Abstract

This paper critiques the ways in which experimental and quasi-experimental research designs are represented as “standard templates” in evaluation/research methods textbooks. This has implications for how social scientists and evaluators are taught and the number of possible research designs that can be applied in programme evaluations. To move the field forward, it is proposed that we revisit the idea of “design elements” introduced in Shadish, Cook and Campbell (2002)’s seminal work. Given the complex context in which many social and educational programmes are situated, new and innovative designs can be “constructed” from design elements. This allows the design of experiments and quasi-experiments to be “flexible enough so that some of them will be relevant for improving clausal claims in almost any research context” (Shadish, Cook & Campbell, 2002,p.xviii). This paper demonstrates how most experimental and quasi-experimental research designs can be reduced to five design elements and introduces a mnemonic to facilitate the design of experiments.