SPSS_Output_lesson20

Log
Log - Log - November 5, 2017

DATASET ACTIVATE DataSet1.

SAVE OUTFILE='C:\MyGithub\N736Fall2017_lesson20\helpmkh.sav'
  /COMPRESSED.
* Encoding: UTF-8.
* ============================================.
* N736 - LESSON 20
* Poisson Regression and Negative Binomial Regression
*
* Melinda Higgins, PhD
* dated 11/5/2017
*
* working with the helpmkh dataset
* ============================================.

* ============================================.
* look at distribution of d1
* number of times hospitalized for medical problems
* this is a good count variable
* check mean and standard deviation
* SD > mean indicates overdispersion
* ============================================.

FREQUENCIES VARIABLES=d1
  /NTILES=4
  /STATISTICS=STDDEV MINIMUM MAXIMUM MEAN
  /HISTOGRAM
  /ORDER=ANALYSIS.

Frequencies
Frequencies - Statistics - November 5, 2017

StatisticsStatistics, table, d1 How many times hospitalized for medical problems (lifetime), 1 layers, 0 levels of column headers and 2 levels of row headers, table with 3 columns and 11 rows

d1 How many times hospitalized for medical problems (lifetime)

N

Valid

453

Missing

0

Mean

3.06

Std. Deviation

6.188

Minimum

0

Maximum

100

Percentiles

25

1.00

50

2.00

75

3.50

Frequencies
Frequencies - d1 How many times hospitalized for medical problems (lifetime) - November 5, 2017

d1 How many times hospitalized for medical problems (lifetime)d1 How many times hospitalized for medical problems (lifetime), table, 1 levels of column headers and 2 levels of row headers, table with 6 columns and 24 rows
		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	0	92	20.3	20.3	20.3
	1	120	26.5	26.5	46.8
	2	92	20.3	20.3	67.1
	3	36	7.9	7.9	75.1
	4	37	8.2	8.2	83.2
	5	18	4.0	4.0	87.2
	6	12	2.6	2.6	89.8
	7	5	1.1	1.1	90.9
	8	11	2.4	2.4	93.4
	9	2	.4	.4	93.8
	10	11	2.4	2.4	96.2
	12	1	.2	.2	96.5
	13	1	.2	.2	96.7
	14	2	.4	.4	97.1
	15	3	.7	.7	97.8
	17	1	.2	.2	98.0
	20	5	1.1	1.1	99.1
	22	1	.2	.2	99.3
	36	1	.2	.2	99.6
	40	1	.2	.2	99.8
	100	1	.2	.2	100.0
	Total	453	100.0	100.0

