lesson17_SPSS_Output

Log
Log - Log - November 5, 2018

GET
  FILE='C:\MyGithub\N736Fall2017_lesson1415\helpmkh.sav'.
DATASET NAME DataSet1 WINDOW=FRONT.
* Encoding: UTF-8.
* ========================================.
* SPSS SYNTAX
* Lesson 17 on Covariates and Interactions
* i.e. moderators
* ========================================.

* Let's look at correlations between some
* conceptual covariates like age, gender,
* and perceived social support among friends
* and associations with potential outcomes of
* interest: pcs, mcs, cesd and indtot

* overall correlations.

CORRELATIONS
  /VARIABLES=age female pss_fr pcs mcs cesd indtot
  /PRINT=TWOTAIL NOSIG
  /MISSING=PAIRWISE.

Correlations
Correlations - Active Dataset - November 5, 2018

[DataSet1] C:\MyGithub\N736Fall2017_lesson1415\helpmkh.sav

Correlations
Correlations - Correlations - November 5, 2018

CorrelationsCorrelations, table, 1 levels of column headers and 2 levels of row headers, table with 9 columns and 25 rows
		age Age at baseline (in years)	female Gender of respondent	pss_fr Perceived Social Support - friends	pcs SF36 Physical Composite Score - Baseline	mcs SF36 Mental Composite Score - Baseline	cesd CESD total score - Baseline	indtot Inventory of Drug Use Consequences (InDue) total score - Baseline
age Age at baseline (in years)	Pearson Correlation	1	.043	.080	-.229^**	.045	.008	.026
	Sig. (2-tailed)		.358	.088	.000	.343	.859	.575
	N	453	453	453	453	453	453	453
female Gender of respondent	Pearson Correlation	.043	1	.067	-.157^**	-.119^*	.180^**	-.261^**
	Sig. (2-tailed)	.358		.155	.001	.011	.000	.000
	N	453	453	453	453	453	453	453
pss_fr Perceived Social Support - friends	Pearson Correlation	.080	.067	1	.077	.138^**	-.184^**	-.198^**
	Sig. (2-tailed)	.088	.155		.104	.003	.000	.000
	N	453	453	453	453	453	453	453
pcs SF36 Physical Composite Score - Baseline	Pearson Correlation	-.229^**	-.157^**	.077	1	.110^*	-.293^**	-.135^**
	Sig. (2-tailed)	.000	.001	.104		.019	.000	.004
	N	453	453	453	453	453	453	453
mcs SF36 Mental Composite Score - Baseline	Pearson Correlation	.045	-.119^*	.138^**	.110^*	1	-.682^**	-.381^**
	Sig. (2-tailed)	.343	.011	.003	.019		.000	.000
	N	453	453	453	453	453	453	453
cesd CESD total score - Baseline	Pearson Correlation	.008	.180^**	-.184^**	-.293^**	-.682^**	1	.336^**
	Sig. (2-tailed)	.859	.000	.000	.000	.000		.000
	N	453	453	453	453	453	453	453
indtot Inventory of Drug Use Consequences (InDue) total score - Baseline	Pearson Correlation	.026	-.261^**	-.198^**	-.135^**	-.381^**	.336^**	1
	Sig. (2-tailed)	.575	.000	.000	.004	.000	.000
	N	453	453	453	453	453	453	453
**. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).

Log
Log - Log - November 5, 2018

* suppose we also consider gender as a potential
* "moderator" of these correlations
* SPLIT the dataset by gender and rerun the
* correlations - look for differences in the
* correlations by gender.

SORT CASES  BY female.
SPLIT FILE LAYERED BY female.

CORRELATIONS
  /VARIABLES=age female pss_fr pcs mcs cesd indtot
  /PRINT=TWOTAIL NOSIG
  /MISSING=PAIRWISE.

