13.07.2015 Views

Contents

Contents

Contents

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

216 PART III: Experimental MethodsFIGURE 6.5Random assignment is not likely to be effective in balancing differences among subjects whensmall numbers of subjects from heterogeneous populations are tested (e.g., newborns). In thissituation, researchers may want to consider the matched groups design.same as that for the random groups design (see Figure 6.5). In most uses of thematched groups design, a pretest task is used to match subjects. The challengeis to select a pretest task (also called a matching task) that equates the groupson a dimension that is relevant to the outcome of the experiment. The matchedgroups design is useful only when a good matching task is available.The most preferred matching task is one that uses the same task that willbe used in the experiment itself. For example, if the dependent variable in theexperiment is blood pressure, participants should be matched on blood pressureprior to the start of the experiment. The matching is accomplished by measuringthe blood pressure of all participants and then forming pairs or triplesor quadruples of participants (depending on the number of conditions in theexperiment) who have identical or very similar blood pressures. Thus, at thestart of the experiment, participants in the different groups will have, on average,equivalent blood pressure. Researchers can then reasonably attribute anygroup differences in blood pressure at the end of the study to the treatment(presuming other potential variables have been held constant or balanced).In some experiments, the primary dependent variable cannot be used tomatch subjects. For example, consider an experiment that teaches participantsdifferent approaches to solving a puzzle. If a pretest were given to see how longit took individuals to solve this puzzle, the participants would likely learn thesolution to the puzzle during the pretest. If so, then it would be impossible toobserve differences in the speed with which different groups of participantssolved the puzzle following the experimental manipulation. In this situationthe next best alternative for a matching task is to use a task from the same classor category as the experimental task. In our problem-solving experiment, participantscould be matched on their performance when solving a different puzzle

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!