Reliability and validity of the Test of Gross Motor Development 3 (German version) : Results

Abstract

Objective: To assess the psychometric properties of the German version of the Test of Gross Motor Development 3 (TGMD-3). Methods: Performances of 117 typically developed children (58 boys; 18 kindergarten and 99 elementary school children; Mage: 7.99 years, SD = 1.98 years) as well as 10 children with developmental coordination disorder (DCD; 8 boys; Mage: 7.88 years, SD = 1.57 years) on 12 gross motor tasks were analyzed on the basis of the respective video recordings. Results: Results indicate an excellent 2-week test–retest (N = 69; total: ICC = .983**; locomotor skills: ICC = .941**; ball skills: ICC = .977**), interrater- (2 raters; N = 30; total: ICC = .960**; locomotor skills: ICC = .884**; ball skills: ICC = .973**) as well as intrarater- (N = 30; total: ICC = .988**; locomotor skills: ICC = .972**; ball skills: ICC = .991**) reliability. Internal consistency of the two subscales (N = 110; locomotor skills: Cronbach’s α = 744; ball skills: Cronbach’s α =.890) ranges between acceptable and good. Confirmatory factor analysis provides evidence for the postulated two-factor model (N = 110; CMIN = 54.83; df = 53; p = .405; CMIN/df = 1.04; CFI = .997; RMSEA = .018; SRMR = .049). Substantial correlations between the TGMD-3 and the M-ABC 2 subscales (N = 110) indicate the presence of convergent (TGMD-3 locomotor skills × M-ABC 2 balance: r = .467**; TGMD-3 ball skills × M-ABC 2 aiming and catching: r = .436**) or divergent (TGMD-3 locomotor skills × M-ABC 2 aiming and catching: r = .334**; TGMD-3 ball skills × M-ABC 2 balance: r = .339**) validity, respectively. Indications for concurrent validity can be found in the relationship between the TGMD-3 and the German youth games total score (N = 91; r = .337**) in elementary school. Performance differences between children with and without DCD (total: F = 3.72; df = 1; p = .035; part. η2 =.171; locomotor skills: F = 4.52; df = 1; p = .024; part. η2 =.201; ball skills: F = 2.50; df = 1; p = .066; part. η2 =.122) support the clinical validity of the TGMD-3 in this subgroup. Conclusion: Our findings provide evidence for the reliability and validity of the German version of the TGMD-3.