Correlations
Correlations - Correlations - November 5, 2018

CorrelationsCorrelations, table, 1 levels of column headers and 3 levels of row headers, table with 10 columns and 47 rows
	female Gender of respondent		age Age at baseline (in years)	female Gender of respondent	pss_fr Perceived Social Support - friends	pcs SF36 Physical Composite Score - Baseline	mcs SF36 Mental Composite Score - Baseline	cesd CESD total score - Baseline	indtot Inventory of Drug Use Consequences (InDue) total score - Baseline
age Age at baseline (in years)	0 Male	Pearson Correlation	1	.^a	.109^*	-.231^**	.055	-.002	.094
		Sig. (2-tailed)		.	.043	.000	.309	.966	.081
		N	346	346	346	346	346	346	346
	1 Female	Pearson Correlation	1	.^a	-.016	-.202^*	.034	.010	-.103
		Sig. (2-tailed)		.	.866	.037	.727	.921	.293
		N	107	107	107	107	107	107	107
female Gender of respondent	0 Male	Pearson Correlation	.^a	.^a	.^a	.^a	.^a	.^a	.^a
		Sig. (2-tailed)	.		.	.	.	.	.
		N	346	346	346	346	346	346	346
	1 Female	Pearson Correlation	.^a	.^a	.^a	.^a	.^a	.^a	.^a
		Sig. (2-tailed)	.		.	.	.	.	.
		N	107	107	107	107	107	107	107
pss_fr Perceived Social Support - friends	0 Male	Pearson Correlation	.109^*	.^a	1	.058	.144^**	-.199^**	-.150^**
		Sig. (2-tailed)	.043	.		.284	.007	.000	.005
		N	346	346	346	346	346	346	346
	1 Female	Pearson Correlation	-.016	.^a	1	.185	.159	-.200^*	-.275^**
		Sig. (2-tailed)	.866	.		.056	.103	.039	.004
		N	107	107	107	107	107	107	107
pcs SF36 Physical Composite Score - Baseline	0 Male	Pearson Correlation	-.231^**	.^a	.058	1	.046	-.243^**	-.113^*
		Sig. (2-tailed)	.000	.	.284		.394	.000	.036
		N	346	346	346	346	346	346	346
	1 Female	Pearson Correlation	-.202^*	.^a	.185	1	.266^**	-.369^**	-.387^**
		Sig. (2-tailed)	.037	.	.056		.006	.000	.000
		N	107	107	107	107	107	107	107
mcs SF36 Mental Composite Score - Baseline	0 Male	Pearson Correlation	.055	.^a	.144^**	.046	1	-.678^**	-.411^**
		Sig. (2-tailed)	.309	.	.007	.394		.000	.000
		N	346	346	346	346	346	346	346
	1 Female	Pearson Correlation	.034	.^a	.159	.266^**	1	-.674^**	-.491^**
		Sig. (2-tailed)	.727	.	.103	.006		.000	.000
		N	107	107	107	107	107	107	107
cesd CESD total score - Baseline	0 Male	Pearson Correlation	-.002	.^a	-.199^**	-.243^**	-.678^**	1	.370^**
		Sig. (2-tailed)	.966	.	.000	.000	.000		.000
		N	346	346	346	346	346	346	346
	1 Female	Pearson Correlation	.010	.^a	-.200^*	-.369^**	-.674^**	1	.486^**
		Sig. (2-tailed)	.921	.	.039	.000	.000		.000
		N	107	107	107	107	107	107	107
indtot Inventory of Drug Use Consequences (InDue) total score - Baseline	0 Male	Pearson Correlation	.094	.^a	-.150^**	-.113^*	-.411^**	.370^**	1
		Sig. (2-tailed)	.081	.	.005	.036	.000	.000
		N	346	346	346	346	346	346	346
	1 Female	Pearson Correlation	-.103	.^a	-.275^**	-.387^**	-.491^**	.486^**	1
		Sig. (2-tailed)	.293	.	.004	.000	.000	.000
		N	107	107	107	107	107	107	107
*. Correlation is significant at the 0.05 level (2-tailed).
**. Correlation is significant at the 0.01 level (2-tailed).
a. Cannot be computed because at least one of the variables is constant.

Log
Log - Log - November 5, 2018

SPLIT FILE OFF.

* look at the differences in the correlations
* by gender for the association between
* mcs and pcs - mental and physical composite
* scores from the SF36 - the association
* is stronger for females.

