RECODE age1 (Lowest thru 48=1) (49 thru 62=2) (63 thru Highest=3) INTO agegrp3.
EXECUTE.
UNIANOVA wm1_1 BY agegrp3 sex1
/METHOD=SSTYPE(3)
/INTERCEPT=INCLUDE
/PLOT=PROFILE(sex1*agegrp3) TYPE=LINE ERRORBAR=NO MEANREFERENCE=NO
YAXIS=AUTO
/PRINT ETASQ DESCRIPTIVE HOMOGENEITY OPOWER
/CRITERIA=ALPHA(.05)
/DESIGN=agegrp3 sex1 agegrp3*sex1.
Univariate Analysis of Variance
[DataSet1] /Users/emilydang/Downloads/Assign4.sav
Between-Subjects Factors
agegrp3
Gender
Value Label
0-48
49-62
63+
female
male
1.00
2.00
3.00
f
m
N
134
136
119
230
159
Descriptive Statistics
Dependent Variable:
agegrp3
Gender
0-48
female
male
Total
49-62
female
male
Total
63+
female
male
Total
Working Memory (1 back)
Mean
Std. Deviation
9.117
1.8565
9.386
1.4111
9.231
1.6810
8.356
2.6213
8.825
2.4662
8.574
2.5521
6.538
3.3297
7.179
3.1024
6.748
3.2579
N
77
57
134
73
63
136
80
39
119
Total
female
male
Total
7.978
8.623
8.242
2.8872
2.4795
2.7430
230
159
389
Levene's Test of Equality of Error Variancesa,b
Levene
Statistic
17.489
13.890
13.890
df1
df2
Working Memory (1 back) Based on Mean
5
383
Based on Median
5
383
Based on Median and with
5
318.279
adjusted df
Based on trimmed mean
17.475
5
383
Tests the null hypothesis that the error variance of the dependent variable is equal across groups.
a. Dependent variable: Working Memory (1 back)
b. Design: Intercept + agegrp3 + sex1 + agegrp3 * sex1
Sig.
.000
.000
.000
.000
Tests of Between-Subjects Effects
Dependent Variable:
Working Memory (1 back)
Type III Sum of
Source
Squares
df
Mean Square
F
Sig.
Partial Eta
Noncent.
Observed
Squared
Parameter
Powerb
432.378a
5
86.476
13.318
.000
.148
66.589
1.000
Intercept
24841.306
1
24841.306
3825.724
.000
.909
3825.724
1.000
agegrp3
347.513
2
173.756
26.760
.000
.123
53.519
1.000
19.393
1
19.393
2.987
.085
.008
2.987
.407
2.050
2
1.025
.158
.854
.001
.316
.074
Error
2486.907
383
6.493
Total
29342.000
389
2919.285
388
Corrected Model
sex1
agegrp3 * sex1
Corrected Total
a. R Squared = .148 (Adjusted R Squared = .137)
b. Computed using alpha = .05
Profile Plots
UNIANOVA wm1_1 BY agegrp3 sex1
/METHOD=SSTYPE(3)
/INTERCEPT=INCLUDE
/POSTHOC=agegrp3(DUNNETT)
/PLOT=PROFILE(sex1*agegrp3) TYPE=LINE ERRORBAR=NO MEANREFERENCE=NO
YAXIS=AUTO
/PRINT ETASQ DESCRIPTIVE HOMOGENEITY OPOWER
/CRITERIA=ALPHA(.05)
/DESIGN=agegrp3 sex1 agegrp3*sex1.
