library(car)
## Loading required package: carData
library(stargazer)
##
## Please cite as:
## Hlavac, Marek (2018). stargazer: Well-Formatted Regression and Summary Statistics Tables.
## R package version 5.2.2. https://CRAN.R-project.org/package=stargazer
library(survey)
## Loading required package: grid
## Loading required package: Matrix
## Loading required package: survival
##
## Attaching package: 'survey'
## The following object is masked from 'package:graphics':
##
## dotchart
library(questionr)
library(foreign)
library(dplyr)
##
## Attaching package: 'dplyr'
## The following object is masked from 'package:car':
##
## recode
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
brfss2017=read.xport("~/Desktop/DataFolder/brfss.XPT ")
#Marital Status; Are you? (marital status): 1 Married, 2 Divorced, 3 Widowed, 4 Seprarated, 5 Never married, 6 Partner, 9 Refuse
brfss2017$marital<-Recode(brfss2017$MARITAL,recodes="1='b-Married';5='a-Single';2:4='c-D/W/S';6='partner';else=NA",as.factor=T)
#Sexual Orientation; Do you consider yourself to be? (sexual orientation): 1 Straight, 2 Gay, 3 Bisexual, 4 Other, 7 Don't know, 9 Refuse
brfss2017$sxorient<-Recode(brfss2017$SXORIENT,recodes="1='a-Straight';2='b-Gay';3='Bisexual';4:7='Other';else=NA",as.factor=T)
#Income
brfss2017$income<-Recode(brfss2017$INCOME2,recodes="1:3='<$20k';4:5='$20k<$35k';6='35k<50k';7='50k<75';8='>75k';else=NA",as.factor=T)
#Age
brfss2017$age<-Recode(brfss2017$X_AGE_G, recodes="1='18-24';2='25-34';3='35-44';4='45-54';5='55-64';6='65+';else=NA",as.factor=T)
#Education
brfss2017$educ<-Recode(brfss2017$EDUCA, recodes="1:2='0Prim'; 3='1somehs'; 4='2hsgrad'; 5='3somecol'; 6='4colgrad';9=NA", as.factor=T)
brfss2017$educ<-relevel(brfss2017$educ, ref='0Prim')
#Cholesterol; Have you EVER been told by a doctor that your cholesterol is high? 1 Yes, 2 No, 7 Dont know, 9 Refuse
brfss2017$choles<-Recode(brfss2017$TOLDHI2,recodes="1='Y';2='N';else=NA",as.factor=T)
#Blood Pressure; Are you currently taking medicine for blood Pressure? 1 Yes, 2 No, 7 Dont know, 9 Refuse
brfss2017$blpressure<-Recode(brfss2017$BPMEDS,recodes="1='Y';2='N';else=NA",as.factor=T)
#Health Insurance; Do you have health care coverage? 1 Yes, 2 No, 7 Dont know, 9 Refuse
brfss2017$hlthinsur<-Recode(brfss2017$HLTHPLN1,recodes="1='Y';2='N';else=NA",as.factor=T)
#DR Checkup; Had a doctor checkup with the last...? 1 year, 2 years, 5 years, 5+ years
brfss2017$checkup<-Recode(brfss2017$CHECKUP1,recodes="1='1 year';2='2 years';3='5 years';4='5+ years';else=NA",as.factor=T)
#Live in a Safe Neighborhood; How safe from crime do you consider your neighborhood to be? 1 Extremely Safe, 2 Safe, 3 Unsafe, 4 Extremely Unsafe, 7 Don't know, 9 Refuse
brfss2017$safeneigh<-Recode(brfss2017$HOWSAFE1,recodes="1:2='Safe';3:4='Unsafe';else=NA",as.factor=T)
#Mental Health; How many days during the past 30 days was your mental health not good? 1-30, 88 None, 77 Don't Know, 99 Refuse
brfss2017$menth<-Recode(brfss2017$MENTHLTH,recodes="88=0;77=NA;99=NA")
#Interaction between marital status and sexual orientation
brfss2017$int<-interaction(brfss2017$marital,brfss2017$sxorient)
brfss2017$int<-as.factor(brfss2017$int)
#Interaction between marital status and sexual orientation recoded to gay people only (gay single and gay married)
#brfss2017$int1<-Recode(brfss2017$int,recodes="0.1=0.1;1.1=1.1;else=NA",as.factor=T)
#Interaction between marital status and sexual orientation recoded to married people only (married straight and married gay)
#brfss2017$int2<-Recode(brfss2017$int,recodes="'M.G'='M.G';'M.S'='M.S';else=NA",as.factor=T)
#Homeowner; Do you own or rent your home? 1 Own, 2 Rent, 3 Other, 7 Don't know, 9 Refuse
brfss2017$ownhome<-Recode(brfss2017$RENTHOM1,recodes="1='Y';2:3='N';else=NA",as.factor=T)
#Smoking; Do you now smoke cigarettes every day, some days, or not at all? 1 Everyday, 2 Some days, 3 Not at all, 7 Don't know, 9 Refuse
brfss2017$smoke<-Recode(brfss2017$SMOKDAY2,recodes="1:2='Y';3='N';else=NA",as.factor=T)
#Was there a time in the past 12 months when you needed to see a doctor but could not because of cost? 1 Yes, 2 No, 7 Don't know, 9 Refuse
brfss2017$medcost<-Recode(brfss2017$MEDCOST,recodes="1='Y';2='N';else=NA",as.factor=T)
#During the past 30 days, on the days when you drank, about how many drinks did you drink on the average?