* test if there is an interaction effect between
* pcs and gender for mcs as the outcome.
* in SPSS for the linear regression module,
* we need to 1st compute an interaction
* effect between pcs and gender. Since gender is
* already coded 0,1, we can just compute the interaction
* without mean-centering. For continuous predictors,
* we should 1st mean center for effects.

compute pcs_x_female = pcs * female.
execute.

* run a regression model
* put in gender - female
* then put in pcs
* then put in the interaction effect between them
* look at the p-values and changes in r2 for each step.
REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS CI(95) R ANOVA COLLIN TOL CHANGE ZPP
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT mcs
  /METHOD=ENTER female
  /METHOD=ENTER pcs
  /METHOD=ENTER pcs_x_female
  /RESIDUALS HISTOGRAM(ZRESID) NORMPROB(ZRESID).

Regression
Regression - Descriptive Statistics - November 5, 2018

Descriptive StatisticsDescriptive Statistics, table, 1 levels of column headers and 1 levels of row headers, table with 4 columns and 6 rows
	Mean	Std. Deviation	N
mcs SF36 Mental Composite Score - Baseline	31.676678300000000	12.839337310000000	453
female Gender of respondent	.24	.425	453
pcs SF36 Physical Composite Score - Baseline	48.048541550000000	10.784602690000000	453
pcs_x_female	10.6330	19.75841	453

Regression
Regression - Correlations - November 5, 2018

CorrelationsCorrelations, table, 1 levels of column headers and 2 levels of row headers, table with 6 columns and 14 rows
		mcs SF36 Mental Composite Score - Baseline	female Gender of respondent	pcs SF36 Physical Composite Score - Baseline	pcs_x_female
Pearson Correlation	mcs SF36 Mental Composite Score - Baseline	1.000	-.119	.110	-.084
	female Gender of respondent	-.119	1.000	-.157	.969
	pcs SF36 Physical Composite Score - Baseline	.110	-.157	1.000	-.039
	pcs_x_female	-.084	.969	-.039	1.000
Sig. (1-tailed)	mcs SF36 Mental Composite Score - Baseline	.	.006	.009	.037
	female Gender of respondent	.006	.	.000	.000
	pcs SF36 Physical Composite Score - Baseline	.009	.000	.	.203
	pcs_x_female	.037	.000	.203	.
N	mcs SF36 Mental Composite Score - Baseline	453	453	453	453
	female Gender of respondent	453	453	453	453
	pcs SF36 Physical Composite Score - Baseline	453	453	453	453
	pcs_x_female	453	453	453	453

Regression
Regression - Variables Entered/Removed - November 5, 2018

Variables Entered/Removed^aVariables Entered/Removed, table, 1 levels of column headers and 1 levels of row headers, table with 4 columns and 7 rows
Model	Variables Entered	Variables Removed	Method
1	female Gender of respondent^b	.	Enter
2	pcs SF36 Physical Composite Score - Baseline^b	.	Enter
3	pcs_x_female^b	.	Enter
a. Dependent Variable: mcs SF36 Mental Composite Score - Baseline
b. All requested variables entered.

Regression
Regression - Model Summary - November 5, 2018

Model Summary^dModel Summary, table, 2 levels of column headers and 1 levels of row headers, table with 10 columns and 10 rows
Model	R	R Square	Adjusted R Square	Std. Error of the Estimate	Change Statistics
Model	R	R Square	Adjusted R Square	Std. Error of the Estimate	R Square Change	F Change	df1	df2	Sig. F Change
1	.119^a	.014	.012	12.762679000000000	.014	6.446	1	451	.011
2	.151^b	.023	.018	12.720657510000000	.009	3.985	1	450	.047
3	.177^c	.031	.025	12.679310190000000	.009	3.940	1	449	.048
a. Predictors: (Constant), female Gender of respondent
b. Predictors: (Constant), female Gender of respondent, pcs SF36 Physical Composite Score - Baseline
c. Predictors: (Constant), female Gender of respondent, pcs SF36 Physical Composite Score - Baseline, pcs_x_female
d. Dependent Variable: mcs SF36 Mental Composite Score - Baseline