Frequencies
Frequencies - Histogram - November 5, 2017

{"style":{"fill":{"g":255,"b":255,"r":255},"outline":{"g":0,"b":0,"r":0,"a":0.0}},"data":[{"id":"dSource","rows":[[0,92],[1,120],[2,92],[3,36],[4,37],[5,18],[6,12],[7,5],[8,11],[9,2],[10,11],[12,1],[13,1],[14,2],[15,3],[17,1],[20,5],[22,1],[36,1],[40,1],[100,1]],"fields":[{"min":0.0,"id":"fVariable","max":100.0,"label":"How many times hospitalized for medical problems (lifetime)","format":{"numericPattern":"###"}},{"min":1.0,"id":"fVariable1","max":120.0,"label":"Frequency"}]},{"id":"summary","source":{"$ref":"dSource"},"output":[{"field":{"id":"count"},"role":"y","parameters":{"summaryFunction":"count"}}],"group":[{"input":{"$ref":"fVariable"},"output":{"id":"Variable","format":{"numericPattern":"###"}}}],"type":"summary"}],"titles":[{"backgroundStyle":{"fill":{"g":0,"b":0,"r":0,"a":0.0},"outline":{"g":0,"b":0,"r":0,"a":0.0}},"content":["Histogram"],"style":{"font":{"weight":"bold","family":"'Helvetica Neue;Helvetica;Arial;SansSerif', sans-serif","size":"12pt"},"fill":{"g":100,"b":157,"r":0},"padding":3.0},"type":"title"}],"copyright":"(C) Copyright IBM Corp. 2011","version":"6.0","size":{"height":500.0,"width":625.0},"grammar":[{"elements":[{"position":[{"field":{"$ref":"count"}},{"field":{"$ref":"Variable"}}],"style":{"width":"100.0%","fill":{"g":178,"b":239,"r":0},"outline":{"g":100,"b":157,"r":0}},"data":{"$ref":"summary"},"type":"interval"}],"coordinates":{"style":{"fill":{"g":255,"b":255,"r":255},"outline":{"g":100,"b":157,"r":0}},"dimensions":[{"axis":[{"title":["Frequency"],"lineStyle":{"stroke":{"width":0.6666667},"fill":{"g":100,"b":157,"r":0}},"tickStyle":{"font":{"weight":"normal","family":"sans-serif","size":"8pt"},"fill":{"g":100,"b":157,"r":0}},"titleStyle":{"font":{"weight":"bold","family":"'Helvetica Neue;Helvetica;Arial;SansSerif', sans-serif","size":"11pt"},"fill":{"g":100,"b":157,"r":0},"padding":6.0},"markStyle":{"stroke":{"width":1.3333334},"fill":{"g":100,"b":157,"r":0}}}],"scale":{"spans":[{"tickDelta":100.0}],"padding":{"right":"5%"}}},{"axis":[{"title":["How many times hospitalized for medical problems (lifetime)"],"lineStyle":{"stroke":{"width":0.6666667},"fill":{"g":100,"b":157,"r":0}},"tickStyle":{"font":{"weight":"normal","family":"sans-serif","size":"8pt"},"fill":{"g":100,"b":157,"r":0}},"titleStyle":{"font":{"weight":"bold","family":"'Helvetica Neue;Helvetica;Arial;SansSerif', sans-serif","size":"11pt"},"fill":{"g":100,"b":157,"r":0},"padding":6.0},"markStyle":{"stroke":{"width":1.3333334},"fill":{"g":100,"b":157,"r":0}}}],"scale":{"spans":[{"tickDelta":30.0}],"padding":{"left":"5%","right":"5%"}}}]}}]}

