified TIP with anthropometric, motor and psychological tests (Elferink-Gemser et al., 2005; Lidor et al., 2005; Mohamed et al., 2009) with the aim to improve the predictive validity of the talent identification. This study evaluates the modified TIP in general and especially the predictive validity of the different tests. Methods: Youth handball players from 20 regional selection handball teams (N = 542, 259 female and 283 male, 14 to 16 years) performed a series of tests. The number of items for the following analyses was reduced to eight potential predictors by ratings. Nomination by the national coaches was the outcome variable. First, a multivariate analysis of variance with the factors nomination and gender was conducted. Results: The results showed significant main effects for nomination (Wilks` Lambda = .87, F[6, 169] = 4.39, p < .001, Eta² = .14) and for gender (Wilks` Lambda = .21, F[6, 169] = 101.37, p < .001, Eta² = .78), but no interactions between both. Therefore stepwise discriminant analyses were carried out separately for each gender. For boys, 30-m sprint is mainly responsibly for a large explained variance (55.9%). For girls, this is the 30-m sprint, ball throwing velocity, and achievement motivation (67.6%). Discussion: The results indicate that singular tests differentiate between nominated and not-nominated female or male youth handball players (Mohamed et al., 2009) but the predictive validity is quite low. The predictors are just not sensitive and specific enough (Lidor et al., 2005). Therefore, TIP will be expanded by tactical tests (Elferink-Gemser et al., 2004) with open skills assuming that an increase of the predictive validity will occur. Additionally, a longitudional study will be conducted to test long-term effects.
© Copyright 2012 17th Annual Congress of the European College of Sport Science (ECSS), Bruges, 4. -7. July 2012. Veröffentlicht von Vrije Universiteit Brussel. Alle Rechte vorbehalten.

Schlagworte: Talent Eignung Auswahl Handball Deutschland Sportverband Test Untersuchungsmethode männlich weiblich Jugend
Notationen: Nachwuchssport Spielsportarten
Veröffentlicht in: 17th Annual Congress of the European College of Sport Science (ECSS), Bruges, 4. -7. July 2012
Herausgeber: R. Meeusen, J. Duchateau, B. Roelands, M. Klass, B. De Geus, S. Baudry, E. Tsolakidis
Veröffentlicht: Brügge Vrije Universiteit Brussel 2012
Seiten: 438
Dokumentenarten: Kongressband, Tagungsbericht
Sprache: Englisch
Level: hoch