brfss2017$drinks<-Recode(brfss2017$AVEDRNK2,recodes="1:5='1-5';5:10='5-10';10:30='10-30';77=NA;99=NA;else=NA",as.factor=T)
#BMI; Four-categories of Body Mass Index (BMI): 1 Underweight, 2 Normal Weight, 3 Overweight, 4 Obese
brfss2017$bmi<-brfss2017$X_BMI5CAT
#Not good mental health days: 1 (0 days), 2 (1-13 days), 3 (14-30 days), 9 Don't know
brfss2017$badmental<-Recode(brfss2017$X_MENT14D,recodes="1=0;2:3=1;9=NA;else=NA",as.factor=T)
#Race/Ethnicity
brfss2017$black<-Recode(brfss2017$racegr3, recodes="2=1; 9=NA; else=0")
brfss2017$white<-Recode(brfss2017$racegr3, recodes="1=1; 9=NA; else=0")
brfss2017$other<-Recode(brfss2017$racegr3, recodes="3:4=1; 9=NA; else=0")
brfss2017$hispanic<-Recode(brfss2017$racegr3, recodes="5=1; 9=NA; else=0")
#Race/Ethnicity; 1 White, 2 Black, 3 Other, 4 Multiracial, 5 Hispanic, 9 Don't know
brfss2017$race<-Recode(brfss2017$X_RACEGR3, recodes="2='black'; 1='awhite'; 3:4='other';5='hispanic'; else=NA",as.factor=T)
brfss2017$race<-relevel(brfss2017$race, ref='awhite')
#Sex; 1 Male, 2 Female
brfss2017$sex<-Recode(brfss2017$SEX,recodes="1=1;2=0;9=NA;else=NA",as.factor=T)
brfss2017$genh<-Recode(brfss2017$GENHLTH,recodes="1:3='aGood';4:5='Bad';else=NA",as.factor=T)
brfss2017$physh<-Recode(brfss2017$PHYSHLTH,recodes="88=0;77=NA;99=NA",as.factor=T)
brfss2017$menth<-Recode(brfss2017$MENTHLTH,recodes="88=0;77=NA;99=NA",as.factor=T)
brfss2017$emosupport<-Recode(brfss2017$EMTSUPRT,recodes="1:2='a-Yes';3='b-Sometimes';4:5='c-No';else=NA",as.factor=T)
brfss2017$satisfed<-Recode(brfss2017$LSATISFY,recodes="1:2='a-Satisfied';3:4='b-Dissatisfied';else=NA",as.factor=T)
brfss2017$employ<-Recode(brfss2017$EMPLOY1,recodes="1:2='Yes';3:4='No';5:8='other';else=NA",as.factor=T)
brfss2017$personalDR<-Recode(brfss2017$PERSDOC2,recodes="1:2='Yes';3='No';else=NA",as.factor=T)
brfss2017$chol<-Recode(brfss2017$CHOLCHK1,recodes="1='N/2yr+';3:7='N/2yr+';2='Within 1yr';else=NA",as.factor=T)
brfss2017$seatblt<-Recode(brfss2017$SEATBELT,recodes="1='Yes';2:7='Sometimes/No';else=NA",as.factor=T)
brfss2017$flushot<-Recode(brfss2017$FLUSHOT6,recode="1='Yes';2='No';else=NA",as.factor=T)
brfss2017$HIVrisk<-Recode(brfss2017$HIVRISK5,recode="1='HighRisk';2='LowRisk';else=NA",as.factor=T)
brfss2017$money4bills<-Recode(brfss2017$SDHMONEY,recode="1='More';2='Exact';3='No';else=NA",as.factor=T)
#str(brfss2017$int1)
#brfss2017 <- brfss2017[complete.cases(brfss2017), ]
#dplyr::select
brfss2017 <- brfss2017 %>%
select(genh,sxorient,marital,badmental,int,sex,age,race,educ,income,smoke,drinks,choles,blpressure,bmi,medcost,hlthinsur,checkup,ownhome,safeneigh,physh,menth,emosupport,satisfed,employ,personalDR,chol,seatblt,flushot,HIVrisk,money4bills,X_STRWT,X_LLCPWT)
brfss2017 <- brfss2017 %>%
subset(sxorient=='b-Gay')
table(brfss2017$sxorient)
##
## a-Straight b-Gay Bisexual Other
## 0 3167 0 0
#Create a survey design
brfss2017%>%
filter(complete.cases(.))