Univariate Analysis of Variance
Between-Subjects Factors
agegrp3
1.00
2.00
Value Label
0-48
49-62
N
134
136
Gender
3.00
f
m
63+
female
male
119
230
159
Descriptive Statistics
Dependent Variable:
agegrp3
Gender
0-48
female
male
Total
49-62
female
male
Total
63+
female
male
Total
Total
female
male
Total
Working Memory (1 back)
Mean
Std. Deviation
9.117
1.8565
9.386
1.4111
9.231
1.6810
8.356
2.6213
8.825
2.4662
8.574
2.5521
6.538
3.3297
7.179
3.1024
6.748
3.2579
7.978
2.8872
8.623
2.4795
8.242
2.7430
N
77
57
134
73
63
136
80
39
119
230
159
389
Levene's Test of Equality of Error Variancesa,b
Levene
Statistic
17.489
13.890
13.890
df1
df2
Working Memory (1 back) Based on Mean
5
383
Based on Median
5
383
Based on Median and with
5
318.279
adjusted df
Based on trimmed mean
17.475
5
383
Tests the null hypothesis that the error variance of the dependent variable is equal across groups.
a. Dependent variable: Working Memory (1 back)
b. Design: Intercept + agegrp3 + sex1 + agegrp3 * sex1
Sig.
.000
.000
.000
.000
Tests of Between-Subjects Effects
Dependent Variable:
Working Memory (1 back)
Type III Sum of
Source
Squares
df
Mean Square
F
Sig.
Partial Eta
Noncent.
Observed
Squared
Parameter
Powerb
432.378a
5
86.476
13.318
.000
.148
66.589
1.000
Intercept
24841.306
1
24841.306
3825.724
.000
.909
3825.724
1.000
agegrp3
347.513
2
173.756
26.760
.000
.123
53.519
1.000
Corrected Model
sex1
19.393
1
19.393
2.987
.085
.008
2.987
.407
2.050
2
1.025
.158
.854
.001
.316
.074
Error
2486.907
383
6.493
Total
29342.000
389
2919.285
388
agegrp3 * sex1
Corrected Total
a. R Squared = .148 (Adjusted R Squared = .137)
b. Computed using alpha = .05
Post Hoc Tests
agegrp3
Multiple Comparisons
Dependent Variable:
Dunnett t (2-sided)a
Working Memory (1 back)
Mean
Difference (I-J)
2.483*
1.826*
95% Confidence Interval
Sig.
Lower Bound
Upper Bound
.000
1.772
3.195
.000
1.117
2.535
(I) agegrp3
(J) agegrp3
Std. Error
0-48
63+
.3210
49-62
63+
.3199
Based on observed means.
The error term is Mean Square(Error) = 6.493.
*. The mean difference is significant at the .05 level.
a. Dunnett t-tests treat one group as a control, and compare all other groups against it.
Profile Plots
RECODE bds1_tot (Lowest thru 2=1) (3 thru 5=2) (6 thru Highest=3) INTO depgrp3.
EXECUTE.
USE ALL.
COMPUTE filter_$=(assign4).
VARIABLE LABELS filter_$ 'assign4 (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMATS filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.
UNIANOVA nartot BY agegrp3 depgrp3
/METHOD=SSTYPE(3)
/INTERCEPT=INCLUDE
/PLOT=PROFILE(depgrp3*agegrp3) TYPE=LINE ERRORBAR=NO MEANREFERENCE=NO
YAXIS=AUTO
/PRINT ETASQ DESCRIPTIVE HOMOGENEITY OPOWER
/CRITERIA=ALPHA(.05)
/DESIGN=agegrp3 depgrp3 agegrp3*depgrp3.
Univariate Analysis of Variance
Between-Subjects Factors
agegrp3
depgrp3
Value Label
0-48
49-62
63+
0-2
3-5
6+
1.00
2.00
3.00
1.00
2.00
3.00
N
45
45
45
45
45
45
Descriptive Statistics
Dependent Variable:
agegrp3
depgrp3
0-48
0-2
3-5
6+
Total
49-62
0-2
3-5
6+
Total
63+
0-2
3-5
6+
Total
Total
0-2
3-5
6+
Total
Average correct on narrative passage
Mean
Std. Deviation
N
.7212
.14360
15
.6848
.10793
15
.7212
.11635
15
.7091
.12197
45
.6727
.10757
15
.6121
.14360
15
.6485
.09637
15
.6444
.11761
45
.6364
.11396
15
.6242
.09004
15
.6364
.09719
15
.6323
.09874
45
.6768
.12494
45
.6404
.11782
45
.6687
.10818
45
.6620
.11736
135
Levene's Test of Equality of Error Variancesa,b
Average correct on
narrative passage
Based on Mean
Based on Median
Based on Median and with
adjusted df
Based on trimmed mean
Levene
Statistic
1.181
1.057
1.057
1.282
df1
df2
8
8
8
126
126
116.480
Sig.
