Psy 330

Experimental Designs:  Between-Subjects Design

 

I.  Experimental Research Strategy

    A. Manipulation of IV

    B. Control of extraneous variables

    C. Basic Experimental Designs

        1.  between-subjects design--Participants serve in only one treatment condition.

        2. within-subjects design--Participants serve in more than one treatment condition. 

    D. Structure of Between-Subjects Designs

A separate group of participants is assigned to each of the different conditions by using random selection.

Groups are presumed to be equal at start so that we can conclude any difference in DV is due to effect of IV. 

In order to assure equivalent groups, the groups must be created equally, treated equally, and composed of equivalent individuals.

 

    E. Advantages & Disadvantages of Between-Subjects Designs

Pro: Each participant is exposed to only one treatment condition. Therefore, score is free from the following influences: (1) practice/experience, (2) fatigue/boredom, and (3) contrast effects. Also, usable for wide variety of research questions.

Cons: (1) Need relatively large N and (2)each score is obtained from a different person--all possessing different characteristics. 

         Individual Differences--e.g., gender, age, personality, background

Caution: Individual differences can become confounding variables and can produce high variability in the scores.

        

II. Limiting Confounding by Individual Differences

        Random Assignment (randomization)

        Matching Groups (matched assignment)

        Holding Variables Constant

 

 

III.  Variability and Individual Differences

Figure 10.3  Results From a Simulated Experiment Comparing the Conditions Using Participants Selected From a Population Where Individual Differences are Relatively Small (p. 253)

Figure 10.4  Results from a Simulated Experiment Comparing Two Conditions using Participants Selected from a Population Where Individual Differences are Relatively Large (p. 254)

       Differences between treatments and variability within treatments

 

        Minimizing variability within treatments

            Standardize Procedures and Treatment Setting

            Limit Individual Differences

            Random Assignment and Matching

            Sample Size

 

IV.  Other Threats to Internal Validity of Between-Subjects Designs

        Assignment Bias

        Differential Attrition

        Diffusion or Imitation of Treatment

        Compensatory Equalization

        Compensatory Rivalry

        Resentful Demoralization

Engagement Activities Ch. 8

1) A researcher has a sample of 30 rats that are all cloned from the same source.  The 30 rats are genetically identical and have been raised in exactly the same environment since birth.  The researcher conducts an experiment, randomly assigning the remaining 10 clones to Treatment A, 10 to Treatment B, and the other 10 to Treatment C.  Explain why the clone experiment is better than a between-subjects study using 30 regular rats that are randomly assigned to the three conditions. In other words, explain how the clone experiment eliminates the basic problems with a between-subjects study.

 

2) A recent survey at a major corporation found that employees who regularly participated in the company fitness program tended to have fewer sick days than employees who did not participate.  However, because the study was not a true experiment, you cannot conclude that regular exercise causes employees to have fewer sick days.

a) Identify another factor (a confounding variable) that might explain why some employees participated in the fitness program, and why those same employees have fewer sick days.

b) Describe the design for a between-subjects experiment that would determine whether participating in the exercise program caused fewer sick days.

c) Describe how the factor you identified in Part a is controlled in your experiment.