aes19 International Evaluation Conference

Gill Westhorp  (Charles Darwin University)

What do we know? What can we know, and how do we know that we know it? These are philosophical questions with real implications for the practice of evaluation. Epistemology is the branch of philosophy dealing with the nature of knowledge. Different epistemologies underpin different approaches in research and evaluation. They have implications for what data is considered to be 'valid', how data can or should be collected, how data is analysed and interpreted, and under what conditions findings are portable to other contexts.

This paper deals with two epistemologies - realist and constructivist - from a realist viewpoint. Some authors have claimed that realists 'are realists ontologically, but constructivists epistemologically'. That is, realists believe that there is a real world, which exists independently of our interpretations of it ("realist ontology"). However, we all construct our own interpretations of it. Knowledge is not a direct representation of reality, but an interpretation of it, constructed in our own heads, and shaped by language, culture, personal experience, and previous learning ("constructivist epistemology"). Knowledge does not exist independently of 'the person who knows'. In radical constructivism, we cannot even be sure that there is a real world. Perhaps we are all just avatars in some giant computer game.

This paper argues that there are areas of overlap, but also areas of distinction between, realist and constructivist epistemology. These distinctions have implications for evaluation practice. It will briefly describe the key assumptions of constructivism, and contrast these with key assumptions in realism. It will use a hypothetical evaluation as an example to discuss differences in: the purposes of constructivist and realist investigation; the nature of the data that is collected; the ways that analysis is undertaken; how 'valuing' is approached and how evaluation adds value; the nature of findings; and the portability of findings.