Psychometric proprieties of the Test of Gross Motor Development-Third Edition in a large sample of Italian children

Abstract

Objectives: The Test of Gross Motor Development-3 (TGMD-3) evaluates fundamental gross motor skills across two domains: locomotor and ball skills. This study aimed to perform a full psychometric assessment of this test in a large sample of Italian pre- and primary school children.

Design: Cross-sectional and test-retest study design.

Method: Children N = 5210; mean age years = 8.38, SD = 1.97; % females = 48 completed three trials, including one practice. Only the scores of the two latter 'formal' trials were recorded for the evaluation. Factorial validity and measurement invariance of TGMD-3 across age and gender groups and test-retest reliability for the overtime measure consistency were tested. Item response theory analysis further tested single items' performances.

Results: Explorative and confirmatory factor analyses confirmed the two-factor structure of the TGMD-3. Multi-group confirmatory factor analyses indicated that there were no significant reductions in model adjustments between the configural, metric and structural invariance solutions for gender and age groups. Test-retest results ranged between 0.967 and 0.990 for both skill sets across age groups. Item response theory analysis using a graded response model showed low standard error and high-test information levels covering a wide spectrum range of both locomotor and ball skills.

Conclusions: These results highlight the strong construct validity and reliability of the TGMD-3 to measure gross motor skills in children across gender and age groups. Item response theory analysis evidenced how the performance criteria included in this test cover a wide range of gross the motor skills spectrum. The use of TGMD-3 may inform motor development programs and support curricular decisions in schools.

Keywords: Item response theory; Motor development; Motor skills; Multigroup analysis.