.315
.397
.398
8
126
.259
Tests the null hypothesis that the error variance of the dependent variable is equal across groups.
a. Dependent variable: Average correct on narrative passage
b. Design: Intercept + agegrp3 + depgrp3 + agegrp3 * depgrp3
Tests of Between-Subjects Effects
Dependent Variable:
Average correct on narrative passage
Type III Sum
Source
of Squares
df
Mean Square
F
Sig.
Partial Eta
Noncent.
Observed
Squared
Parameter
Powerb
.196a
8
.024
1.870
.070
.106
14.963
.768
Intercept
59.155
1
59.155
4518.385
.000
.973
4518.385
1.000
agegrp3
.153
2
.077
5.854
.004
.085
11.709
.867
depgrp3
.033
2
.016
1.253
.289
.020
2.506
.268
agegrp3 * depgrp3
.010
4
.002
.187
.945
.006
.748
.089
Error
1.650
126
.013
Total
61.000
135
1.845
134
Corrected Model
Corrected Total
a. R Squared = .106 (Adjusted R Squared = .049)
b. Computed using alpha = .05
Profile Plots
UNIANOVA nartot BY agegrp3 depgrp3
/METHOD=SSTYPE(3)
/INTERCEPT=INCLUDE
/POSTHOC=depgrp3(DUNNETT)
/PLOT=PROFILE(depgrp3*agegrp3) TYPE=LINE ERRORBAR=NO MEANREFERENCE=NO
YAXIS=AUTO
/PRINT ETASQ DESCRIPTIVE HOMOGENEITY OPOWER
/CRITERIA=ALPHA(.05)
/DESIGN=agegrp3 depgrp3 agegrp3*depgrp3.
Univariate Analysis of Variance
Between-Subjects Factors
agegrp3
1.00
2.00
Value Label
0-48
49-62
N
45
45
depgrp3
3.00
1.00
2.00
3.00
63+
0-2
3-5
6+
45
45
45
45
Descriptive Statistics
Dependent Variable:
agegrp3
depgrp3
0-48
0-2
3-5
6+
Total
49-62
0-2
3-5
6+
Total
63+
0-2
3-5
6+
Total
Total
0-2
3-5
6+
Total
Average correct on narrative passage
Mean
Std. Deviation
N
.7212
.14360
15
.6848
.10793
15
.7212
.11635
15
.7091
.12197
45
.6727
.10757
15
.6121
.14360
15
.6485
.09637
15
.6444
.11761
45
.6364
.11396
15
.6242
.09004
15
.6364
.09719
15
.6323
.09874
45
.6768
.12494
45
.6404
.11782
45
.6687
.10818
45
.6620
.11736
135
Levene's Test of Equality of Error Variancesa,b
Levene
Statistic
1.181
1.057
1.057
Average correct on
narrative passage
df1
df2
Based on Mean
8
126
Based on Median
8
126
Based on Median and with
8
116.480
adjusted df
Based on trimmed mean
1.282
8
126
Tests the null hypothesis that the error variance of the dependent variable is equal across groups.
a. Dependent variable: Average correct on narrative passage
b. Design: Intercept + agegrp3 + depgrp3 + agegrp3 * depgrp3
Tests of Between-Subjects Effects
Dependent Variable:
Average correct on narrative passage
Sig.
.315
.397
.398
.259
Type III Sum
Source
of Squares
df
Mean Square
F
Sig.
Partial Eta
Noncent.