Log
Log - Log - November 5, 2017

* ============================================.
* look at correlations of d1 with demographics
* and other predictors - we'll focus on pcs
* ============================================.

CORRELATIONS
  /VARIABLES=d1 age female pss_fr pcs mcs indtot sexrisk
  /PRINT=TWOTAIL NOSIG
  /MISSING=PAIRWISE.

Correlations
Correlations - Correlations - November 5, 2017

CorrelationsCorrelations, table, 1 levels of column headers and 2 levels of row headers, table with 10 columns and 28 rows
		d1 How many times hospitalized for medical problems (lifetime)	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	indtot Inventory of Drug Use Consequences (InDue) total score - Baseline	sexrisk Risk Assessment Battery (RAB) sex risk score - Baseline
d1 How many times hospitalized for medical problems (lifetime)	Pearson Correlation	1	.161^**	.038	-.048	-.258^**	-.093^*	.032	.036
	Sig. (2-tailed)		.001	.415	.310	.000	.049	.492	.442
	N	453	453	453	453	453	453	453	453
age Age at baseline (in years)	Pearson Correlation	.161^**	1	.043	.080	-.229^**	.045	.026	-.120^*
	Sig. (2-tailed)	.001		.358	.088	.000	.343	.575	.011
	N	453	453	453	453	453	453	453	453
female Gender of respondent	Pearson Correlation	.038	.043	1	.067	-.157^**	-.119^*	-.261^**	.092
	Sig. (2-tailed)	.415	.358		.155	.001	.011	.000	.052
	N	453	453	453	453	453	453	453	453
pss_fr Perceived Social Support - friends	Pearson Correlation	-.048	.080	.067	1	.077	.138^**	-.198^**	-.113^*
	Sig. (2-tailed)	.310	.088	.155		.104	.003	.000	.016
	N	453	453	453	453	453	453	453	453
pcs SF36 Physical Composite Score - Baseline	Pearson Correlation	-.258^**	-.229^**	-.157^**	.077	1	.110^*	-.135^**	.024
	Sig. (2-tailed)	.000	.000	.001	.104		.019	.004	.612
	N	453	453	453	453	453	453	453	453
mcs SF36 Mental Composite Score - Baseline	Pearson Correlation	-.093^*	.045	-.119^*	.138^**	.110^*	1	-.381^**	-.106^*
	Sig. (2-tailed)	.049	.343	.011	.003	.019		.000	.024
	N	453	453	453	453	453	453	453	453
indtot Inventory of Drug Use Consequences (InDue) total score - Baseline	Pearson Correlation	.032	.026	-.261^**	-.198^**	-.135^**	-.381^**	1	.113^*
	Sig. (2-tailed)	.492	.575	.000	.000	.004	.000		.016
	N	453	453	453	453	453	453	453	453
sexrisk Risk Assessment Battery (RAB) sex risk score - Baseline	Pearson Correlation	.036	-.120^*	.092	-.113^*	.024	-.106^*	.113^*	1
	Sig. (2-tailed)	.442	.011	.052	.016	.612	.024	.016
	N	453	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, 2017