## genh sxorient marital badmental int sex age race
## 1 aGood b-Gay a-Single 0 a-Single.b-Gay 1 45-54 awhite
## 2 aGood b-Gay c-D/W/S 0 c-D/W/S.b-Gay 1 55-64 awhite
## 3 aGood b-Gay c-D/W/S 0 c-D/W/S.b-Gay 0 65+ awhite
## 4 aGood b-Gay a-Single 0 a-Single.b-Gay 1 55-64 awhite
## 5 aGood b-Gay b-Married 0 b-Married.b-Gay 0 65+ awhite
## 6 aGood b-Gay a-Single 0 a-Single.b-Gay 0 65+ awhite
## 7 aGood b-Gay b-Married 0 b-Married.b-Gay 1 65+ awhite
## 8 aGood b-Gay c-D/W/S 0 c-D/W/S.b-Gay 1 65+ awhite
## 9 aGood b-Gay a-Single 0 a-Single.b-Gay 1 55-64 awhite
## 10 aGood b-Gay a-Single 0 a-Single.b-Gay 1 65+ awhite
## 11 aGood b-Gay c-D/W/S 1 c-D/W/S.b-Gay 0 65+ awhite
## 12 aGood b-Gay a-Single 0 a-Single.b-Gay 0 35-44 awhite
## 13 aGood b-Gay a-Single 0 a-Single.b-Gay 1 35-44 awhite
## 14 aGood b-Gay a-Single 0 a-Single.b-Gay 0 65+ awhite
## 15 aGood b-Gay b-Married 0 b-Married.b-Gay 1 55-64 awhite
## 16 aGood b-Gay a-Single 1 a-Single.b-Gay 1 55-64 awhite
## 17 Bad b-Gay c-D/W/S 1 c-D/W/S.b-Gay 0 45-54 awhite
## 18 aGood b-Gay c-D/W/S 0 c-D/W/S.b-Gay 1 65+ awhite
## 19 Bad b-Gay a-Single 1 a-Single.b-Gay 0 55-64 awhite
## 20 aGood b-Gay a-Single 1 a-Single.b-Gay 1 25-34 awhite
## 21 aGood b-Gay partner 0 partner.b-Gay 1 65+ awhite
## 22 aGood b-Gay b-Married 0 b-Married.b-Gay 0 45-54 awhite
## 23 aGood b-Gay b-Married 0 b-Married.b-Gay 0 55-64 awhite
## 24 aGood b-Gay c-D/W/S 0 c-D/W/S.b-Gay 1 65+ awhite
## 25 Bad b-Gay a-Single 0 a-Single.b-Gay 1 55-64 awhite
## 26 aGood b-Gay a-Single 1 a-Single.b-Gay 1 55-64 awhite
## 27 aGood b-Gay a-Single 1 a-Single.b-Gay 0 25-34 awhite
## 28 aGood b-Gay b-Married 1 b-Married.b-Gay 1 25-34 awhite
## 29 aGood b-Gay partner 0 partner.b-Gay 0 55-64 awhite
## 30 aGood b-Gay c-D/W/S 0 c-D/W/S.b-Gay 1 55-64 awhite
## 31 aGood b-Gay b-Married 1 b-Married.b-Gay 0 35-44 other
## educ income smoke drinks choles blpressure bmi medcost hlthinsur
## 1 4colgrad >75k N 1-5 N Y 4 N Y
## 2 4colgrad >75k N 1-5 Y Y 4 N Y
## 3 4colgrad >75k N 1-5 Y Y 4 N Y
## 4 4colgrad >75k N 1-5 N Y 4 N Y
## 5 2hsgrad $20k<$35k N 1-5 Y Y 4 N Y
## 6 4colgrad 50k<75 N 1-5 N Y 4 N Y
## 7 4colgrad >75k N 1-5 Y Y 3 N Y
## 8 3somecol $20k<$35k N 1-5 N Y 3 N Y
## 9 4colgrad >75k N 1-5 N Y 2 N Y
## 10 4colgrad $20k<$35k N 1-5 N N 2 N Y
## 11 3somecol <$20k N 1-5 N Y 3 N Y
## 12 4colgrad 35k<50k Y 1-5 Y N 4 Y Y
## 13 4colgrad >75k N 1-5 N Y 3 N Y
## 14 3somecol <$20k Y 1-5 Y Y 1 N Y
## 15 4colgrad >75k N 1-5 Y Y 3 N Y
## 16 1somehs <$20k Y 1-5 Y Y 4 N Y
## 17 3somecol <$20k N 1-5 N Y 4 N Y
## 18 4colgrad 35k<50k Y 1-5 N Y 2 N Y
## 19 0Prim <$20k Y 1-5 Y Y 1 N Y
## 20 4colgrad 50k<75 Y 5-10 Y N 4 N N
## 21 3somecol $20k<$35k Y 1-5 N Y 3 N Y
## 22 4colgrad >75k N 1-5 N N 4 N Y
## 23 4colgrad 50k<75 N 1-5 N Y 4 N Y
## 24 1somehs 35k<50k N 1-5 Y Y 4 N Y
## 25 4colgrad <$20k Y 1-5 Y Y 4 N Y
## 26 2hsgrad <$20k N 1-5 N Y 4 N Y
## 27 4colgrad >75k N 1-5 N N 2 N Y
## 28 3somecol $20k<$35k N 1-5 N N 3 N Y
## 29 4colgrad >75k N 1-5 Y Y 2 N Y
## 30 3somecol <$20k N 1-5 Y Y 4 N Y
## 31 3somecol >75k Y 1-5 N Y 4 Y Y
## checkup ownhome safeneigh physh menth emosupport satisfed
## 1 2 years Y Safe 0 0 a-Yes a-Satisfied
## 2 1 year Y Safe 0 0 a-Yes a-Satisfied
## 3 1 year Y Safe 0 0 a-Yes a-Satisfied
## 4 1 year Y Safe 0 0 a-Yes a-Satisfied
## 5 1 year Y Safe 0 0 a-Yes a-Satisfied
## 6 1 year Y Safe 7 0 a-Yes a-Satisfied
## 7 1 year Y Safe 0 0 a-Yes a-Satisfied
## 8 1 year Y Safe 0 0 b-Sometimes a-Satisfied
## 9 1 year Y Safe 30 0 a-Yes a-Satisfied
## 10 1 year Y Safe 0 0 b-Sometimes a-Satisfied
## 11 2 years Y Safe 0 2 c-No a-Satisfied
## 12 5 years Y Safe 5 0 a-Yes a-Satisfied
## 13 1 year N Safe 5 0 a-Yes a-Satisfied
## 14 2 years Y Safe 0 0 c-No a-Satisfied
## 15 1 year Y Safe 0 0 a-Yes a-Satisfied
## 16 1 year N Safe 30 30 c-No b-Dissatisfied
## 17 1 year Y Safe 28 15 b-Sometimes a-Satisfied
## 18 1 year Y Safe 0 0 b-Sometimes b-Dissatisfied
## 19 1 year N Safe 20 10 a-Yes a-Satisfied
## 20 1 year N Safe 2 1 a-Yes a-Satisfied
## 21 1 year N Safe 5 0 c-No a-Satisfied
## 22 1 year Y Safe 0 0 a-Yes a-Satisfied
## 23 1 year Y Safe 0 0 a-Yes a-Satisfied
