ABSTRACT
The distinguishing feature of social scientific investigation is its inability to conduct controlled investigation. The (theoretical) method of the ‘natural’ scientist is to scrutinize hypothesized cause-and-effect relationships by performing pairs of matched experiments. To one group, the ‘experimental group’, an agent will be applied which is suspected of inducing a particular change. The group’s progress will be monitored to see if the change does actually occur. To ensure that the change is not merely the product of the passage of time, or some other extraneous factor, progress will also be monitored in a ‘control group’. That is, a group will be chosen whose characteristics have been, as far as is conceivably possible, matched to those of the experimental group in all relevant respects.
