Chapter 9 Using Between-Subjects and WithinSubjects Experimental Designs © 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Experimental Design Used when your goal is to establish causal relationships between variables and you can manipulate variables. To manipulate the independent variable you set its value to at least two different values (levels). You can manipulate it: Quantitatively (amount of exposure to same variable) Qualitatively (use different exposures) Adapted from © 2005 The McGraw-Hill Companies, Inc.. Types of Experimental Designs Between-Subjects Design Within-Subjects Design Different groups of subjects are randomly assigned to the levels of your independent variable Data are averaged for analysis A single group of subjects is exposed to all levels of the independent variable Data are averaged for analysis Single-Subject Design Single subject, or small group of subjects is (are) exposed to all levels of the independent variable Data are not averaged for analysis; the behavior of single subjects is evaluated Adapted from © 2005 The McGraw-Hill Companies, Inc.. The Problem of Error Variance Error variance is the variability among scores not caused by the independent variable Error variance is common to all three experimental designs Error variance is handled differently in each design Sources of error variance Individual differences among subjects Environmental conditions not constant across levels of the independent variable Fluctuations in the physical/mental state of an individual subject Adapted from © 2005 The McGraw-Hill Companies, Inc.. Handling Error Variance Taking steps to reduce error variance Increasing the effectiveness of the independent variable Hold extraneous variables constant by treating subjects as similarly as possible Match subjects on crucial characteristics Strong manipulations yield less error variance than weak manipulations (e.g. Greater increase in dosage of medication) Randomizing error variance across groups Distribute error variance equivalently across levels of the independent variable Accomplished with random assignment of subjects to levels of the independent Adapted from © 2005 The McGraw-Hill Companies, Inc.. Statistical analysis Random assignment tends to equalize error variance across groups, but not guarantee that it will You can estimate the probability that observed differences are due to error variance by using inferential statistics Adapted from © 2005 The McGraw-Hill Companies, Inc.. Between-Subjects Designs I. Single-Factor Randomized Groups Design The randomized two-group design (see page 266) Randomly assign to two groups (experimental and control) Expose the two groups to different levels of indep. Variable Hold extraneous variables constant Compare the two means Advantages include: Simple, requires fewer subjects, no pretesting required, analysis is simple Disadvantages include: Provides limited amount of information about effect of independent variable (see example page 267 text) Limited sensitivity to effect when subjects differ greatly in characteristics that influence their performance on dep. measure Limited at detecting limits of an effect (need more levels) Adapted from © 2005 The McGraw-Hill Companies, Inc.. Between-Subjects Designs Cont. The randomized multiple group design Additional levels of the independent variable can be added to form a MULTIGROUP DESIGN If different levels of the independent variable represent quantitative differences, the design is a PARAMETRIC DESIGN If different levels of the independent variable represent qualitative differences, the design is a NONPARAMETRIC DESIGN When you manipulate your indep variable quantitatively you are using a parametric design Parametric- refers to the systematic variation of the amount of the independent variable. A variation of this method/design is the multiple control group design Control Group Placebo Group Treatment Group Adapted from © 2005 The McGraw-Hill Companies, Inc.. Between-Subjects Designs Cont. II. Matched-Groups Designs (see page 270) Steps Obtain a sample of subjects Measure the subjects for a certain characteristic (e.g., intelligence) that you feel may relate to the dependent variable Match the subjects according to the characteristic (e.g., pair subjects with similar intelligence test scores) to form pairs of similar subjects Randomly assign one subject from each pair of subjects to the control group and the other to the experimental group Carry out the experiment in the same manner as a randomized group experiment Advantages Distributes the characteristic evenly across treatments. Allows you to control subject variables that obscure results. May require fewer subjects Disadvantages Less statistical power in analysis used which decreases ability to detect differences. Adapted from © 2005 The McGraw-Hill Companies, Inc.. Between-Subjects Designs Cont. The matched-pairs design Equivalent to the randomized multi-group design. The matched multigroup design Adapted from © 2005 The McGraw-Hill Companies, Inc.. Within-Subjects Designs Subjects are not randomly assigned to treatment conditions The same subjects are used in all conditions Closely related to the matched-groups design Advantages Reduces error variance due to individual differences among subjects across treatment groups Reduced error variance results in a more powerful design Effects of independent variable are more likely to be detected Adapted from © 2005 The McGraw-Hill Companies, Inc.. Disadvantages More demanding on subjects, especially in complex designs Subject attrition is a problem Carryover effects: Exposure to a previous treatment affects performance in a subsequent treatment Adapted from © 2005 The McGraw-Hill Companies, Inc.. Sources of Carryover Learning Fatigue Learning a task in the first treatment may affect performance in the second Fatigue from earlier treatments may affect performance in later treatments Habituation Repeated exposure to a stimulus may lead to unresponsiveness to that stimulus Adapted from © 2005 The McGraw-Hill Companies, Inc.. Sensitization Contrast Exposure to a stimulus may make a subject respond more strongly to another Subjects may compare treatments, which may affect behavior Adaptation If a subject undergoes adaptation (e.g., dark adaptation), then earlier results may differ from later ones Adapted from © 2005 The McGraw-Hill Companies, Inc.. Dealing With Carryover Effects Counterbalancing The various treatments are presented in a different order for different subjects May be complete or partial The Latin Square Design Used when you make the number of treatment orders equal to the number of treatments Adapted from © 2005 The McGraw-Hill Companies, Inc.. Taking Steps to Minimize Carryover Techniques such as pre-training, practice sessions, or rest periods between treatments can reduce some forms of carryover Make Treatment Order an Independent Variable Allows you to measure the size of carryover effects, which can be taken into account in future experiments Adapted from © 2005 The McGraw-Hill Companies, Inc.. Example of a Counterbalanced Single-Factor Design With Three Treatments Subjects First Treatment Administered Second Treatment Administered Third Treatment Administered S1 1 2 3 S2 1 3 2 S3 2 1 3 S4 2 3 1 S5 3 1 2 S6 3 2 1 Adapted from © 2005 The McGraw-Hill Companies, Inc.. When to Use a Within-Subjects Design A within-subjects design may be best when Subject variables are correlated with the dependent variable It is important to economize on participants or subjects You want to assess the effects of increasing exposure on behavior Adapted from © 2005 The McGraw-Hill Companies, Inc.. Factorial Designs Adding a second independent variable to a singlefactor design results in a FACTORIAL DESIGN Two components can be assessed The MAIN EFFECT of each independent variable The separate effect of each independent variable Analogous to separate experiments involving those variables The INTERACTION between independent variables When the effect of one independent variable changes over levels of a second Adapted from © 2005 The McGraw-Hill Companies, Inc.. Example of An Interaction Value of the Dependent Variable Level 1 Level 2 12 10 8 6 4 2 0 Level 1 Level 2 Level of Independent Variable A Adapted from © 2005 The McGraw-Hill Companies, Inc.. Higher-Order Factorial Designs More than two independent variables are included in a higher-order factorial design As factors are added, the complexity of the experimental design increases The number of possible main effects and interactions increases The number of subjects required increases The volume of materials and amount of time needed to complete the experiment increases Adapted from © 2005 The McGraw-Hill Companies, Inc..