Regression
Regression - ANOVA - November 5, 2018

ANOVA^aANOVA, table, 1 levels of column headers and 2 levels of row headers, table with 7 columns and 15 rows
Model		Sum of Squares	df	Mean Square	F	Sig.
1	Regression	1049.984	1	1049.984	6.446	.011^b
	Residual	73461.575	451	162.886
	Total	74511.559	452
2	Regression	1694.752	2	847.376	5.237	.006^c
	Residual	72816.807	450	161.815
	Total	74511.559	452
3	Regression	2328.116	3	776.039	4.827	.003^d
	Residual	72183.443	449	160.765
	Total	74511.559	452
a. Dependent Variable: mcs SF36 Mental Composite Score - Baseline
b. Predictors: (Constant), female Gender of respondent
c. Predictors: (Constant), female Gender of respondent, pcs SF36 Physical Composite Score - Baseline
d. Predictors: (Constant), female Gender of respondent, pcs SF36 Physical Composite Score - Baseline, pcs_x_female

Regression
Regression - Coefficients - November 5, 2018

Coefficients^aCoefficients, table, 2 levels of column headers and 2 levels of row headers, table with 14 columns and 13 rows
Model		Unstandardized Coefficients		Standardized Coefficients	t	Sig.	95.0% Confidence Interval for B		Correlations			Collinearity Statistics
Model		B	Std. Error	Beta	t	Sig.	Lower Bound	Upper Bound	Zero-order	Partial	Part	Tolerance	VIF
1	(Constant)	32.523	.686		47.401	.000	31.175	33.872
1	female Gender of respondent	-3.584	1.412	-.119	-2.539	.011	-6.359	-.810	-.119	-.119	-.119	1.000	1.000
2	(Constant)	27.031	2.835		9.533	.000	21.458	32.603
	female Gender of respondent	-3.139	1.425	-.104	-2.203	.028	-5.939	-.339	-.119	-.103	-.103	.975	1.025
	pcs SF36 Physical Composite Score - Baseline	.112	.056	.094	1.996	.047	.002	.223	.110	.094	.093	.975	1.025
3	(Constant)	29.849	3.163		9.437	.000	23.633	36.065
	female Gender of respondent	-15.620	6.446	-.517	-2.423	.016	-28.288	-2.951	-.119	-.114	-.113	.047	21.124
	pcs SF36 Physical Composite Score - Baseline	.055	.063	.046	.866	.387	-.069	.178	.110	.041	.040	.769	1.300
	pcs_x_female	.272	.137	.419	1.985	.048	.003	.542	-.084	.093	.092	.048	20.638
a. Dependent Variable: mcs SF36 Mental Composite Score - Baseline

Regression
Regression - Excluded Variables - November 5, 2018

Excluded Variables^aExcluded Variables, table, 2 levels of column headers and 2 levels of row headers, table with 9 columns and 9 rows
Model		Beta In	t	Sig.	Partial Correlation	Collinearity Statistics
Model		Beta In	t	Sig.	Partial Correlation	Tolerance	VIF	Minimum Tolerance
1	pcs SF36 Physical Composite Score - Baseline	.094^b	1.996	.047	.094	.975	1.025	.975
1	pcs_x_female	.503^b	2.684	.008	.126	.061	16.275	.061
2	pcs_x_female	.419^c	1.985	.048	.093	.048	20.638	.047
a. Dependent Variable: mcs SF36 Mental Composite Score - Baseline
b. Predictors in the Model: (Constant), female Gender of respondent
c. Predictors in the Model: (Constant), female Gender of respondent, pcs SF36 Physical Composite Score - Baseline

Regression
Regression - Collinearity Diagnostics - November 5, 2018

Collinearity Diagnostics^aCollinearity Diagnostics, table, 2 levels of column headers and 2 levels of row headers, table with 8 columns and 13 rows
Model	Dimension	Eigenvalue	Condition Index	Variance Proportions
Model	Dimension	Eigenvalue	Condition Index	(Constant)	female Gender of respondent	pcs SF36 Physical Composite Score - Baseline	pcs_x_female
1	1	1.486	1.000	.26	.26
1	2	.514	1.700	.74	.74
2	1	2.307	1.000	.01	.07	.01
	2	.670	1.856	.01	.89	.01
	3	.023	9.999	.99	.05	.98
3	1	2.906	1.000	.00	.00	.00	.00
	2	1.046	1.667	.01	.01	.01	.01
	3	.035	9.085	.25	.25	.25	.28
	4	.013	15.240	.74	.74	.74	.70
a. Dependent Variable: mcs SF36 Mental Composite Score - Baseline

