ABSTRACT
Designs can be classified according to the number of IVs and DVs that they contain and the aims of the research. They can involve the same participants in more than one condition or they can employ different participants in different conditions. Designs also differ in the degree to which they measure participants at different stages in a process. Although it is possible to maximise the internal validity of a design in laboratory experiments, much research is conducted outside the laboratory. In this case, researchers have to choose the most internally valid design which is available to them in the circumstances. No design is perfect, but some are more appropriate than others to answer a particular research question. Where possible it is best to allocate participants to the different conditions on a random basis.