## 24 1 year Y Safe 10 0 a-Yes a-Satisfied
## 25 1 year N Safe 7 0 a-Yes a-Satisfied
## 26 1 year Y Safe 3 7 a-Yes a-Satisfied
## 27 1 year Y Safe 0 5 a-Yes a-Satisfied
## 28 5+ years N Safe 1 1 a-Yes a-Satisfied
## 29 1 year Y Safe 0 0 a-Yes a-Satisfied
## 30 1 year Y Unsafe 2 0 a-Yes a-Satisfied
## 31 1 year N Safe 0 7 a-Yes a-Satisfied
## employ personalDR chol seatblt flushot HIVrisk money4bills
## 1 Yes Yes N/2yr+ Yes No LowRisk More
## 2 Yes Yes Within 1yr Yes Yes HighRisk Exact
## 3 Yes Yes Within 1yr Yes No LowRisk More
## 4 other Yes Within 1yr Yes Yes LowRisk More
## 5 other Yes Within 1yr Yes Yes LowRisk Exact
## 6 other Yes Within 1yr Yes Yes LowRisk More
## 7 other Yes Within 1yr Yes No LowRisk More
## 8 other Yes Within 1yr Yes No LowRisk Exact
## 9 other Yes Within 1yr Yes No LowRisk More
## 10 Yes Yes Within 1yr Yes Yes HighRisk No
## 11 Yes No N/2yr+ Yes Yes LowRisk Exact
## 12 Yes No N/2yr+ Yes No LowRisk Exact
## 13 Yes Yes Within 1yr Yes No HighRisk More
## 14 other Yes N/2yr+ Yes No LowRisk More
## 15 Yes Yes Within 1yr Yes Yes LowRisk More
## 16 No Yes Within 1yr Yes Yes LowRisk No
## 17 other Yes Within 1yr Yes Yes LowRisk No
## 18 other Yes Within 1yr Yes No LowRisk Exact
## 19 other Yes Within 1yr Yes Yes LowRisk No
## 20 Yes Yes Within 1yr Yes No HighRisk Exact
## 21 other Yes Within 1yr Yes Yes LowRisk Exact
## 22 Yes Yes Within 1yr Yes Yes LowRisk Exact
## 23 other Yes Within 1yr Yes Yes LowRisk More
## 24 other Yes Within 1yr Sometimes/No No HighRisk More
## 25 other Yes Within 1yr Yes No LowRisk Exact
## 26 Yes Yes N/2yr+ Sometimes/No No LowRisk More
## 27 Yes Yes Within 1yr Yes No HighRisk More
## 28 Yes No N/2yr+ Yes No HighRisk Exact
## 29 other Yes Within 1yr Yes No LowRisk More
## 30 other Yes Within 1yr Yes No LowRisk Exact
## 31 Yes Yes Within 1yr Yes No LowRisk More
## X_STRWT X_LLCPWT
## 1 21.74020 538.30762
## 2 21.74020 95.06265
## 3 13.30360 53.45304
## 4 22.82476 79.98952
## 5 21.74020 287.86580
## 6 21.74020 64.76206
## 7 13.14636 98.82987
## 8 21.74020 137.28997
## 9 13.14636 56.78258
## 10 67.43929 230.58637
## 11 67.43929 522.52499
## 12 67.43929 469.54269
## 13 67.43929 261.12408
## 14 67.43929 516.14330
## 15 67.43929 214.49235
## 16 67.43929 589.95251
## 17 67.43929 730.02556
## 18 67.43929 249.68454
## 19 27.94482 199.01761
## 20 27.94482 113.94125
## 21 11.58969 36.81910
## 22 27.94482 93.95368
## 23 27.94482 83.59105
## 24 12.37312 122.76381
## 25 97.75916 168.77126
## 26 39.23976 157.68665
## 27 126.24787 585.74339
## 28 494.53985 2133.25829
## 29 154.35401 363.37455
## 30 349.16801 2386.52211
## 31 154.35401 384.96429
options(survey.lonely.psu = "adjust")
des<-svydesign(ids = ~1,strata=~X_STRWT,weights = ~X_LLCPWT,data = brfss2017)
table(brfss2017$genh)
##
## aGood Bad
## 2569 595
#MODEL ONE
#General Health by Marital Status
fit.logit1<-svyglm(genh~marital,design=des,family=binomial)
## Warning in eval(family$initialize): non-integer #successes in a binomial
## glm!
summary(fit.logit1)
##
## Call:
## svyglm(formula = genh ~ marital, design = des, family = binomial)
##
## Survey design:
## svydesign(ids = ~1, strata = ~X_STRWT, weights = ~X_LLCPWT, data = brfss2017)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.76399 0.11489 -15.354 < 2e-16 ***
## maritalb-Married -0.09431 0.23311 -0.405 0.685827
## maritalc-D/W/S 0.86897 0.23562 3.688 0.000231 ***
## maritalpartner 0.19801 0.27475 0.721 0.471164
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1.003631)
##
## Number of Fisher Scoring iterations: 4
#MODEL TWO - includes biological demographic variables
#General Health by marital status and sex, age, race
fit.logit2<-svyglm(genh~marital+sex+age+race,design=des,family=binomial)
## Warning in eval(family$initialize): non-integer #successes in a binomial
## glm!
summary(fit.logit2)
##
## Call:
## svyglm(formula = genh ~ marital + sex + age + race, design = des,
## family = binomial)
##
## Survey design:
## svydesign(ids = ~1, strata = ~X_STRWT, weights = ~X_LLCPWT, data = brfss2017)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -2.08866 0.26918 -7.759 1.24e-14 ***