Regression
Regression - Residuals Statistics - November 5, 2018

Residuals Statistics^aResiduals Statistics, table, 1 levels of column headers and 1 levels of row headers, table with 6 columns and 7 rows
	Minimum	Maximum	Mean	Std. Deviation	N
Predicted Value	21.391777040000000	36.831836700000004	31.676678300000000	2.269515227000000	453
Residual	-27.169780730000000	32.866035460000000	.000000000000003	12.637162780000000	453
Std. Predicted Value	-4.532	2.271	.000	1.000	453
Std. Residual	-2.143	2.592	.000	.997	453
a. Dependent Variable: mcs SF36 Mental Composite Score - Baseline

Charts
Charts - *zresid Histogram - November 5, 2018

Charts
Charts - *zresid Normal P-P Plot - November 5, 2018

Log
Log - Log - November 5, 2018

* notice that gender is a significant predictor
* conceptually it is a significant "covariate"
* you can see this from Model 1 (BLOCK 1)
* AFTER adjusting for gender, pcs is a
* significant predictor of mcs - see Model 2
* However, once you add the interaction term
* you'll notice that pcs_x_female is significant
* and the pcs term is no longer significant
* there is a significant interaction between
* pcs and gender - gender is a significant moderator
* of the effect between pcs and mcs.

* let's try this using an ANCOVA approach
* Analysis of Covariance Approach
* set mcs = dependent variable
* set female = fixed factor (because it is categorical)
* set pcs = covariate (because it is continuous)
* note that model defaults to full factorial
* but his does NOT include the interaction effect
* so select custom and build a model
* with the 2 main effects plus the interaction effect
* see /DESIGN= in the SYNTAX below
* in the options be sure to check the box
* for parameter effects to see the coefficients
* similar to the regression output.

UNIANOVA mcs BY female WITH pcs
  /METHOD=SSTYPE(3)
  /INTERCEPT=INCLUDE
  /PRINT=ETASQ DESCRIPTIVE PARAMETER HOMOGENEITY
  /CRITERIA=ALPHA(.05)
  /DESIGN=female pcs female*pcs.

Univariate Analysis of Variance
Univariate Analysis of Variance - Between-Subjects Factors - November 5, 2018

Between-Subjects FactorsBetween-Subjects Factors, table, 1 levels of column headers and 2 levels of row headers, table with 4 columns and 4 rows
		Value Label	N
female Gender of respondent	0	Male	346
female Gender of respondent	1	Female	107

Univariate Analysis of Variance
Univariate Analysis of Variance - Descriptive Statistics - November 5, 2018

Descriptive StatisticsDescriptive Statistics, table, Dependent Variable, mcs SF36 Mental Composite Score - Baseline, 1 layers, 1 levels of column headers and 1 levels of row headers, table with 4 columns and 6 rows

mcs SF36 Mental Composite Score - Baseline

female Gender of respondent

Mean

Std. Deviation

N

0 Male

32.523313200000004

12.870236490000000

346

1 Female

28.938961720000000

12.406154990000001

107

Total

31.676678300000000

12.839337310000000

453

Univariate Analysis of Variance
Univariate Analysis of Variance - Levene's Test of Equality of Error Variances - November 5, 2018

Levene's Test of Equality of Error Variances^aLevene's Test of Equality of Error Variances, table, Dependent Variable, mcs SF36 Mental Composite Score - Baseline, 1 layers, 1 levels of column headers and 0 levels of row headers, table with 4 columns and 6 rows

mcs SF36 Mental Composite Score - Baseline

F

df1

df2

Sig.

