Lenses and their blind spots: A comparative framework for evaluation paradigms

Abstract

Evaluation approaches, or paradigms, are best understood as lenses: ways of looking at evaluands that are chosen, not inevitable. Framing them this way presents evaluation traditions as options to be selected rather than positions to be defended, inviting evaluators to ask not ‘which paradigm is correct?’ but ‘which lens best serves this evaluation’s context and purpose?’ The lens metaphor offers two practical benefits. First, it promotes flexibility, encouraging evaluators to expand their repertoire and guiding them towards the right lens for each situation. Second, it offers a common framework for exploring what evaluation is, making convergence and divergence across paradigms more apparent and improving evaluative discourse and practice. The paper compares six approaches: a-theoretical (or accountability), experimental, realist, constructivist, pragmatic, and systems evaluation, across 10 dimensions, including purpose, causality, validity, and each lens’s blind spot. The aim is not to resolve paradigm debates but to structure them usefully.