variations of anova
DESCRIPTION
Variations of ANOVA. Repeated Measures ANOVA. Used when the research design contains one factor on which participants are measured more than twice (dependent, or within-groups design). Similar to the paired-samples t -test. Computing Repeated Measures ANOVA in SPSS. - PowerPoint PPT PresentationTRANSCRIPT
Repeated Measures ANOVA
• Used when the research design contains one factor on which participants are measured more than twice (dependent, or within-groups design).
• Similar to the paired-samples t-test
Computing Repeated Measures ANOVA in SPSS
• Go to Analyze General Linear Model Repeated Measures
• In the repeated measures define factor(s) window, name the factor and enter the number of levels click Add click Define
• In the Repeated Measures dialog box, click on the first level of your variable and move it to the __?__(1) space in the within-subjects variables window continue to do this for all of the remaining levels of the variable
• Click Options Move factor 1 to the Display Means for window and select Compare Main Effects also select Descriptive Statistics and Estimates of Effect Size.
• Click Continue Click OK
Interpreting the OutputDescriptive Statistics
18.8000 2.42605 15
16.4667 2.66905 15
12.6000 3.77586 15
No Alcohol
Three Beers
Six Beers
Mean Std. Deviation N
Multivariate Testsb
.821 29.715a 2.000 13.000 .000
.179 29.715a 2.000 13.000 .000
4.572 29.715a 2.000 13.000 .000
4.572 29.715a 2.000 13.000 .000
Pillai's Trace
Wilks' Lambda
Hotelling's Trace
Roy's Largest Root
EffectBEER
Value F Hypothesis df Error df Sig.
Exact statistica.
Design: Intercept Within Subjects Design: BEER
b.
The descriptive statistics box provides the mean, standard deviation, and number of participants for each measurement time.
This box is generated because three (or more) columns of measurements are being compared. This only needs to be interpreted when those columns of measurements correspond to separate variables (multivariate designs).
Main Analysis
The row you are interested in is the row which has the name of your variable in it. The between df appear in this row; the within degrees of freedom appear in the error row. F is your test statistic, and Sig is its probability. Partial eta squared is the effect size statistic for the F-ratio.
Tests of Within-Subjects Effects
Measure: MEASURE_1
294.178 2 147.089 47.254 .000 .771
294.178 1.409 208.741 47.254 .000 .771
294.178 1.520 193.523 47.254 .000 .771
294.178 1.000 294.178 47.254 .000 .771
87.156 28 3.113
87.156 19.730 4.417
87.156 21.282 4.095
87.156 14.000 6.225
Sphericity Assumed
Greenhouse-Geisser
Huynh-Feldt
Lower-bound
Sphericity Assumed
Greenhouse-Geisser
Huynh-Feldt
Lower-bound
Sourcebeer
Error(beer)
Type III Sumof Squares df Mean Square F Sig.
Partial EtaSquared
Post Hoc TestsPairwise Comparisons
Measure: MEASURE_1
2.333* .410 .000 1.454 3.213
6.200* .794 .000 4.497 7.903
-2.333* .410 .000 -3.213 -1.454
3.867* .668 .000 2.434 5.300
-6.200* .794 .000 -7.903 -4.497
-3.867* .668 .000 -5.300 -2.434
(J) BEER2
3
1
3
1
2
(I) BEER1
2
3
MeanDifference
(I-J) Std. Error Sig.a
Lower Bound Upper Bound
95% Confidence Interval forDifference
a
Based on estimated marginal means
The mean difference is significant at the .05 level.*.
Adjustment for multiple comparisons: Least Significant Difference (equivalent to noadjustments).
a.
Pairwise Comparisons provide the mean difference between each measurement time and its significance.
Factorial ANOVA
• A special case of ANOVA in which there is more than one independent variable (IV) being explored.
• Because there are multiple IVs, factorial designs have multiple hypotheses which are analyzed by multiple F tests: one for each main effect (IV); and one for each possible interaction between the IVs.
Looking for Main Effects
• Main Effect: the action of a single IV in an experiment
Looking for Interactions
• Interaction: the effect of one IV changes across the levels of another IV
• Higher-Order Interaction: an interaction effect involving more than two IVs
Laying Out a Factorial Design• Design Matrix: a visual representation of the research
design• Hint: If you can’t draw it, you can’t interpret it!
M F A ULow X X X X
Male Mod X X X X X XHigh X X X XLow X X X X X
Female Mod X X X X XHigh X X X X X
X X X X
Describing the Design
• Shorthand Notation: a system that uses numbers to describe the design of a factorial study
Within-Subjects Factorial Designs
• Within-Subjects Factorial Design: a factorial design in which subjects receive all conditions in the experiment
Mixed Designs
• Mixed Design: a factorial design that combines within-subjects and between-subjects factors
Computing Factorial ANOVA in SPSS• Analyze General Linear Model Univariate• Move the independent variables to the Fixed Factor(s) box
Move the dependent variable to the Dependent Variable box
• Click Options highlight the independent variables and the interaction term in the Factor(s) box and move it to the Display Means for box Under Display, check descriptive statistics, homogeneity tests, and estimates of effect size. Note that the significance level is already set at 0.05. Click Continue.
• Click OK.
Interpreting the OutputDescriptive Statistics
Dependent Variable: PLAYTIME
5.0000 1.22474 5
10.0000 1.22474 5
7.5000 2.87711 10
15.0000 3.67423 5
35.0000 3.93700 5
25.0000 11.13553 10
10.0000 5.86894 10
22.5000 13.45982 10
16.2500 11.96871 20
AGE4 years
6 years
Total
4 years
6 years
Total
4 years
6 years
Total
Social ConditionAlone
Parents
Total
Mean Std. Deviation N
The descriptive statistics box provides the means, standard deviations, and Ns for each main effect, as well as all interactions.
Levene's Test of Equality of Error Variancesa
Dependent Variable: playtime
2.469 3 16 .099F df1 df2 Sig.
Tests the null hypothesis that the error variance of thedependent variable is equal across groups.
Design: Intercept+soccond+age+soccond * agea.
Levene’s test is designed to compare the error variance of the dependent variable across groups. We do not want this result to be significant.
Main Analysis
There are three hypotheses being tested here (one for each main effect and one for the interaction). Thus, there are three separate F-tests conducted. The between degrees of freedom, as well as the F-ratio, its significance, and associated effect size, are located on the rows with the variable names. The within degrees of freedom is located with the error term.
Tests of Between-Subjects Effects
Dependent Variable: playtime
2593.750a 3 864.583 108.073 .000 .953
5281.250 1 5281.250 660.156 .000 .976
1531.250 1 1531.250 191.406 .000 .923
781.250 1 781.250 97.656 .000 .859
281.250 1 281.250 35.156 .000 .687
128.000 16 8.000
8003.000 20
2721.750 19
SourceCorrected Model
Intercept
soccond
age
soccond * age
Error
Total
Corrected Total
Type III Sumof Squares df Mean Square F Sig.
Partial EtaSquared
R Squared = .953 (Adjusted R Squared = .944)a.