NONPAR CORR
  /VARIABLES=d1 age female pss_fr pcs mcs indtot sexrisk
  /PRINT=BOTH TWOTAIL NOSIG
  /MISSING=PAIRWISE.

Nonparametric Correlations
Nonparametric Correlations - Correlations - November 5, 2017

CorrelationsCorrelations, table, 1 levels of column headers and 3 levels of row headers, table with 11 columns and 52 rows
			d1 How many times hospitalized for medical problems (lifetime)	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	indtot Inventory of Drug Use Consequences (InDue) total score - Baseline	sexrisk Risk Assessment Battery (RAB) sex risk score - Baseline
Kendall's tau_b	d1 How many times hospitalized for medical problems (lifetime)	Correlation Coefficient	1.000	.162^**	.100^*	-.069^*	-.238^**	-.141^**	.124^**	.051
		Sig. (2-tailed)	.	.000	.016	.048	.000	.000	.000	.152
		N	453	453	453	453	453	453	453	453
	age Age at baseline (in years)	Correlation Coefficient	.162^**	1.000	.033	.043	-.139^**	.020	.045	-.077^*
		Sig. (2-tailed)	.000	.	.398	.198	.000	.533	.166	.022
		N	453	453	453	453	453	453	453	453
	female Gender of respondent	Correlation Coefficient	.100^*	.033	1.000	.054	-.138^**	-.099^*	-.216^**	.047
		Sig. (2-tailed)	.016	.398	.	.178	.000	.010	.000	.240
		N	453	453	453	453	453	453	453	453
	pss_fr Perceived Social Support - friends	Correlation Coefficient	-.069^*	.043	.054	1.000	.047	.088^**	-.133^**	-.100^**
		Sig. (2-tailed)	.048	.198	.178	.	.147	.007	.000	.003
		N	453	453	453	453	453	453	453	453
	pcs SF36 Physical Composite Score - Baseline	Correlation Coefficient	-.238^**	-.139^**	-.138^**	.047	1.000	.089^**	-.092^**	.021
		Sig. (2-tailed)	.000	.000	.000	.147	.	.005	.004	.530
		N	453	453	453	453	453	453	453	453
	mcs SF36 Mental Composite Score - Baseline	Correlation Coefficient	-.141^**	.020	-.099^*	.088^**	.089^**	1.000	-.238^**	-.074^*
		Sig. (2-tailed)	.000	.533	.010	.007	.005	.	.000	.024
		N	453	453	453	453	453	453	453	453
	indtot Inventory of Drug Use Consequences (InDue) total score - Baseline	Correlation Coefficient	.124^**	.045	-.216^**	-.133^**	-.092^**	-.238^**	1.000	.088^**
		Sig. (2-tailed)	.000	.166	.000	.000	.004	.000	.	.009
		N	453	453	453	453	453	453	453	453
	sexrisk Risk Assessment Battery (RAB) sex risk score - Baseline	Correlation Coefficient	.051	-.077^*	.047	-.100^**	.021	-.074^*	.088^**	1.000
		Sig. (2-tailed)	.152	.022	.240	.003	.530	.024	.009	.
		N	453	453	453	453	453	453	453	453
Spearman's rho	d1 How many times hospitalized for medical problems (lifetime)	Correlation Coefficient	1.000	.220^**	.114^*	-.095^*	-.327^**	-.199^**	.166^**	.069
		Sig. (2-tailed)	.	.000	.015	.043	.000	.000	.000	.141
		N	453	453	453	453	453	453	453	453
	age Age at baseline (in years)	Correlation Coefficient	.220^**	1.000	.040	.062	-.207^**	.029	.066	-.110^*
		Sig. (2-tailed)	.000	.	.399	.190	.000	.541	.163	.019
		N	453	453	453	453	453	453	453	453
	female Gender of respondent	Correlation Coefficient	.114^*	.040	1.000	.063	-.169^**	-.121^*	-.258^**	.055
		Sig. (2-tailed)	.015	.399	.	.178	.000	.010	.000	.241
		N	453	453	453	453	453	453	453	453
	pss_fr Perceived Social Support - friends	Correlation Coefficient	-.095^*	.062	.063	1.000	.067	.126^**	-.185^**	-.137^**
		Sig. (2-tailed)	.043	.190	.178	.	.157	.007	.000	.004
		N	453	453	453	453	453	453	453	453
	pcs SF36 Physical Composite Score - Baseline	Correlation Coefficient	-.327^**	-.207^**	-.169^**	.067	1.000	.144^**	-.134^**	.031
		Sig. (2-tailed)	.000	.000	.000	.157	.	.002	.004	.514
		N	453	453	453	453	453	453	453	453
	mcs SF36 Mental Composite Score - Baseline	Correlation Coefficient	-.199^**	.029	-.121^*	.126^**	.144^**	1.000	-.343^**	-.104^*
		Sig. (2-tailed)	.000	.541	.010	.007	.002	.	.000	.027
		N	453	453	453	453	453	453	453	453
	indtot Inventory of Drug Use Consequences (InDue) total score - Baseline	Correlation Coefficient	.166^**	.066	-.258^**	-.185^**	-.134^**	-.343^**	1.000	.126^**
		Sig. (2-tailed)	.000	.163	.000	.000	.004	.000	.	.007
		N	453	453	453	453	453	453	453	453
	sexrisk Risk Assessment Battery (RAB) sex risk score - Baseline	Correlation Coefficient	.069	-.110^*	.055	-.137^**	.031	-.104^*	.126^**	1.000
		Sig. (2-tailed)	.141	.019	.241	.004	.514	.027	.007	.
		N	453	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, 2017

* ============================================.
* run Poisson regression - intercept only model
* this is the NULL model - no predictors
* ============================================.

* Generalized Linear Models.
GENLIN d1
  /MODEL INTERCEPT=YES
DISTRIBUTION=POISSON LINK=LOG
  /CRITERIA METHOD=FISHER(1) SCALE=1 COVB=MODEL MAXITERATIONS=100 MAXSTEPHALVING=5
    PCONVERGE=1E-006(ABSOLUTE) SINGULAR=1E-012 ANALYSISTYPE=3(WALD) CILEVEL=95 CITYPE=WALD
    LIKELIHOOD=FULL
  /MISSING CLASSMISSING=EXCLUDE
  /PRINT CPS DESCRIPTIVES MODELINFO FIT SUMMARY SOLUTION (EXPONENTIATED).

