Method effects on theory testing: the case of organizational coupling in education

Shows how methodological choices are not theoretically neutral. Draws attention to ways in which different analyses of the same data may affect inferences about teachers' policy coupling in school systems. In this case study of the Jerusalem school system the authors used three statistical procedures to assess teachers' perceived policy alignment among three organizational levels (teachers; schools; and the Israel Ministry of Education (IMOE)). Analyses using descriptive statistics show that the perceived policies at the three levels are similar, thus giving some support to the theory of tight coupling. Smallest Space Analysis shows that there are close connections between the teachers' own policies and those they impute to schools, but not those they impute to the IMOE. These findings support a moderate view of organizational coupling. Finally, variance component analyses find almost no consensus in schools regarding policies. In contrast to the other approaches, these models support a loose coupling hypothesis. In overview, shows how methodological choices affect the support given to rival theoretical hypotheses. Suggests that theoretical looseness with regard to explicit falsification conditions is at the root of contrasting evidence about teacher coupling in school systems.