Multilevel modeling and mixed longitudinal design in the research in Physical Education and Sport Medicine

Abstract

The main aim of this study was to introduce the use of multilevel modeling in the data from a mixed longitudinal study concerning growth, development and health of Portuguese adolescents. The sample comprises 5155 subjects from 4 age cohorts: 10 to 12, 12 to 14, 14 to 16 and 16 to 18 years. Variables chosen for analysis were handgrip strength and percentage of total body fat. Multilevel modeling analysis showed a quadratic trajectory of adolescents' handgrip strength performance, with higher annual increasing in boys. The percentage of total body fat was positively associated with the static strength and there were marked interindividual differences. The use of a mixed longitudinal design allowed a faster data collection process; in addition, multilevel modeling analysis was sufficiently flexible and robust to accommodate, in a single model, aspects of individual and interindividual changes in static strength development over the time.