An evaluation of two pilot projects of inclusion of students from special schools for moderate and severe learning difficulties in mainstream primary and secondary schools in south-west England stimulated the development of a methodology and tools to measure levels and quality of inclusive activity. The unresolved theoretical debate about what constitutes inclusion makes pragmatic approaches desirable for evaluation of practice. The need for contextual, attitudinal and institutional factors to be taken into account is discussed. It is argued that comparison of perceptions helps to clarify some of the factors that may be influential in successful development of inclusion in schools. A phenomenological approach places value on the affective component of "voice" as a "true" measure of success. A formative model for evaluation, with multiple data sources, is utilized to structure a complex data collection and evaluative process.