## maritalb-Married -0.47988 0.27255 -1.761 0.078413 .
## maritalc-D/W/S 0.40622 0.26122 1.555 0.120057
## maritalpartner -0.01209 0.30836 -0.039 0.968716
## sex1 -0.45024 0.18392 -2.448 0.014434 *
## age25-34 0.52009 0.33479 1.553 0.120435
## age35-44 0.80618 0.37225 2.166 0.030431 *
## age45-54 0.92709 0.35536 2.609 0.009138 **
## age55-64 1.01325 0.33699 3.007 0.002667 **
## age65+ 1.22290 0.34903 3.504 0.000467 ***
## raceblack 0.06559 0.25640 0.256 0.798104
## racehispanic 0.23475 0.27767 0.845 0.397946
## raceother -0.32375 0.29244 -1.107 0.268384
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1.010438)
##
## Number of Fisher Scoring iterations: 4
regTermTest(fit.logit2, test.terms = ~sex+age+race, method="Wald", df = NULL)
## Wald test for sex age race
## in svyglm(formula = genh ~ marital + sex + age + race, design = des,
## family = binomial)
## F = 2.728861 on 9 and 2461 df: p= 0.0036133
####MODEL THREE - includes biological demographic variables and social demographic variables
#General Health by Marital Status plus sex, age, race, plus educ, income
fit.logit3<-svyglm(genh~marital+sex+age+race+educ+income,design=des,family=binomial)
## Warning in eval(family$initialize): non-integer #successes in a binomial
## glm!
summary(fit.logit3)
##
## Call:
## svyglm(formula = genh ~ marital + sex + age + race + educ + income,
## design = des, family = binomial)
##
## Survey design:
## svydesign(ids = ~1, strata = ~X_STRWT, weights = ~X_LLCPWT, data = brfss2017)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.0479 1.0319 -1.016 0.309955
## maritalb-Married 0.1036 0.2974 0.348 0.727609
## maritalc-D/W/S 0.2963 0.2967 0.999 0.318074
## maritalpartner 0.2714 0.3246 0.836 0.403167
## sex1 -0.4393 0.1937 -2.268 0.023450 *
## age25-34 0.5866 0.3824 1.534 0.125173
## age35-44 0.9393 0.4097 2.293 0.021948 *
## age45-54 1.2165 0.3856 3.155 0.001626 **
## age55-64 1.2605 0.3674 3.431 0.000613 ***
## age65+ 1.2612 0.3970 3.177 0.001510 **
## raceblack -0.5118 0.2887 -1.773 0.076383 .
## racehispanic -0.1761 0.3163 -0.557 0.577682
## raceother -0.4417 0.3387 -1.304 0.192401
## educ1somehs 0.0633 1.0286 0.062 0.950938
## educ2hsgrad -0.2050 0.9777 -0.210 0.833946
## educ3somecol -0.3705 0.9853 -0.376 0.706963
## educ4colgrad -0.5396 1.0004 -0.539 0.589691
## income>75k -2.1150 0.3442 -6.144 9.54e-10 ***
## income$20k<$35k -0.7819 0.2663 -2.936 0.003359 **
## income35k<50k -0.6766 0.3089 -2.190 0.028629 *
## income50k<75 -1.1454 0.3058 -3.746 0.000185 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1.012634)
##
## Number of Fisher Scoring iterations: 5
regTermTest(fit.logit3, test.terms = ~sex+age+race+educ+income, method="Wald", df = NULL)
## Wald test for sex age race educ income
## in svyglm(formula = genh ~ marital + sex + age + race + educ + income,
## design = des, family = binomial)
## F = 4.904605 on 17 and 2183 df: p= 1.6174e-10
#MODEL FOUR -includes biological and social demographic variables plus marriage behavior benefit variables
fit.logit4<-svyglm(genh~marital+sex+age+race+educ+income+smoke+drinks+bmi,design=des,family=binomial)
## Warning in eval(family$initialize): non-integer #successes in a binomial
## glm!
summary(fit.logit4)
##
## Call:
## svyglm(formula = genh ~ marital + sex + age + race + educ + income +
## smoke + drinks + bmi, design = des, family = binomial)
##
## Survey design:
## svydesign(ids = ~1, strata = ~X_STRWT, weights = ~X_LLCPWT, data = brfss2017)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.09019 1.69540 -0.643 0.520511
## maritalb-Married -0.37293 0.45732 -0.815 0.415201
## maritalc-D/W/S 0.53286 0.51027 1.044 0.296876
## maritalpartner 0.25934 0.47837 0.542 0.587973
## sex1 -0.44957 0.37925 -1.185 0.236428
## age25-34 0.73967 0.74645 0.991 0.322215
## age35-44 0.49478 0.82676 0.598 0.549810
## age45-54 1.63604 0.78452 2.085 0.037553 *
## age55-64 0.89379 0.76199 1.173 0.241379
## age65+ 1.65724 0.80671 2.054 0.040478 *
## raceblack -0.70918 0.60365 -1.175 0.240641
## racehispanic 0.29432 0.58306 0.505 0.613942
## raceother -0.57160 0.53258 -1.073 0.283689
## educ1somehs -1.26233 1.50372 -0.839 0.401617
## educ2hsgrad -1.12342 1.34913 -0.833 0.405423
## educ3somecol -1.17521 1.31509 -0.894 0.371958
## educ4colgrad -0.93540 1.34326 -0.696 0.486534
## income>75k -2.04647 0.59591 -3.434 0.000645 ***
## income$20k<$35k -0.13749 0.50576 -0.272 0.785848
## income35k<50k -0.54114 0.62158 -0.871 0.384405
## income50k<75 -1.19088 0.58038 -2.052 0.040716 *
## smokeY 0.44289 0.37968 1.166 0.243989
## drinks10-30 -0.36645 0.86829 -0.422 0.673188
## drinks5-10 0.26220 0.48599 0.540 0.589769
## bmi 0.07803 0.21592 0.361 0.717968
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1.11061)
##
## Number of Fisher Scoring iterations: 5
regTermTest(fit.logit4, test.terms = ~sex+age+race+educ+income+smoke+drinks+bmi, method="Wald", df = NULL)
## Wald test for sex age race educ income smoke drinks bmi
## in svyglm(formula = genh ~ marital + sex + age + race + educ + income +
## smoke + drinks + bmi, design = des, family = binomial)
## F = 2.82611 on 21 and 487 df: p= 3.4977e-05
#MODEL FIVE - includes biological and social demographic variables plus marriage behavior benefit variables marriage benefit medical variables
fit.logit5<-svyglm(genh~marital+sex+age+race+educ+income+smoke+drinks+bmi+medcost+hlthinsur+checkup,design=des,family=binomial)