Observed
Squared
Parameter
Powerb
.196a
8
.024
1.870
.070
.106
14.963
.768
Intercept
59.155
1
59.155
4518.385
.000
.973
4518.385
1.000
agegrp3
.153
2
.077
5.854
.004
.085
11.709
.867
depgrp3
.033
2
.016
1.253
.289
.020
2.506
.268
agegrp3 * depgrp3
.010
4
.002
.187
.945
.006
.748
.089
Error
1.650
126
.013
Total
61.000
135
1.845
134
Corrected Model
Corrected Total
a. R Squared = .106 (Adjusted R Squared = .049)
b. Computed using alpha = .05
Post Hoc Tests
depgrp3
Multiple Comparisons
Dependent Variable:
Dunnett t (2-sided)a
Average correct on narrative passage
Mean
Difference (I-J)
.0081
-.0283
95% Confidence Interval
Sig.
Lower Bound
Upper Bound
.922
-.0459
.0620
.397
-.0822
.0257
(I) depgrp3
(J) depgrp3
Std. Error
0-2
6+
.02412
3-5
6+
.02412
Based on observed means.
The error term is Mean Square(Error) = .013.
a. Dunnett t-tests treat one group as a control, and compare all other groups against it.
Profile Plots
UNIANOVA totalpre BY agegrp3 depgrp3
/METHOD=SSTYPE(3)
/INTERCEPT=INCLUDE
/PLOT=PROFILE(depgrp3*agegrp3) TYPE=LINE ERRORBAR=NO MEANREFERENCE=NO
YAXIS=AUTO
/PRINT ETASQ DESCRIPTIVE HOMOGENEITY OPOWER
/CRITERIA=ALPHA(.05)
/DESIGN=agegrp3 depgrp3 agegrp3*depgrp3.
Univariate Analysis of Variance
Between-Subjects Factors
agegrp3
1.00
2.00
3.00
Value Label
0-48
49-62
63+
N
45
45
45
depgrp3
1.00
2.00
3.00
0-2
3-5
6+
45
45
45
Descriptive Statistics
Dependent Variable:
agegrp3
depgrp3
0-48
0-2
3-5
6+
Total
49-62
0-2
3-5
6+
Total
63+
0-2
3-5
6+
Total
Total
0-2
3-5
6+
Total
Total # of prescription drugs
Mean
Std. Deviation
.3333
.61721
.6667
.89974
.7333
1.03280
.5778
.86573
1.1333
1.30201
1.4667
1.68466
2.0000
1.81265
1.5333
1.61808
.7333
1.16292
2.3333
1.91485
1.5333
.99043
1.5333
1.53149
.7333
1.09545
1.4889
1.67362
1.4222
1.40598
1.2148
1.44235
N
15
15
15
45
15
15
15
45
15
15
15
45
45
45
45
135
Levene's Test of Equality of Error Variancesa,b
Levene
Statistic
3.359
2.180
2.180
Total # of prescription
drugs
df1
df2
Based on Mean
8
126
Based on Median
8
126
Based on Median and with
8
111.566
adjusted df
Based on trimmed mean
3.075
8
126
Tests the null hypothesis that the error variance of the dependent variable is equal across groups.
a. Dependent variable: Total # of prescription drugs
b. Design: Intercept + agegrp3 + depgrp3 + agegrp3 * depgrp3
Tests of Between-Subjects Effects
Dependent Variable:
Total # of prescription drugs
Sig.
.002
.033
.034
.003
Type III Sum of
Source
Squares
df
Mean Square
F
Sig.
Partial Eta
Noncent.
