@ 2012 Wadsworth, Cengage Learning Chapter 10 Extending the Logic of Experimentation: Within-Subjects and Matched-Subjects 2012 Wadsworth,

Slides:



Advertisements
Similar presentations
Chapter 9 Choosing the Right Research Design Chapter 9.
Advertisements

CHAPTER 8 EXPERIMENTAL DESIGN.
Questions What is the relationship between ‘research designs’ and ‘research strategies’? Which method of experiments, within subjects or between subjects.
Slides to accompany Weathington, Cunningham & Pittenger (2010), Chapter 14: Correlated Groups Designs 1.
Randomized Experimental Design
Control Any means used to rule out threats to validity Example –Hypothesis: Rats learned to press a bar when a light was turned on. –Data for 10 rats bar.
Design of Experiments and Analysis of Variance
Selection of Research Participants: Sampling Procedures
Chapter 9: Within Designs
EXPERIMENTAL DESIGNS Criteria for Experiments
CHAPTER 8 EXPERIMENTAL DESIGN.
The Psychologist as Detective, 4e by Smith/Davis © 2007 Pearson Education Chapter Twelve: Designing, Conducting, Analyzing, and Interpreting Experiments.
Experimental Design: Between and Within factors Psych 231: Research Methods in Psychology.
Using Between-Subjects and Within-Subjects Experimental Designs
PSYC512: Research Methods PSYC512: Research Methods Lecture 11 Brian P. Dyre University of Idaho.
Experimental Designs Psych 231: Research Methods in Psychology.
Basic Logic of Experimentation The design of an Internally valid experimental procedure requires us to: Form Equivalent Groups Treat Groups Identically.
Experimental Design: Between and within factors Psych 231: Research Methods in Psychology.
Experimental Design: Between and Within factors Psych 231: Research Methods in Psychology.
PSYC512: Research Methods PSYC512: Research Methods Lecture 12 Brian P. Dyre University of Idaho.
© 2001 Dr. Laura Snodgrass, Ph.D.1 Basic Experimental Design Common Problems Assigning Participants to Groups Single variable experiments –bivalent –multivalent.
Factorial Experiments Factorial Design = experiment in which more than one IV (factor) at a time is manipulated Uses all possible combinations of the levels.
Experimental Design: Between and within factors Psych 231: Research Methods in Psychology.
Chapter 9 Experimental Research Gay, Mills, and Airasian
McGraw-Hill © 2006 The McGraw-Hill Companies, Inc. All rights reserved. Experimental Research Chapter Thirteen.
Experimental Research
Smith/Davis (c) 2005 Prentice Hall Chapter Ten Designing and Conducting, Experiments with Two Groups PowerPoint Presentation created by Dr. Susan R. Burns.
Chapter 8 Experimental Research
Chapter 8 Experimental Design.
BHS Methods in Behavioral Sciences I May 19, 2003 Chapter 9 (Ray) Within-Subjects Designs (Cont.), Matched-Subjects Procedures.
Research Methods in Psychology
Matched Pairs, Within-Subjects, and Mixed Designs
Chapter 8 Experimental Design: Dependent Groups and Mixed Groups Designs.
Which Test Do I Use? Statistics for Two Group Experiments The Chi Square Test The t Test Analyzing Multiple Groups and Factorial Experiments Analysis of.
Single-Factor Experimental Designs
@ 2012 Wadsworth, Cengage Learning Chapter 9 Applying the Logic of Experimentation: Between-Subjects 2012 Wadsworth, Cengage Learning.
Psychology 301 Chapters & Differences Between Two Means Introduction to Analysis of Variance Multiple Comparisons.
1 Experimental Design. 2  Single Factor - One treatment with several levels.  Multiple Factors - More than one treatment with several levels each. 
A Within-Subjects Experiment: Homophone Priming of Proper Names
©2010 John Wiley and Sons Chapter 3 Research Methods in Human-Computer Interaction Chapter 3- Experimental Design.
Introduction section of article
McGraw-Hill/Irwin Copyright © 2011 by The McGraw-Hill Companies, Inc. All rights reserved. Using Between-Subjects and Within- Subjects Experimental Designs.
The Psychologist as Detective, 4e by Smith/Davis © 2007 Pearson Education Chapter Six: The Basics of Experimentation I: Variables and Control Chapter Six:
Copyright © 2011 by The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill/Irwin Using Specialized Research Designs.
Chapter Six: The Basics of Experimentation I: Variables and Control.
 Descriptive Methods ◦ Observation ◦ Survey Research  Experimental Methods ◦ Independent Groups Designs ◦ Repeated Measures Designs ◦ Complex Designs.
Choosing the right research design
Smith/Davis (c) 2005 Prentice Hall Chapter Fifteen Inferential Tests of Significance III: Analyzing and Interpreting Experiments with Multiple Independent.
Chapter Eight: Quantitative Methods
ITEC6310 Research Methods in Information Technology Instructor: Prof. Z. Yang Course Website: /itec6310.htm Office:
Experiment Basics: Designs Psych 231: Research Methods in Psychology.
ANOVA Overview of Major Designs. Between or Within Subjects Between-subjects (completely randomized) designs –Subjects are nested within treatment conditions.
HYPOTHESIS TESTING FOR DIFFERENCES BETWEEN MEANS AND BETWEEN PROPORTIONS.
Experiment Basics: Designs Psych 231: Research Methods in Psychology.
8 Experimental Research Design.
7 Control Techniques in Experimental Research.
METHODS IN BEHAVIORAL RESEARCH
Research Methods: Concepts and Connections First Edition
Experimental Research Designs
Experimental Design-Chapter 8
Between-Subjects, within-subjects, and factorial Experimental Designs
Within- Subjects Design
BHS Methods in Behavioral Sciences I
Between-Subjects Experimental Designs
Chapter 8 Experimental Design.
Latin Square Designs KNNL – Sections
Single-Variable, Correlated-Groups Designs
Correlated-Groups and Single-Subject Designs
Repeated Measures Balancing Practice Effects with an Incomplete Design
CHAPTER 8 EXPERIMENTAL DESIGN © 2012 The McGraw-Hill Companies, Inc.
Presentation transcript:

@ 2012 Wadsworth, Cengage Learning Chapter 10 Extending the Logic of Experimentation: Within-Subjects and Matched-Subjects 2012 Wadsworth, Cengage Learning

Topics 1.Within-Subjects Designs 2.Mixed Designs 3.Matched-Subjects Designs

@ 2012 Wadsworth, Cengage Learning Within-Subjects Designs

@ 2012 Wadsworth, Cengage Learning Within-Subjects Designs The participant’s own performance is the basis of comparison Compare the performances of the same set of participants on the dependent variable following different treatments Let’s look at an experiment – First as a between-subjects experiment – Then as a within-subjects experiment

@ 2012 Wadsworth, Cengage Learning Within-Subjects Designs (cont’d.) If we identify the individual participants, the design is:

@ 2012 Wadsworth, Cengage Learning Within-Subjects Designs (cont’d.) We can perform this same study as a within-subjects design

@ 2012 Wadsworth, Cengage Learning Table 10.1 Analysis of Variance F Table for the Basketball Study as a Between-Subjects Design and as a Within-Subjects Design

@ 2012 Wadsworth, Cengage Learning Within-Subjects Designs (cont’d.) Potential problem in basketball study – The results from the no-feedback condition might have a potential carryover effect on the feed-back condition To control for this potential problem: – Use a counterbalancing procedure Intrasubject counterbalancing

@ 2012 Wadsworth, Cengage Learning An Illustration of Within-Subjects Research Interaction effect – Crashes were caused by a visual illusion but only when the pilots were distracted from instruments Perform a simple 2 X 3 X 2 design

@ 2012 Wadsworth, Cengage Learning An Illustration of Within-Subjects Research (cont’d.) Dependent variable: amount of landing error Using a between-subjects design – Randomly assign pilots to one of the four possible conditions (A 1 B 1, A 1 B 2, A 2 B 1, A 2 B 2 ) Is a between-subjects design the best alternative? Use a within-subjects design – Each cell is composed of the same participants

@ 2012 Wadsworth, Cengage Learning Within-Subjects Designs Advantages of within-subjects designs: – Ensures that all groups are equal on every factor at the beginning of the experiment – The total number of research participants can be reduced dramatically – Statistically more sensitive to changes in the treatment effect