4.000

1

451

.046

Tests the null hypothesis that the error variance of the dependent variable is equal across groups.

a. Design: Intercept + female + pcs + female * pcs

Univariate Analysis of Variance
Univariate Analysis of Variance - Tests of Between-Subjects Effects - November 5, 2018

Tests of Between-Subjects EffectsTests of Between-Subjects Effects, table, Dependent Variable, mcs SF36 Mental Composite Score - Baseline, 1 layers, 1 levels of column headers and 1 levels of row headers, table with 7 columns and 12 rows

mcs SF36 Mental Composite Score - Baseline

Source

Type III Sum of Squares

df

Mean Square

F

Sig.

Partial Eta Squared

Corrected Model

2328.116^a

3

776.039

4.827

.003

.031

Intercept

7516.967

1

7516.967

46.758

.000

.094

female

943.910

1

943.910

5.871

.016

.013

pcs

1243.360

1

1243.360

7.734

.006

.017

female * pcs

633.364

1

633.364

3.940

.048

.009

Error

72183.443

449

160.765

Total

529057.172

453

Corrected Total

74511.559

452

a. R Squared = .031 (Adjusted R Squared = .025)

Univariate Analysis of Variance
Univariate Analysis of Variance - Parameter Estimates - November 5, 2018

Parameter EstimatesParameter Estimates, table, Dependent Variable, mcs SF36 Mental Composite Score - Baseline, 1 layers, 2 levels of column headers and 1 levels of row headers, table with 8 columns and 11 rows

mcs SF36 Mental Composite Score - Baseline

Parameter

B

Std. Error

t

Sig.

95% Confidence Interval

Partial Eta Squared

Lower Bound

Upper Bound

Intercept

14.230

5.617

2.533

.012

3.191

25.268

.014

[female=0]

15.620

6.446

2.423

.016

2.951

28.288

.013

[female=1]

0^a

.

pcs

.327

.122

2.683

.008

.087

.566

.016

[female=0] * pcs

-.272

.137

-1.985

.048

-.542

-.003

.009

[female=1] * pcs

0^a

.

a. This parameter is set to zero because it is redundant.

Log
Log - Log - November 5, 2018

* compare the parameter estimates table from the ANCOVA
* to the coefficients tables from the regression
* what do you notice? The intercept and interaction
* coefficients are different - why?
* the default contrast computed in SPSS actually flips
* the coding for males and females

* this has to do with contrast coding in SPSS
* so let's flip the coding for gender
* compute males to set male=1 and female=0.
compute male = female=0.
execute.

* rerun the ANCOVA model.
UNIANOVA mcs BY male WITH pcs
  /METHOD=SSTYPE(3)
  /INTERCEPT=INCLUDE
  /PRINT=ETASQ DESCRIPTIVE PARAMETER HOMOGENEITY
  /CRITERIA=ALPHA(.05)
  /DESIGN=male pcs male*pcs.

Univariate Analysis of Variance
Univariate Analysis of Variance - Between-Subjects Factors - November 5, 2018

Between-Subjects FactorsBetween-Subjects Factors, table, 1 levels of column headers and 2 levels of row headers, table with 3 columns and 4 rows
		N
male	.00	107
male	1.00	346

Univariate Analysis of Variance
Univariate Analysis of Variance - Descriptive Statistics - November 5, 2018

Descriptive StatisticsDescriptive Statistics, table, Dependent Variable, mcs SF36 Mental Composite Score - Baseline, 1 layers, 1 levels of column headers and 1 levels of row headers, table with 4 columns and 6 rows

mcs SF36 Mental Composite Score - Baseline

male

Mean

Std. Deviation

N

.00

28.938961720000000

12.406154990000001

107

1.00

32.523313200000004

12.870236490000000

346

Total

31.676678300000000

12.839337310000000

453

Univariate Analysis of Variance
Univariate Analysis of Variance - Levene's Test of Equality of Error Variances - November 5, 2018

Levene's Test of Equality of Error Variances^aLevene's Test of Equality of Error Variances, table, Dependent Variable, mcs SF36 Mental Composite Score - Baseline, 1 layers, 1 levels of column headers and 0 levels of row headers, table with 4 columns and 6 rows

mcs SF36 Mental Composite Score - Baseline

F

df1

df2

Sig.