Observed
Squared
Parameter
Powerb
Corrected Model
53.704a
8
6.713
3.758
.001
.193
30.065
.984
Intercept
199.230
1
199.230
111.536
.000
.470
111.536
1.000
agegrp3
27.393
2
13.696
7.668
.001
.109
15.335
.944
depgrp3
15.748
2
7.874
4.408
.014
.065
8.816
.751
agegrp3 * depgrp3
10.563
4
2.641
1.478
.213
.045
5.914
.448
Error
225.067
126
1.786
Total
478.000
135
Corrected Total
278.770
134
a. R Squared = .193 (Adjusted R Squared = .141)
b. Computed using alpha = .05
Profile Plots
UNIANOVA totalpre BY agegrp3 depgrp3
/METHOD=SSTYPE(3)
/INTERCEPT=INCLUDE
/POSTHOC=depgrp3(DUNNETT)
/PLOT=PROFILE(depgrp3*agegrp3) TYPE=LINE ERRORBAR=NO MEANREFERENCE=NO
YAXIS=AUTO
/PRINT ETASQ DESCRIPTIVE HOMOGENEITY OPOWER
/CRITERIA=ALPHA(.05)
/DESIGN=agegrp3 depgrp3 agegrp3*depgrp3.
Univariate Analysis of Variance
Between-Subjects Factors
agegrp3
depgrp3
Value Label
0-48
49-62
63+
0-2
3-5
6+
1.00
2.00
3.00
1.00
2.00
3.00
N
45
45
45
45
45
45
Descriptive Statistics
Dependent Variable:
agegrp3
depgrp3
0-48
0-2
3-5
6+
Total
49-62
0-2
3-5
6+
Total
63+
0-2
3-5
6+
Total
Total
0-2
3-5
Total # of prescription drugs
Mean
Std. Deviation
.3333
.61721
.6667
.89974
.7333
1.03280
.5778
.86573
1.1333
1.30201
1.4667
1.68466
2.0000
1.81265
1.5333
1.61808
.7333
1.16292
2.3333
1.91485
1.5333
.99043
1.5333
1.53149
.7333
1.09545
1.4889
1.67362
N
15
15
15
45
15
15
15
45
15
15
15
45
45
45
6+
Total
1.4222
1.2148
1.40598
1.44235
45
135
Levene's Test of Equality of Error Variancesa,b
Levene
Statistic
3.359
2.180
2.180
Total # of prescription
drugs
df1
df2
Based on Mean
8
126
Based on Median
8
126
Based on Median and with
8
111.566
adjusted df
Based on trimmed mean
3.075
8
126
Tests the null hypothesis that the error variance of the dependent variable is equal across groups.
a. Dependent variable: Total # of prescription drugs
b. Design: Intercept + agegrp3 + depgrp3 + agegrp3 * depgrp3
Sig.
.002
.033
.034
.003
Tests of Between-Subjects Effects
Dependent Variable:
Total # of prescription drugs
Type III Sum of
Source
Squares
df
Mean Square
F
Sig.
Partial Eta
Noncent.
Observed
Squared
Parameter
Powerb
Corrected Model
53.704a
8
6.713
3.758
.001
.193
30.065
.984
Intercept
199.230
1
199.230
111.536
.000
.470
111.536
1.000
agegrp3
27.393
2
13.696
7.668
.001
.109
15.335
.944
depgrp3
15.748
2
7.874
4.408
.014
.065
8.816
.751
agegrp3 * depgrp3
10.563
4
2.641
1.478
.213
.045
5.914
.448
Error
225.067
126
1.786
Total
478.000
135
Corrected Total
278.770
134
a. R Squared = .193 (Adjusted R Squared = .141)
b. Computed using alpha = .05
Post Hoc Tests
depgrp3
Multiple Comparisons
Dependent Variable:
Dunnett t (2-sided)a
Total # of prescription drugs
Mean
Difference (I-J)
-.6889*
.0667
95% Confidence Interval
Sig.
Lower Bound
Upper Bound
.030
-1.3192
-.0586
.960
-.5637
.6970
(I) depgrp3
(J) depgrp3
Std. Error
0-2
6+
.28176
3-5
6+
.28176
Based on observed means.
The error term is Mean Square(Error) = 1.786.
*. The mean difference is significant at the .05 level.
a. Dunnett t-tests treat one group as a control, and compare all other groups against it.
Profile Plots
GET
FILE='/Users/emilydang/Downloads/Lesson_25_Data_File_1.sav'.
DATASET NAME DataSet2 WINDOW=FRONT.