Sei sulla pagina 1di 21

Chapter 10: Experimental Design

PSYC 3F40

Assigning Participants (Ps) to Conditions


Three basic types of experimental designs: randomized groups (between-person) design Ps are assigned randomly to conditions matched subjects design Ps are matched into blocks on the basis of a relevant variable (one that correlates with the DV), then randomly assigned from blocks to conditions repeated measures (within-person) design each P serves in all experimental conditions

Posttest and Pretest-Posttest Designs

Posttest only design dependent variable is measured only after the manipulation of the independent variable

Pretest-posttest design dependent variable is measured twice, both before and after the experimental manipulation (repeated measures design)

Advantages of Pretest-Posttest Designs

Can determine that the experimental conditions did not differ on the dependent variable at the beginning of the experiment Can see how much the independent variable changed behavior from pretest to posttest More powerful than posttest-only designs

Each participant serves as his/her own control Removes error variance due to individual differences

Disadvantage of Pretest-Posttest Designs

Pretest sensitization administering the pretest may lead participants to respond differently to the independent variable than they would had they not been pretested Pretest-posttest designs are not essential. Posttest-only designs are adequate to determine whether the independent variable has an effect on the dependent variable.

One-Way & Factorial Designs

A one-way experimental design involves manipulating only one independent variable (with at least two levels).

A factorial

design involves manipulating two or more independent variables (referred to as factors).

Describing the Size and Structure of Factorial Designs


A 2

x 2 factorial (read 2-by-2) is a design with two independent variables, each with two levels. x 3 factorial has two independent variables, each with three levels. x 2 x 4 factorial has three independent variables, two with two levels, and one with four levels.

A 3

A 2

Assigning Participants to Conditions in a Factorial Design (1)

Same options as with a one-way design:


Randomized groups factorial design Matched groups factorial design Repeated measures (within-person) factorial design

Also: Mixed factorial design participants are randomly assigned to only one level of some independent variable(s) but receive every level of other independent variable(s); also called a between-within design

Main Effects and Interactions

Main effects:

The effect of an independent variable while ignoring the effects of all other independent variables in the design. A factorial design will have as many main effects as there are independent variables. When the effect of one independent variable differs across the levels of another independent variable. E.g., the effect of variable A is different under one level of variable B than it is under another level of variable B.

Interaction:

Graph of an Interaction
35 Dependent Variable 30 25 20 15 10 5 0 Condition A1 Condition A2 Independent Variable A
Condition B1 Condition B2

Variable A had a different effect on participants in Condition B1 than on those in Condition B2.

Higher-Order Designs

Three-way designs examine:

the main effects of three independent variables


three two-way interactions the A X B interaction (ignoring C), the A X C interaction (ignoring B), the B X C interaction (ignoring A). The three-way interaction of A X B X C

The interaction between two variables depends on the level of the another variable. E.g., The A X B interaction is stronger in condition C1 than C2

Example of a 3-way interaction

Ersner,Hershfield, Mikils, Sullivan, & Carstenson (2008, Study 1) examined time perspective and mixed emotions

Perceiving that one has limited time left leads to:


The anticipation of loss A focus on emotionally meaningful goals

These two tendencies may yield complex, mixed emotional experiences (e.g., positive and negative emotions co-occuring)

Does time perspective cause mixed emotional experience? Is this moderated by age?

Example of a 3-way interaction

3 x 2 x 2 Mixed factorial design

3 guided imagery trials (within-person or repeated-measures


factor):

1) Imagine a specific, meaningful location 2) Imagine the location in 2 months 3) Imagine the location in 4 months

2 conditions (between-person factor):


1) Control condition 2) Experimental condition: Imagine the location in 4 months, visiting it for the last time

2 ages (between-person quasi-independent variable):


60 younger (M = 20 years) Ps 60 older (M= 77 years) Ps

DV: Mixed emotions = MINIMUM[Happiness, Sadness]

2-way interaction
Time x Condition interaction, F(2, 115) = 12.33, p < .001
visiting it for the last time

3-way interaction!

3 x 2 x 2 mixed factorial design

3 guided imagery trials 2 between-person conditions (control vs. experimental) 2 ages

Is the 2-way interaction moderated by age?


No: The 3-way interaction was not significant. The manipulation affected young and old participants similarly.

Participant Variables in Factorial Designs


Expericorr

factorial designs include both independent variables (that are manipulated) and participant variables (that are measured).

Independent Variable Level 1 Men Women Level 2

The previous 3-way interaction example was an expericorr design

Uses of Expericorr (or Mixed) Designs


Determine whether effects of the independent variable generalize to participants with particular characteristics Examine interactions (how personal characteristics relate to behavior under different experimental conditions) Reduce error variance by accounting for individual differences among participants

Classifying Participants into Groups in Expericorr Designs

Median-split procedure participants who score below the median on the participant variable are classified as low, and participants scoring above the median are classified as high

Ex: some attachment research

Extreme groups procedure use only participants who score very high or low on the participant variable (such as lowest and highest 25%)

Ex: identifying lonely participants

Problems with classifying participants using median splits


Splitting

participants on a continuous variable with a median split or extreme groups procedure may bias the results by missing effects that are actually present or obtaining effects that are statistical artifacts. of splitting participants into groups, researchers often use multiple regression analyses that allow them to keep the participant variable continuous.

Instead

Problems with classifying participants using median splits


Continuous variability provides useful information. Treating a normally distributed variable as categorical disregards information.

Type 1

Type 1I

Cautions in Interpreting Results from Expericorr Designs


If

the manipulated independent variable affects the dependent variable, we can conclude that the independent variable caused this effect. However, because participant variables are measured rather than manipulated, we cannot infer causation. If a participant variable is involved in an interaction, we say that it moderates participants reactions to the independent variable (rather than causes them).

Potrebbero piacerti anche