Mod 5 - Graphing homework
2
2
Module 4 – Mixed ANOVA
Martha Ramsey
Saint Leo University
Research Methods II: PSY 535
Instructor Keith Burton
July 31, 2022
GLM male_smiles female_smiles BY gender_participant
/WSFACTOR=Smiles 2 Polynomial
/METHOD=SSTYPE(3)
/POSTHOC=gender_participant(TUKEY)
/PLOT=PROFILE(Smiles*gender_participant) TYPE=LINE ERRORBAR=NO MEANREFERENCE=NO YAXIS=AUTO
/EMMEANS=TABLES(Smiles) COMPARE ADJ(LSD)
/EMMEANS=TABLES(gender_participant*Smiles)
/PRINT=DESCRIPTIVE ETASQ OPOWER HOMOGENEITY
/CRITERIA=ALPHA(.05)
/WSDESIGN=Smiles
/DESIGN=gender_participant.
General Linear Model
[DataSet0] C:\Users\HP 840 G3\Downloads\Smile data.sav
Warnings
Post hoc tests are not performed for gender_participant because there are fewer than three groups.
Within-Subjects Factors
Measure: MEASURE_1
Smiles
Dependent Variable
1
male_smiles
2
female_smiles
Between-Subjects Factors
N
gender_participant
f
26
m
23
Descriptive Statistics
gender_participant
Mean
Std. Deviation
N
male_smiles
f
8.6923
2.63468
26
m
9.9565
2.82003
23
Total
9.2857
2.76887
49
female_smiles
f
6.2308
2.77572
26
m
5.2174
1.99901
23
Total
5.7551
2.47092
49
Box's Test of Equality of Covariance Matrices a
Box's M
10.215
F
3.247
df1
3
df2
818359.675
Sig.
.021
Tests the null hypothesis that the observed covariance matrices of the dependent variables are equal across groups.a
a. Design: Intercept + gender_participant
Within Subjects Design: Smiles
Multivariate Tests a
Effect
Value
F
Hypothesis df
Error df
Smiles
Pillai's Trace
.448
38.183b
1.000
47.000
Wilks' Lambda
.552
38.183b
1.000
47.000
Hotelling's Trace
.812
38.183b
1.000
47.000
Roy's Largest Root
.812
38.183b
1.000
47.000
Smiles * gender_participant
Pillai's Trace
.075
3.820b
1.000
47.000
Wilks' Lambda
.925
3.820b
1.000
47.000
Hotelling's Trace
.081
3.820b
1.000
47.000
Roy's Largest Root
.081
3.820b
1.000
47.000
Multivariate Tests a
Effect
Sig.
Partial Eta Squared
Noncent. Parameter
Smiles
Pillai's Trace
.000
.448
38.183
Wilks' Lambda
.000
.448
38.183
Hotelling's Trace
.000
.448
38.183
Roy's Largest Root
.000
.448
38.183
Smiles * gender_participant
Pillai's Trace
.057
.075
3.820
Wilks' Lambda
.057
.075
3.820
Hotelling's Trace
.057
.075
3.820
Roy's Largest Root
.057
.075
3.820
Multivariate Tests a
Effect
Observed Powerc
Smiles
Pillai's Trace
1.000
Wilks' Lambda
1.000
Hotelling's Trace
1.000
Roy's Largest Root
1.000
Smiles * gender_participant
Pillai's Trace
.482
Wilks' Lambda
.482
Hotelling's Trace
.482
Roy's Largest Root
.482
a. Design: Intercept + gender_participant
Within Subjects Design: Smiles
b. Exact statistic
c. Computed using alpha = .05
Mauchly's Test of Sphericity a
Measure: MEASURE_1
Within Subjects Effect
Mauchly's W
Approx. Chi-Square
df
Sig.
Epsilonb
Greenhouse-Geisser
Smiles
1.000
.000
0
.
1.000
Mauchly's Test of Sphericity a
Measure: MEASURE_1
Within Subjects Effect
Epsilon
Huynh-Feldt
Lower-bound
Smiles
1.000
1.000
Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed dependent variables is proportional to an identity matrix.a
a. Design: Intercept + gender_participant
Within Subjects Design: Smiles
b. May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in the Tests of Within-Subjects Effects table.
Tests of Within-Subjects Effects
Measure: MEASURE_1
Source
Type III Sum of Squares
df
Mean Square
Smiles
Sphericity Assumed
316.389
1
316.389
Greenhouse-Geisser
316.389
1.000
316.389
Huynh-Feldt
316.389
1.000
316.389
Lower-bound
316.389
1.000
316.389
Smiles * gender_participant
Sphericity Assumed
31.654
1
31.654
Greenhouse-Geisser
31.654
1.000
31.654
Huynh-Feldt
31.654
1.000
31.654
Lower-bound
31.654
1.000
31.654
Error(Smiles)
Sphericity Assumed
389.448
47
8.286
Greenhouse-Geisser
389.448
47.000
8.286
Huynh-Feldt
389.448
47.000
8.286
Lower-bound
389.448
47.000
8.286
Tests of Within-Subjects Effects
Measure: MEASURE_1
Source
F
Sig.
