Paul Stapleton's (2006) critique of quantitative research brings to the surface some common interpretive problems arising when experimental and quasi-experimental research designs are compared. While Stapleton may be correct in pointing out the superiority of experimental research designs because they best eliminate the influence of extraneous factors, his construction of factitious intervening variables do little to support his argument that naturalistic field studies are invalidated by imaginary spurious influences on the outcomes of interest. In this reply, the differences between randomized experiments and real-world effectiveness studies are also reviewed with the aim of stressing the potential practical utility of the latter. While it is well-known that naturalistic field studies rarely provide conclusive results, it is argued that they can still provide useful evidence to justify inferences about worthwhile assessment practices.