Generalized Linear Models
Generalized Linear Models - Model Information - November 5, 2017

Model InformationModel Information, table, 0 levels of column headers and 1 levels of row headers, table with 2 columns and 4 rows
Dependent Variable	d1 How many times hospitalized for medical problems (lifetime)
Probability Distribution	Poisson
Link Function	Log

Generalized Linear Models
Generalized Linear Models - Case Processing Summary - November 5, 2017

Case Processing SummaryCase Processing Summary, table, 1 levels of column headers and 1 levels of row headers, table with 3 columns and 5 rows
	N	Percent
Included	453	100.0%
Excluded	0	0.0%
Total	453	100.0%

Generalized Linear Models
Generalized Linear Models - Continuous Variable Information - November 5, 2017

Continuous Variable InformationContinuous Variable Information, table, 1 levels of column headers and 2 levels of row headers, table with 7 columns and 3 rows
		N	Minimum	Maximum	Mean	Std. Deviation
Dependent Variable	d1 How many times hospitalized for medical problems (lifetime)	453	0	100	3.06	6.188

Generalized Linear Models
Generalized Linear Models - Goodness of Fit - November 5, 2017

Goodness of Fit^aGoodness of Fit, table, 1 levels of column headers and 1 levels of row headers, table with 4 columns and 14 rows
	Value	df	Value/df
Deviance	2261.889	452	5.004
Scaled Deviance	2261.889	452
Pearson Chi-Square	5656.091	452	12.513
Scaled Pearson Chi-Square	5656.091	452
Log Likelihood^b	-1638.126
Akaike's Information Criterion (AIC)	3278.252
Finite Sample Corrected AIC (AICC)	3278.261
Bayesian Information Criterion (BIC)	3282.368
Consistent AIC (CAIC)	3283.368
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept)
a. Information criteria are in smaller-is-better form.
b. The full log likelihood function is displayed and used in computing information criteria.

Generalized Linear Models
Generalized Linear Models - Omnibus Test - November 5, 2017

Omnibus Test^aOmnibus Test, table, 1 levels of column headers and 0 levels of row headers, table with 3 columns and 5 rows
Likelihood Ratio Chi-Square	df	Sig.
.000	.	.
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept)
a. Compares the fitted model against the intercept-only model.

Generalized Linear Models
Generalized Linear Models - Tests of Model Effects - November 5, 2017

Tests of Model EffectsTests of Model Effects, table, 2 levels of column headers and 1 levels of row headers, table with 4 columns and 5 rows
Source	Type III
Source	Wald Chi-Square	df	Sig.
(Intercept)	1733.278	1	.000
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept)

Generalized Linear Models
Generalized Linear Models - Parameter Estimates - November 5, 2017

Parameter EstimatesParameter Estimates, table, 2 levels of column headers and 1 levels of row headers, table with 11 columns and 7 rows
Parameter	B	Std. Error	95% Wald Confidence Interval		Hypothesis Test			Exp(B)	95% Wald Confidence Interval for Exp(B)
Parameter	B	Std. Error	Lower	Upper	Wald Chi-Square	df	Sig.	Exp(B)	Lower	Upper
(Intercept)	1.118	.0269	1.066	1.171	1733.278	1	.000	3.060	2.903	3.225
(Scale)	1^a
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept)
a. Fixed at the displayed value.

Log
Log - Log - November 5, 2017

* ============================================.
* Poisson Regression - pcs as predictor for d1
* ============================================.

* Generalized Linear Models.
GENLIN d1 WITH pcs
  /MODEL pcs INTERCEPT=YES
DISTRIBUTION=POISSON LINK=LOG
  /CRITERIA METHOD=FISHER(1) SCALE=1 COVB=MODEL MAXITERATIONS=100 MAXSTEPHALVING=5
    PCONVERGE=1E-006(ABSOLUTE) SINGULAR=1E-012 ANALYSISTYPE=3(WALD) CILEVEL=95 CITYPE=WALD
    LIKELIHOOD=FULL
  /MISSING CLASSMISSING=EXCLUDE
  /PRINT CPS DESCRIPTIVES MODELINFO FIT SUMMARY SOLUTION (EXPONENTIATED).

