Between subjects designs are experimental studies in which each participant is assessed with regard to a single dependent variable (DV) after having been presented with one level of the independent variable (IV). A key factor of between subjects designs is independence: each participant has only one score on the dependent variable, which is unrelated to other participants’ scores. Moreover, because of this independence, an individual score is unlikely to be influenced by external factors that introduce error into the study. Between subjects designs are often quick and efficient, which means that participants do not get bored, and researchers do not have to worry about attrition—losing participants over the course of the study.
However, between subjects designs are not without their disadvantages. The primary concern for these types of studies is non-equivalent groups. It is possible that important personal characteristics are not evenly distributed across all experimental groups, which introduces into the study an additional difference above and beyond the level of independent variable. In order to avoid this, researchers rely upon a number of strategies including random assignment, matched assignment, constant variables, and large sample sizes.
It may also cause some problems if participants, particularly those in the control group, are aware of the purpose and conditions of a study. Compensatory equalization, compensatory rivalry, and resentful demoralization all can affect control participants’ actions, attitudes, and approach to the study, undermining the validity of the research.
Interpretation of the results of a between subjects study depends upon the number of groups in the study. The simplest design compares two groups, and researchers would simply compare the mean dependent measure score of participants in one condition to that of participants in the other condition. If this overlap between these two conditions is sufficiently small, we can reject the null hypothesis—the expectation that there will be no difference based on condition.
Oftentimes psychologists want to use more than two conditions in their research. Instead of comparing two groups along a single dependent variable, they might design the study to have three or more levels of the independent variable. In this case, we compare the variability that exists between groups to the variability within groups. The greater the variability between groups, and the less the variability within, the less overlap there is between conditions, and the more likely the effect of condition is significant.
Factorial designs are slightly more complicated between subjects designs as they include more than one independent variable with a single dependent variable. In these studies we must include analyses that consider main effects (the effects of each of the independent variables alone), but also interactions (the combined effect of the different levels of our independent variables).
Additional Online Resources
Video on between-subjects design: http://study.com/academy/lesson/between-subjects-designs-definition-examples.html#lesson
Online tutorial about factorial research design from SUNY-Brockport: http://www.acs.brockport.edu/~mdesroch/Factorial3/
Podcast on factorial designs by Michael Britt: http://www.thepsychfiles.com/2008/03/episode-52-repost-research-design-part-2-factorial-designs/
Test your knowledge of the keywords and definitions in the chapter.
Interactive Quiz for Chapter 9
Instructions: For each question, click on the radio button beside your answer. When you have completed the entire quiz, click the “Submit my answers” button at the bottom of the page to receive your results.