another anonymous reviewer suggested several additional methodological tests that could be conducted, including a confirmatory factor analysis to confirm or revise the subscale structure. other research could include locating centers about to undergo significant renovation and conducting pre- and post-renovation cpers scores, and examining cpers in relation to socio-economic status, as previous research has suggested this has a robust correlation to teaching characteristics (salaries, educational qualifications, etc.) and educational philosophies. the latter could be especially interesting. there is considerable evidence in other sub-fields of environment-behavior studies that considering moderating constructs in combination with physical environmental independent variables leads to stronger predictions of the dependent variables and greater overall understanding of the environment-behavior nexus, or, in this case, the child-environment nexus.