4.000

1

451

.046

Tests the null hypothesis that the error variance of the dependent variable is equal across groups.

a. Design: Intercept + male + pcs + male * pcs

Univariate Analysis of Variance
Univariate Analysis of Variance - Tests of Between-Subjects Effects - November 5, 2018

Tests of Between-Subjects EffectsTests of Between-Subjects Effects, table, Dependent Variable, mcs SF36 Mental Composite Score - Baseline, 1 layers, 1 levels of column headers and 1 levels of row headers, table with 7 columns and 12 rows

mcs SF36 Mental Composite Score - Baseline

Source

Type III Sum of Squares

df

Mean Square

F

Sig.

Partial Eta Squared

Corrected Model

2328.116^a

3

776.039

4.827

.003

.031

Intercept

7516.967

1

7516.967

46.758

.000

.094

male

943.910

1

943.910

5.871

.016

.013

pcs

1243.360

1

1243.360

7.734

.006

.017

male * pcs

633.364

1

633.364

3.940

.048

.009

Error

72183.443

449

160.765

Total

529057.172

453

Corrected Total

74511.559

452

a. R Squared = .031 (Adjusted R Squared = .025)

Univariate Analysis of Variance
Univariate Analysis of Variance - Parameter Estimates - November 5, 2018

Parameter EstimatesParameter Estimates, table, Dependent Variable, mcs SF36 Mental Composite Score - Baseline, 1 layers, 2 levels of column headers and 1 levels of row headers, table with 8 columns and 11 rows

mcs SF36 Mental Composite Score - Baseline

Parameter

B

Std. Error

t

Sig.

95% Confidence Interval

Partial Eta Squared

Lower Bound

Upper Bound

Intercept

29.849

3.163

9.437

.000

23.633

36.065

.166

[male=.00]

-15.620

6.446

-2.423

.016

-28.288

-2.951

.013

[male=1.00]

0^a

.

pcs

.055

.063

.866

.387

-.069

.178

.002

[male=.00] * pcs

.272

.137

1.985

.048

.003

.542

.009

[male=1.00] * pcs

0^a

.

a. This parameter is set to zero because it is redundant.

Log
Log - Log - November 5, 2018

* let's make a plot of mcs by pcs by gender.

GRAPH
/SCATTERPLOT(BIVAR)=pcs WITH mcs BY female
/MISSING=LISTWISE.

Graph
Graph - Scatter of mcs pcs by female - November 5, 2018

Log
Log - Log - November 5, 2018

* and let's overlay the fitted lines for each gender
* we can compute these by hand.
* regression line
*     mcs = 29.849 - 15.620(female) + 0.055(pcs) + 0.272(pcs*female)
* when female=0 for MALES
*     mcs = 29.849 + 0.055(pcs)
* when female=1 for FEMALES
*     mcs = [29.849 - 15.620] + [0.055 + 0.272](pcs)
*     mcs = 14.229 + 0.327(pcs).

	Undo - Undoes the last change in the document.
	Edit - Open the Editor tool for tables and charts. Certain editing options are only available when you are connected to an Internet server.
	Save - Creates a new copy of the Web Report with the saved changes.
	Print - Prints the current object when in Object View and all objects in Page View.
	Page View - Switches the Web Report to display all the tables and charts on a single page.
	Object View - Switches the Web Report so that each table or chart is displayed one at a time.

Some of this functionality is only available when connected to an Internet server.
	Create a chart - Create a chart from the selected cells in the table.
	Pivot and Sort - Transpose, sort, and pivot the table.
	Background color - The background color of the selected cells.
	Text Color and Style - Font color, style, and size.
	Number Format - Font color, style, and size.

All of this functionality is only available when connected to an Internet server.
	Chart Size - Change the height and width of the chart
	Background color - The background color of the selected object.
	Border and Line Style - The color and thickness of the line or border.
	Text Color and Style - Font color, style, and size.
	Number Format - Font color, style, and size.
	Axis Properties - Change the scale and display axis titles and ticks.



Lock aspect ratio