@ 2012 Wadsworth, Cengage Learning Within-Subjects Designs (cont’d.) Disadvantages of within-subjects design – Within-subjects design is not appropriate when: The treatment has a lasting effect or The purpose of the study is to test for a lasting effect – Extremely sensitive to time-related effects Order effects: effects brought about through continued repetition of the tasks – Fatigue effects: a decline in performance – Practice effects: an improvement in performance

@ 2012 Wadsworth, Cengage Learning Counterbalancing Conditions for complete counterbalancing: – Each condition must occur equally often – Each condition must precede and follow all other conditions an equal number of times Intragroup counterbalancing: ensures that every possible sequence appears at each presentation of the treatment

@ 2012 Wadsworth, Cengage Learning Counterbalancing (cont’d.) Latin square design – Special case of incomplete counterbalancing – Common use: psychopharmacological research Counterbalancing will not control for a differential order effect

@ 2012 Wadsworth, Cengage Learning Repeated Measures One common application of within-subjects designs – Repeated measures as one factor in a factorial design – Useful in studying psychological processes that occur over time – Widely used in studying human and animal learning processes

@ 2012 Wadsworth, Cengage Learning Mixed Designs

@ 2012 Wadsworth, Cengage Learning Mixed Designs Designs that include both “within” and “between” components Example: – Assume that we want to perform a biofeedback experiment to determine the effect of feedback on our ability to control heart rate

@ 2012 Wadsworth, Cengage Learning Mixed Designs (cont’d.) Such an experiment is diagrammed as:

@ 2012 Wadsworth, Cengage Learning Mixed Designs (cont’d.) Figure 10.1

@ 2012 Wadsworth, Cengage Learning Matched-Subjects Designs

@ 2012 Wadsworth, Cengage Learning Matched Subjects Design By using a matched-subjects design, we: – Reap some of the advantages of within-subjects designs and – Take advantage of the random assignment of participants that is possible with a between- subjects design Matching can be used in either of two ways: – As a control procedure – As an experimental procedure

@ 2012 Wadsworth, Cengage Learning Matching as a Control Procedure When a particular individual variable or characteristic has a high correlation with the dependent variable – Equal groups may be obtained by matching along this characteristic

@ 2012 Wadsworth, Cengage Learning Matching as a Control Procedure (cont’d.) Two steps in forming matched groups of participants: – Pairs of participants are matched on some measure that is correlated with performance on the dependent variable – One member of each pair is assigned randomly to either the experimental or the control group; the other member is then assigned to the other group

@ 2012 Wadsworth, Cengage Learning Figure 10.2 The six major steps in a matched-subjects design used to study changes in brain chemistry resulting from prolonged sensory deprivation. (1) Rank order all subjects on the aspect of blood chemistry that is known to be correlated with brain neurochemistry. (2) Form pairs of subjects on the basis of this rank order. (3) Randomly assign one member of each pair to the experimental group and one member to the control group. (4) Conduct the experimental treatments. (5) Conduct neurochemical analyses. (6) Compare the results for the experimental and control animals. Note that, except for our ranking and matching procedures, the design is similar to the completely randomized designs discussed previously

@ 2012 Wadsworth, Cengage Learning Matching as an Experimental Procedure Randomized block design: resulting procedure when the matching factor is analyzed Time estimation study factorial design

@ 2012 Wadsworth, Cengage Learning Matching as an Experimental Procedure (cont’d.) Advantages resulting from the prior matching of participants to groups – Knowledge gained from the interaction effect – Matching ensures that the groups in a study are equal before the treatment is introduced – Matching participants reduces the within-groups variance (error variance)

@ 2012 Wadsworth, Cengage Learning Terminology Glass and Stanley (1970) suggest: – Reserve the term blocking for cases in which the matching takes place on a nominal-scale factor – The use of twins be considered matching on a nominal-scale factor – When the matching uses ordinal measurements, use the term stratifying – When interval or ratio scales are used, use the term leveling

@ 2012 Wadsworth, Cengage Learning Summary Within-subjects designs: every participant serves in every group and receives all levels of the independent variable Mixed design: combines both “within” and “between” designs Matched-subjects procedure may be used either as a control procedure or as an experimental procedure