Applied multivariate statistics for the social sciences
Applied multivariate statistics for the social sciences
Behaviour & Information Technology - Computer Support for Learning Communities
Hi-index | 0.00 |
Research about the effect of cooperative learning settings faces the challenge of dealing with hierarchical data where observations regarding the learners are not stochastically independent. Standard methods like ANOVAs cannot deal with such data adequately. The article introduces multilevel modelling (MLM) as a statistical approach adequate for nested data. MLM allows taking into account interactions between group-level variables and individual-level variables. But MLM requires large sample sizes, and thus many studies fail to adopt MLM. For such studies, it is proposed that some additional statistics be presented.