Partial Eta Squared
Smiles
Sphericity Assumed
38.183
.000
.448
Greenhouse-Geisser
38.183
.000
.448
Huynh-Feldt
38.183
.000
.448
Lower-bound
38.183
.000
.448
Smiles * gender_participant
Sphericity Assumed
3.820
.057
.075
Greenhouse-Geisser
3.820
.057
.075
Huynh-Feldt
3.820
.057
.075
Lower-bound
3.820
.057
.075
Error(Smiles)
Sphericity Assumed
Greenhouse-Geisser
Huynh-Feldt
Lower-bound
Tests of Within-Subjects Effects
Measure: MEASURE_1
Source
Noncent. Parameter
Observed Powera
Smiles
Sphericity Assumed
38.183
1.000
Greenhouse-Geisser
38.183
1.000
Huynh-Feldt
38.183
1.000
Lower-bound
38.183
1.000
Smiles * gender_participant
Sphericity Assumed
3.820
.482
Greenhouse-Geisser
3.820
.482
Huynh-Feldt
3.820
.482
Lower-bound
3.820
.482
Error(Smiles)
Sphericity Assumed
Greenhouse-Geisser
Huynh-Feldt
Lower-bound
a. Computed using alpha = .05
Tests of Within-Subjects Contrasts
Measure: MEASURE_1
Source
Smiles
Type III Sum of Squares
df
Mean Square
F
Sig.
Smiles
Linear
316.389
1
316.389
38.183
.000
Smiles * gender_participant
Linear
31.654
1
31.654
3.820
.057
Error(Smiles)
Linear
389.448
47
8.286
Tests of Within-Subjects Contrasts
Measure: MEASURE_1
Source
Smiles
Partial Eta Squared
Noncent. Parameter
Observed Powera
Smiles
Linear
.448
38.183
1.000
Smiles * gender_participant
Linear
.075
3.820
.482
Error(Smiles)
Linear
a. Computed using alpha = .05
Levene's Test of Equality of Error Variances a
Levene Statistic
df1
df2
Sig.
male_smiles
Based on Mean
.274
1
47
.603
Based on Median
.257
1
47
.614
Based on Median and with adjusted df
.257
1
46.984
.614
Based on trimmed mean
.264
1
47
.610
female_smiles
Based on Mean
1.113
1
47
.297
Based on Median
.525
1
47
.472
Based on Median and with adjusted df
.525
1
37.991
.473
Based on trimmed mean
.874
1
47
.355
Tests the null hypothesis that the error variance of the dependent variable is equal across groups.a
a. Design: Intercept + gender_participant
Within Subjects Design: Smiles
Tests of Between-Subjects Effects
Measure: MEASURE_1
Transformed Variable: Average
Source
Type III Sum of Squares
df
Mean Square
F
Sig.
Intercept
5527.404
1
5527.404
1084.369
.000
gender_participant
.384
1
.384
.075
.785
Error
239.575
47
5.097
Tests of Between-Subjects Effects
Measure: MEASURE_1
Transformed Variable: Average
Source
Partial Eta Squared
Noncent. Parameter
Observed Powera
Intercept
.958
1084.369
1.000
gender_participant
.002
.075
.058
Error
a. Computed using alpha = .05
Estimated Marginal Means
1. Smiles
Estimates
Measure: MEASURE_1
Smiles
Mean
Std. Error
95% Confidence Interval
Lower Bound
Upper Bound
1
9.324
.390
8.540
10.108
2
5.724
.350
5.021
6.428
Pairwise Comparisons
Measure: MEASURE_1
(I) Smiles
(J) Smiles
Mean Difference (I-J)
Std. Error
Sig.b
95% Confidence Interval for Differenceb
Lower Bound
Upper Bound
1
2
3.600*
.583
.000
2.428
4.772
2
1
-3.600*
.583
.000
-4.772
-2.428
Based on estimated marginal means
*. The mean difference is significant at the .05 level.
b. Adjustment for multiple comparisons: Least Significant Difference (equivalent to no adjustments).
Multivariate Tests
Value
F
Hypothesis df
Error df
Sig.
Partial Eta Squared
Pillai's trace
.448
38.183a
1.000
47.000
.000
.448
Wilks' lambda
.552
38.183a
1.000
47.000
.000
.448
Hotelling's trace
.812
38.183a
1.000
47.000
.000
.448
Roy's largest root
.812
38.183a
1.000
47.000
.000
.448
Multivariate Tests
Noncent. Parameter
Observed Powerb
Pillai's trace
38.183
1.000
Wilks' lambda
38.183
1.000
Hotelling's trace
38.183
1.000
Roy's largest root
38.183
1.000
Each F tests the multivariate effect of Smiles. These tests are based on the linearly independent pairwise comparisons among the estimated marginal means.
a. Exact statistic
b. Computed using alpha = .05
2. gender_participant * Smiles
Measure: MEASURE_1
gender_participant
Smiles
Mean
Std. Error
95% Confidence Interval
Lower Bound
Upper Bound
f
1
8.692
.534
7.618
9.767
2
6.231
.479
5.267
7.195
m
1
9.957
.568
8.814
11.099
2
5.217
.509
4.193
6.242
Profile Plots