## Warning in eval(family$initialize): non-integer #successes in a binomial
## glm!
summary(fit.logit5)
##
## Call:
## svyglm(formula = genh ~ marital + sex + age + race + educ + income +
## smoke + drinks + bmi + medcost + hlthinsur + checkup, design = des,
## family = binomial)
##
## Survey design:
## svydesign(ids = ~1, strata = ~X_STRWT, weights = ~X_LLCPWT, data = brfss2017)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -3.24857 1.63455 -1.987 0.047445 *
## maritalb-Married -0.13397 0.43818 -0.306 0.759937
## maritalc-D/W/S 0.67140 0.51602 1.301 0.193847
## maritalpartner 0.03095 0.50205 0.062 0.950864
## sex1 -0.28874 0.34022 -0.849 0.396485
## age25-34 1.15696 0.74390 1.555 0.120548
## age35-44 0.68900 0.79211 0.870 0.384828
## age45-54 2.06463 0.78729 2.622 0.009009 **
## age55-64 1.38682 0.77907 1.780 0.075695 .
## age65+ 2.46466 0.84363 2.921 0.003649 **
## raceblack -0.66682 0.67638 -0.986 0.324703
## racehispanic 0.05070 0.53994 0.094 0.925222
## raceother 0.09799 0.56782 0.173 0.863062
## educ1somehs -1.32517 1.58826 -0.834 0.404498
## educ2hsgrad -0.70928 1.20755 -0.587 0.557235
## educ3somecol -0.52006 1.19919 -0.434 0.664719
## educ4colgrad -0.53345 1.23004 -0.434 0.664714
## income>75k -2.39273 0.66967 -3.573 0.000389 ***
## income$20k<$35k -0.60036 0.55689 -1.078 0.281546
## income35k<50k -1.19463 0.57432 -2.080 0.038053 *
## income50k<75 -1.35534 0.64492 -2.102 0.036115 *
## smokeY 0.68895 0.37570 1.834 0.067310 .
## drinks10-30 0.07147 0.84985 0.084 0.933012
## drinks5-10 0.19304 0.53533 0.361 0.718563
## bmi 0.04824 0.20627 0.234 0.815183
## medcostY 1.86801 0.46969 3.977 8.06e-05 ***
## hlthinsurY 1.14366 0.66443 1.721 0.085853 .
## checkup2 years -1.25231 0.56869 -2.202 0.028137 *
## checkup5 years -0.65965 0.63665 -1.036 0.300672
## checkup5+ years 0.44459 0.60820 0.731 0.465145
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 0.9953725)
##
## Number of Fisher Scoring iterations: 6
regTermTest(fit.logit5, test.terms = ~sex+age+race+educ+income+smoke+drinks+bmi+medcost+hlthinsur+checkup, method="Wald", df = NULL)
## Wald test for sex age race educ income smoke drinks bmi medcost hlthinsur checkup
## in svyglm(formula = genh ~ marital + sex + age + race + educ + income +
## smoke + drinks + bmi + medcost + hlthinsur + checkup, design = des,
## family = binomial)
## F = 3.340716 on 26 and 477 df: p= 1.1173e-07
#MODEL SIX - includes biological and social demographic variables plus marriage behavior, medical, and social benefit variables
fit.logit6<-svyglm(genh~marital+sex+age+race+educ+income+smoke+drinks+bmi+medcost+hlthinsur+checkup+ownhome+safeneigh,design=des,family=binomial)
## Warning in eval(family$initialize): non-integer #successes in a binomial
## glm!
summary(fit.logit6)
##
## Call:
## svyglm(formula = genh ~ marital + sex + age + race + educ + income +
## smoke + drinks + bmi + medcost + hlthinsur + checkup + ownhome +
## safeneigh, design = des, family = binomial)
##
## Survey design:
## svydesign(ids = ~1, strata = ~X_STRWT, weights = ~X_LLCPWT, data = brfss2017)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.57516 4.50679 0.793 0.429353