Generalized Linear Models
Generalized Linear Models - Model Information - November 5, 2017

Model InformationModel Information, table, 0 levels of column headers and 1 levels of row headers, table with 2 columns and 4 rows
Dependent Variable	d1 How many times hospitalized for medical problems (lifetime)
Probability Distribution	Poisson
Link Function	Log

Generalized Linear Models
Generalized Linear Models - Case Processing Summary - November 5, 2017

Case Processing SummaryCase Processing Summary, table, 1 levels of column headers and 1 levels of row headers, table with 3 columns and 5 rows
	N	Percent
Included	453	100.0%
Excluded	0	0.0%
Total	453	100.0%

Generalized Linear Models
Generalized Linear Models - Continuous Variable Information - November 5, 2017

Continuous Variable InformationContinuous Variable Information, table, 1 levels of column headers and 2 levels of row headers, table with 7 columns and 4 rows
		N	Minimum	Maximum	Mean	Std. Deviation
Dependent Variable	d1 How many times hospitalized for medical problems (lifetime)	453	0	100	3.06	6.188
Covariate	pcs SF36 Physical Composite Score - Baseline	453	14.0742912292480	74.8063278198242	48.048541551131024	10.784602685414228

Generalized Linear Models
Generalized Linear Models - Goodness of Fit - November 5, 2017

Goodness of Fit^aGoodness of Fit, table, 1 levels of column headers and 1 levels of row headers, table with 4 columns and 14 rows
	Value	df	Value/df
Deviance	1899.702	451	4.212
Scaled Deviance	1899.702	451
Pearson Chi-Square	3178.676	451	7.048
Scaled Pearson Chi-Square	3178.676	451
Log Likelihood^b	-1457.032
Akaike's Information Criterion (AIC)	2918.065
Finite Sample Corrected AIC (AICC)	2918.091
Bayesian Information Criterion (BIC)	2926.296
Consistent AIC (CAIC)	2928.296
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept), pcs
a. Information criteria are in smaller-is-better form.
b. The full log likelihood function is displayed and used in computing information criteria.

Generalized Linear Models
Generalized Linear Models - Omnibus Test - November 5, 2017

Omnibus Test^aOmnibus Test, table, 1 levels of column headers and 0 levels of row headers, table with 3 columns and 5 rows
Likelihood Ratio Chi-Square	df	Sig.
362.187	1	.000
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept), pcs
a. Compares the fitted model against the intercept-only model.

Generalized Linear Models
Generalized Linear Models - Tests of Model Effects - November 5, 2017

Tests of Model EffectsTests of Model Effects, table, 2 levels of column headers and 1 levels of row headers, table with 4 columns and 6 rows
Source	Type III
Source	Wald Chi-Square	df	Sig.
(Intercept)	916.659	1	.000
pcs	365.062	1	.000
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept), pcs

Generalized Linear Models
Generalized Linear Models - Parameter Estimates - November 5, 2017

Parameter EstimatesParameter Estimates, table, 2 levels of column headers and 1 levels of row headers, table with 11 columns and 8 rows
Parameter	B	Std. Error	95% Wald Confidence Interval		Hypothesis Test			Exp(B)	95% Wald Confidence Interval for Exp(B)
Parameter	B	Std. Error	Lower	Upper	Wald Chi-Square	df	Sig.	Exp(B)	Lower	Upper
(Intercept)	3.206	.1059	2.998	3.414	916.659	1	.000	24.679	20.054	30.372
pcs	-.046	.0024	-.051	-.041	365.062	1	.000	.955	.950	.959
(Scale)	1^a
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept), pcs
a. Fixed at the displayed value.

Log
Log - Log - November 5, 2017

* ============================================.
* Negative Binomial Regression - pcs as predictor for d1
* compare goodness of fit stats
* ============================================.

* Generalized Linear Models.
GENLIN d1 WITH pcs
  /MODEL pcs INTERCEPT=YES
DISTRIBUTION=NEGBIN(MLE) LINK=LOG
  /CRITERIA METHOD=FISHER(1) SCALE=1 COVB=MODEL MAXITERATIONS=100 MAXSTEPHALVING=5
    PCONVERGE=1E-006(ABSOLUTE) SINGULAR=1E-012 ANALYSISTYPE=3(WALD) CILEVEL=95 CITYPE=WALD
    LIKELIHOOD=FULL
  /MISSING CLASSMISSING=EXCLUDE
  /PRINT CPS DESCRIPTIVES MODELINFO FIT SUMMARY SOLUTION (EXPONENTIATED).

