13.07.2015 Views

Contents

Contents

Contents

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

324 PART IV: Applied ResearchThe dashed line indicates that the treatment and comparison groups were notformed by assigning participants randomly to conditions.By adding a comparison group, researchers can control threats to internalvalidity due to history, maturation, testing, instrumentation, and regression. Abrief review of the logic of experimental design will help show why this occurs.We wish to begin an experiment with two similar groups; then one groupreceives the treatment and the other does not. If the two groups’ posttest scoresdiffer following treatment, we first must rule out alternative explanations beforewe can claim that treatment caused the difference. If the groups are truly comparable,and both groups have similar experiences (except for the treatment),then we can assume that history, maturation, testing, instrumentation, andregression effects occur to both groups equally. Thus, we may assume that bothgroups change naturally at the same rate (maturation), experience the sameeffect of multiple testing, or are exposed to the same external events (history).If these effects are experienced in the same way by both groups, they cannotpossibly be used to account for group differences on posttest measures. Therefore,they no longer are threats to internal validity. Thus, researchers gain atremendous advantage in their ability to make causal claims simply by adding acomparison group. These causal claims, however, depend critically on formingcomparable groups at the start of the study, and ensuring that the groups thenhave comparable experiences, except for the treatment. Because this is difficultto realize in practice, as we’ll see, threats to internal validity due to additiveeffects with selection typically are not eliminated in this design.As you approach the end of a course on research methods in psychology,you might appreciate learning about the results of a nonequivalent controlgroup design that examined the effect of taking a research methods course onreasoning about real-life events (VanderStoep & Shaughnessy, 1997). Studentsenrolled in two sections of a research methods course (and who happened to beusing an edition of this textbook) were compared with students in two sectionsof a developmental psychology course on their performance on a test emphasizingmethodological reasoning about everyday events. Students in both kindsof classes were administered tests at the beginning and at the end of the semester.Results revealed that research methods students showed greater improvementthan did students in the control group. Taking a research methods courseimproved students’ ability to think critically about real-life events.With that bit of encouraging news in mind, let us now examine in detailanother study using a nonequivalent control group design. This will give us theopportunity to review both the specific strengths and limitations of this quasiexperimentalprocedure.Nonequivalent Control Group Design:The Langer and Rodin Study• Quasi-experiments often assess the overall effectiveness of a treatmentthat has many components; follow-up research may then determine whichcomponents are critical for achieving the treatment effect.

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

Saved successfully!

Ooh no, something went wrong!