## maritalb-Married -2.57658 0.86902 -2.965 0.003726 **
## maritalc-D/W/S 0.78224 0.96944 0.807 0.421497
## maritalpartner -1.50125 0.93981 -1.597 0.113098
## sex1 -1.81042 0.61458 -2.946 0.003946 **
## age25-34 0.63990 1.58806 0.403 0.687784
## age35-44 -3.28065 2.01465 -1.628 0.106355
## age45-54 2.53663 1.47103 1.724 0.087499 .
## age55-64 1.07662 1.70687 0.631 0.529530
## age65+ 3.63355 1.65046 2.202 0.029825 *
## raceblack -3.13185 1.44631 -2.165 0.032557 *
## racehispanic 0.17554 1.17695 0.149 0.881716
## raceother 3.06107 1.25008 2.449 0.015947 *
## educ1somehs -7.19155 4.05792 -1.772 0.079177 .
## educ2hsgrad -5.10949 3.58191 -1.426 0.156617
## educ3somecol -5.13069 3.69867 -1.387 0.168245
## educ4colgrad -6.09217 3.80900 -1.599 0.112650
## income>75k -0.81916 1.06054 -0.772 0.441567
## income$20k<$35k 0.01275 0.91843 0.014 0.988953
## income35k<50k -1.57516 1.01629 -1.550 0.124090
## income50k<75 0.09722 1.08261 0.090 0.928610
## smokeY -0.38653 0.62550 -0.618 0.537904
## drinks10-30 2.57456 1.25531 2.051 0.042693 *
## drinks5-10 -0.69588 1.03353 -0.673 0.502190
## bmi 0.10430 0.39970 0.261 0.794630
## medcostY 3.12381 0.87010 3.590 0.000499 ***
## hlthinsurY 0.84812 1.25612 0.675 0.501000
## checkup2 years -1.59237 0.97526 -1.633 0.105431
## checkup5 years -3.49687 1.73770 -2.012 0.046671 *
## checkup5+ years -0.49521 0.86452 -0.573 0.567957
## ownhomeY -1.04495 0.82067 -1.273 0.205647
## safeneighUnsafe 0.25143 0.89546 0.281 0.779418
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 3.107481)
##
## Number of Fisher Scoring iterations: 7
regTermTest(fit.logit6, test.terms = ~sex+age+race+educ+income+smoke+drinks+bmi+medcost+hlthinsur+checkup+ownhome+safeneigh, method="Wald", df = NULL)
## Wald test for sex age race educ income smoke drinks bmi medcost hlthinsur checkup ownhome safeneigh
## in svyglm(formula = genh ~ marital + sex + age + race + educ + income +
## smoke + drinks + bmi + medcost + hlthinsur + checkup + ownhome +
## safeneigh, design = des, family = binomial)
## F = 2.095336 on 28 and 108 df: p= 0.0036576
stargazer(fit.logit1, fit.logit2, fit.logit3,fit.logit4,fit.logit5,fit.logit6,type = "html", style="demography", covariate.labels =c("Married","D/W/S","partner","Male","25-34","35-44","45-54","55-64","65+","Black","Hisp","Other","someHS","HSgrad","someCol","Colgrad","75+","20-35","35-50","50-75","smokeY","drinks10-30","drinks5-10","bmi","medcostY","hlthinsY","CU2","CU5","CU5+","ownhome","safeneighUnsaf"), ci = T )
|
|
|
|
genh
|
|
|
Model 1
|
Model 2
|
Model 3
|
Model 4
|
Model 5
|
Model 6
|
|
|
|
Married
|
-0.094
|
-0.480
|
0.104
|
-0.373
|
-0.134
|
-2.577**
|
|
|
(-0.551, 0.363)
|
(-1.014, 0.054)
|
(-0.479, 0.686)
|
(-1.269, 0.523)
|
(-0.993, 0.725)
|
(-4.280, -0.873)
|
|
D/W/S
|
0.869***
|
0.406
|
0.296
|
0.533
|
0.671
|
0.782
|
|
|
(0.407, 1.331)
|
(-0.106, 0.918)
|
(-0.285, 0.878)
|
(-0.467, 1.533)
|
(-0.340, 1.683)
|
(-1.118, 2.682)
|
|
partner
|
0.198
|
-0.012
|
0.271
|
0.259
|
0.031
|
-1.501
|
|
|
(-0.340, 0.737)
|
(-0.616, 0.592)
|
(-0.365, 0.908)
|
(-0.678, 1.197)
|
(-0.953, 1.015)
|
(-3.343, 0.341)
|
|
Male
|
|
-0.450*
|
-0.439*
|
-0.450
|
-0.289
|
-1.810**
|
|
|
|
(-0.811, -0.090)
|
(-0.819, -0.060)
|
(-1.193, 0.294)
|
(-0.956, 0.378)
|
(-3.015, -0.606)
|
|
25-34
|
|
0.520
|
0.587
|
0.740
|
1.157
|
0.640
|
|
|
|
(-0.136, 1.176)
|
(-0.163, 1.336)
|
(-0.723, 2.203)
|
(-0.301, 2.615)
|
(-2.473, 3.752)
|
|
35-44
|
|
0.806*
|
0.939*
|
0.495
|
0.689
|
-3.281
|
|
|
|
(0.077, 1.536)
|
(0.136, 1.742)
|
(-1.126, 2.115)
|
(-0.864, 2.242)
|
(-7.229, 0.668)
|
|
45-54
|
|
0.927**
|
1.217**
|
1.636*
|
2.065**
|
2.537
|
|
|
|
(0.231, 1.624)
|
(0.461, 1.972)
|
(0.098, 3.174)
|
(0.522, 3.608)
|
(-0.347, 5.420)
|
|
55-64