Generalized Linear Models
Generalized Linear Models - Model Information - November 5, 2017

Model InformationModel Information, table, 0 levels of column headers and 1 levels of row headers, table with 2 columns and 4 rows
Dependent Variable	d1 How many times hospitalized for medical problems (lifetime)
Probability Distribution	Negative binomial (MLE)
Link Function	Log

Generalized Linear Models
Generalized Linear Models - Case Processing Summary - November 5, 2017

Case Processing SummaryCase Processing Summary, table, 1 levels of column headers and 1 levels of row headers, table with 3 columns and 5 rows
	N	Percent
Included	453	100.0%
Excluded	0	0.0%
Total	453	100.0%

Generalized Linear Models
Generalized Linear Models - Continuous Variable Information - November 5, 2017

Continuous Variable InformationContinuous Variable Information, table, 1 levels of column headers and 2 levels of row headers, table with 7 columns and 4 rows
		N	Minimum	Maximum	Mean	Std. Deviation
Dependent Variable	d1 How many times hospitalized for medical problems (lifetime)	453	0	100	3.06	6.188
Covariate	pcs SF36 Physical Composite Score - Baseline	453	14.0742912292480	74.8063278198242	48.048541551131024	10.784602685414228

Generalized Linear Models
Generalized Linear Models - Goodness of Fit - November 5, 2017

Goodness of Fit^aGoodness of Fit, table, 1 levels of column headers and 1 levels of row headers, table with 4 columns and 14 rows
	Value	df	Value/df
Deviance	475.653	450	1.057
Scaled Deviance	475.653	450
Pearson Chi-Square	730.027	450	1.622
Scaled Pearson Chi-Square	730.027	450
Log Likelihood^b	-984.664
Akaike's Information Criterion (AIC)	1975.329
Finite Sample Corrected AIC (AICC)	1975.382
Bayesian Information Criterion (BIC)	1987.676
Consistent AIC (CAIC)	1990.676
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept), pcs
a. Information criteria are in smaller-is-better form.
b. The full log likelihood function is displayed and used in computing information criteria.

Generalized Linear Models
Generalized Linear Models - Omnibus Test - November 5, 2017

Omnibus Test^aOmnibus Test, table, 1 levels of column headers and 0 levels of row headers, table with 3 columns and 5 rows
Likelihood Ratio Chi-Square	df	Sig.
80.697	1	.000
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept), pcs
a. Compares the fitted model against the intercept-only model.

Generalized Linear Models
Generalized Linear Models - Tests of Model Effects - November 5, 2017

Tests of Model EffectsTests of Model Effects, table, 2 levels of column headers and 1 levels of row headers, table with 4 columns and 6 rows
Source	Type III
Source	Wald Chi-Square	df	Sig.
(Intercept)	180.959	1	.000
pcs	83.302	1	.000
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept), pcs

Generalized Linear Models
Generalized Linear Models - Parameter Estimates - November 5, 2017

Parameter EstimatesParameter Estimates, table, 2 levels of column headers and 1 levels of row headers, table with 11 columns and 9 rows
Parameter	B	Std. Error	95% Wald Confidence Interval		Hypothesis Test			Exp(B)	95% Wald Confidence Interval for Exp(B)
Parameter	B	Std. Error	Lower	Upper	Wald Chi-Square	df	Sig.	Exp(B)	Lower	Upper
(Intercept)	3.127	.2324	2.671	3.582	180.959	1	.000	22.796	14.455	35.950
pcs	-.044	.0049	-.054	-.035	83.302	1	.000	.957	.948	.966
(Scale)	1^a
(Negative binomial)	.910	.0855	.757	1.094
Dependent Variable: How many times hospitalized for medical problems (lifetime) Model: (Intercept), pcs
a. Fixed at the displayed value.

	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



Display Axis Title
Display Ticks