10 Experimental Research OneWay Designs What types of

10 Experimental Research: One-Way Designs • What types of evidence allow us to conclude that one variable causes another variable? • How do experimental research designs allow the demonstration of causal relationships between independent and dependent variables? • How is equivalence among the level of the independent variable created in experiments? • How does the ANOVA test hypotheses about differences between the experimental conditions? • What are repeated-measures experimental designs? • How are the results of experimental research designs presented in the research report? • What are the advantages and disadvantages of experimental designs versus correlational research?

Demonstration of Causality Association If there is a causal relationship between IV and DV, there must be a strong correlation between them. Temporal Priority IV must be an antecedent of the DV. Control of Common-causal Variable The influence of common-causal variables that may have produced spurious relationships between IV and DV should be ruled out.

One-Way Experimental Design Example: Violent cartoons increase children’s aggressive behaviors. 1) Define a independent variable and its levels (experimental condition) 2) Create equivalence either through use of different participants (between-participant designs) or through use of the same participants in each of experimental conditions (repeatedmeasure designs = within-participant designs) 3) Assign participants to each level randomly. 4) Select a dependent variable

Analysis of Variance (ANOVA) A statistical procedure that is specially designed to compare the means of the dependent variables across the levels of an experimental research design (the independent variable). Example: Violent cartoons increase children’s aggressive behaviors. IV: Violent Cartoons vs. Nonviolent Cartoons DV: Children’s aggressive behaviors

Hypothesis Testing in Experimental Design Null Hypothesis H 0: Mean. Violent cartoons = Mean. Nonviolent Cartoons Research Hypothesis H 1: Mean. Violent cartoons > Mean. Nonviolent Cartoons There is a specific difference among the conditions so that Mean(violent) is greater than M(nonviolent).

Between-Groups and Within-Groups Variance Estimates Variance A measure of dispersion of the scores on a variable. The ANOVA compares the variance of the means of the dependent variable between the different levels to the variance of individuals on the dependent variable within each of the conditions Between-Group Variance The variance among the condition means Within-Group Variance The variance within the conditions

Data Violent 5 3 4 5 Mean(violent) =4 Nonviolent 2 1 Within-Groups Variance 3 2 1 3 Mean(violent) Total mean =2 =3 Between Group Variance

F value F= Between-groups variance Within-groups variance As the Between-groups variance increases in comparison to the Within-groups variance F increases P value will be less than alpha

ANOVA Summary Table Source DV: Between Aggressive Within Play Total Sum of Squares 14. 40 49. 78 64. 18 df Mean Square 1 14. 40 38 1. 31 F p-value 10. 98 . 002

Presentation of Experiment Results Degree of freedom (# of Within-Groups - 1) Degree of freedom (# of Between-Groups - 1) p value F value There were significant differences on rated aggression across the levels of the cartoon condition, F (1, 38) = 10. 98, p <. 01. Children who viewed the violent cartoons (M = 2. 89) were rated as playing more aggressively than children who had viewed the nonviolent cartoons (M = 1. 52) The mean value of the nonviolent condition The mean value of the violent condition

Between-Participants Designs and Repeated-Measure Designs B-P Design R-M Design Ps Ps Violent Cartoons Aggressive Play Nonviolent Cartoons Violent Cartoons Aggressive Play

Advantages and Disadvantages of Repeated-Measure Designs 1) Increase Statistical Power 2) Economize Participants X 1) The first measure will influence the second measure. (carryover) 2) Participants might become fatigued in the second measure. 3) Participants performance might improve on the task over time through practice.

Counterbalancing Arranging the order in which the conditions of a repeated-measures design are experienced. Latin Square Designs A method of counterbalancing the order of conditions so that each condition appears in each order but also follows equally often after each of the order conditions. P 1 P 2 P 3 S 1 A B C S 2 B C A S 3 C A B

Advantages and Disadvantages of Experiments The experiment design allow us to draw conclusions about causal relationships between the independent and dependent variables. X The experiment design can not manipulate a person’s sex, race, intelligence, family variables, and religious background. X The experiment design will not observe participant’s reaction exactly as they would behave if observed outside of the lab X The experiment design necessarily oversimplify things.