|
|
1.013**
|
1.260***
|
0.894
|
1.387
|
1.077
|
|
|
|
(0.353, 1.674)
|
(0.540, 1.981)
|
(-0.600, 2.387)
|
(-0.140, 2.914)
|
(-2.269, 4.422)
|
|
65+
|
|
1.223***
|
1.261**
|
1.657*
|
2.465**
|
3.634*
|
|
|
|
(0.539, 1.907)
|
(0.483, 2.039)
|
(0.076, 3.238)
|
(0.811, 4.118)
|
(0.399, 6.868)
|
|
Black
|
|
0.066
|
-0.512
|
-0.709
|
-0.667
|
-3.132*
|
|
|
|
(-0.437, 0.568)
|
(-1.078, 0.054)
|
(-1.892, 0.474)
|
(-1.992, 0.659)
|
(-5.967, -0.297)
|
|
Hisp
|
|
0.235
|
-0.176
|
0.294
|
0.051
|
0.176
|
|
|
|
(-0.309, 0.779)
|
(-0.796, 0.444)
|
(-0.848, 1.437)
|
(-1.008, 1.109)
|
(-2.131, 2.482)
|
|
Other
|
|
-0.324
|
-0.442
|
-0.572
|
0.098
|
3.061*
|
|
|
|
(-0.897, 0.249)
|
(-1.106, 0.222)
|
(-1.615, 0.472)
|
(-1.015, 1.211)
|
(0.611, 5.511)
|
|
someHS
|
|
|
0.063
|
-1.262
|
-1.325
|
-7.192
|
|
|
|
|
(-1.953, 2.079)
|
(-4.210, 1.685)
|
(-4.438, 1.788)
|
(-15.145, 0.762)
|
|
HSgrad
|
|
|
-0.205
|
-1.123
|
-0.709
|
-5.109
|
|
|
|
|
(-2.121, 1.711)
|
(-3.768, 1.521)
|
(-3.076, 1.657)
|
(-12.130, 1.911)
|
|
someCol
|
|
|
-0.370
|
-1.175
|
-0.520
|
-5.131
|
|
|
|
|
(-2.302, 1.561)
|
(-3.753, 1.402)
|
(-2.870, 1.830)
|
(-12.380, 2.119)
|
|
Colgrad
|
|
|
-0.540
|
-0.935
|
-0.533
|
-6.092
|
|
|
|
|
(-2.500, 1.421)
|
(-3.568, 1.697)
|
(-2.944, 1.877)
|
(-13.558, 1.373)
|
|
75+
|
|
|
-2.115***
|
-2.046***
|
-2.393***
|
-0.819
|
|
|
|
|
(-2.790, -1.440)
|
(-3.214, -0.879)
|
(-3.705, -1.080)
|
(-2.898, 1.259)
|
|
20-35
|
|
|
-0.782**
|
-0.137
|
-0.600
|
0.013
|
|
|
|
|
(-1.304, -0.260)
|
(-1.129, 0.854)
|
(-1.692, 0.491)
|
(-1.787, 1.813)
|
|
35-50
|
|
|
-0.677*
|
-0.541
|
-1.195*
|
-1.575
|
|
|
|
|
(-1.282, -0.071)
|
(-1.759, 0.677)
|
(-2.320, -0.069)
|
(-3.567, 0.417)
|
|
50-75
|
|
|
-1.145***
|
-1.191*
|
-1.355*
|
0.097
|
|
|
|
|
(-1.745, -0.546)
|
(-2.328, -0.053)
|
(-2.619, -0.091)
|
(-2.025, 2.219)
|
|
smokeY
|
|
|
|
0.443
|
0.689
|
-0.387
|
|
|
|
|
|
(-0.301, 1.187)
|
(-0.047, 1.425)
|
(-1.612, 0.839)
|
|
drinks10-30
|
|
|
|
-0.366
|
0.071
|
2.575*
|
|
|
|
|
|
(-2.068, 1.335)
|
(-1.594, 1.737)
|
(0.114, 5.035)
|
|
drinks5-10
|
|
|
|
0.262
|
0.193
|
-0.696
|
|
|
|
|
|
(-0.690, 1.215)
|
(-0.856, 1.242)
|
(-2.722, 1.330)
|
|
bmi
|
|
|
|
0.078
|
0.048
|
0.104
|
|
|
|
|
|
(-0.345, 0.501)
|
(-0.356, 0.453)
|
(-0.679, 0.888)
|
|
medcostY
|
|
|
|
|
1.868***
|
3.124***
|
|
|
|
|
|
|
(0.947, 2.789)
|
(1.418, 4.829)
|
|
hlthinsY
|
|
|
|
|
1.144
|
0.848
|
|
|
|
|
|
|
(-0.159, 2.446)
|
(-1.614, 3.310)
|
|
CU2
|
|
|
|
|
-1.252*
|
-1.592
|
|
|
|
|
|
|
(-2.367, -0.138)
|
(-3.504, 0.319)
|
|
CU5
|
|
|
|
|
-0.660
|
-3.497*
|
|
|
|
|
|
|
(-1.907, 0.588)
|
(-6.903, -0.091)
|
|
CU5+
|
|
|
|
|
0.445
|
-0.495
|
|
|
|
|
|
|
(-0.747, 1.637)
|
(-2.190, 1.199)
|
|
ownhome
|
|
|
|
|
|
-1.045
|
|
|
|
|
|
|
|
(-2.653, 0.564)
|
|
safeneighUnsaf
|
|
|
|
|
|
0.251
|
|
|
|
|
|
|
|
(-1.504, 2.006)
|
|
Constant
|
-1.764***
|
-2.089***
|
-1.048
|
-1.090
|
-3.249*
|
3.575
|
|
|
(-1.989, -1.539)
|
(-2.616, -1.561)
|
(-3.070, 0.974)
|
(-4.413, 2.233)
|
(-6.452, -0.045)
|
(-5.258, 12.408)
|
|
N
|
3,143
|
3,100
|
2,811
|
889
|
881
|
258
|
|
Log Likelihood
|
-1,268.652
|
-1,222.661
|
-974.462
|
-293.681
|
-258.416
|
-40.791
|
|
AIC
|
2,545.304
|
2,471.323
|
1,990.924
|
637.362
|
576.833
|
145.581
|
|
|
|
p < .05; p < .01; p < .001
|
car::vif(fit.logit6)
## GVIF Df GVIF^(1/(2*Df))
## marital 21.514784 3 1.667719
## sex 2.199551 1 1.483088
## age 540.611020 5 1.876240
## race 20.224544 3 1.650618
## educ 42.315357 4 1.597027
## income 42.912141 4 1.599825
## smoke 2.364973 1 1.537847
## drinks 9.860776 2 1.772057
## bmi 3.000467 1 1.732186
## medcost 4.652115 1 2.156876
## hlthinsur 5.177888 1 2.275497
## checkup 11.534037 3 1.503131
## ownhome 5.499614 1 2.345126
## safeneigh 2.463188 1 1.569455