Recode Data
library(questionr)
library(dplyr)
ds2<-ds1%>%
mutate(
date = ISOdate(year,month,day = 1),
covid = as.factor(if_else(year==2020 & month>=4 | year==2021 & month<4,1,0)),
covidint = as.factor(if_else(year==2019 & month<=9,"1precov_Q32019",
if_else(year==2019 & month>=10,"2precov_Q42019",
if_else(year==2020 & month<=3,"3precov_Q12020",
if_else(year==2020 & month>=4 & month<=6,"4earlycov_Q22020",
if_else(year==2020 & month>=7 & month<=9,"5midcov_Q32020",
if_else(year==2020 & month>=10,"6latecov_Q42020",
if_else(year==2021 & month<=3,"7latercov_Q12021", "NA")))))))),
covidint=relevel(covidint, ref = "3precov_Q12020"),
covid19unaw = Recode(covidunaw, recodes = "1='0'; 2= '1'; else=NA", as.factor=T),
hhinc =Recode(faminc,recodes="100:490='1_lt15k';500:600='2_15-25k';710:720='3_25-35k';
730:740='4_35-50k';810:830='5_50k-75k';840:843='6_75kplus';999='NA'", as.factor = T),
hhinc2=Recode(faminc,recodes="100:490=1;500:600=2;710:720=3;
730:740=4;810:830=5;840:843=6;999=NA", as.factor = T),
hhinc2=relevel(hhinc2, ref = "1"),
hhincnum=car::Recode(faminc,recodes="100:490=1;500:600=2;710:720=3;
730:740=4;810:830=5;840:843=6;999=NA", as.factor = F),
agegrp=cut(age, breaks=c(16,24,34,44,54,64)),
agegrp=relevel(agegrp, ref = "(44,54]"),
male=Recode(sex,recodes="2='0';1='1';9=NA", as.factor = T),
male=relevel(male, ref = "1"),
raceeth=as.factor(if_else(race==100 & hispan==000,"NH_wht",
if_else(race==200 & hispan==000, "NH_blk",
if_else(!race%in%c('100','200','999') & hispan==000, "NH_other",
if_else(!hispan%in%c('000','901','902'), "Hisp","NA"))))
),
raceeth=relevel(raceeth, ref = "NH_wht"),
employst=Recode( empstat,recodes="1='Mil';10='emp';12='emp_nwklwk';20:22='unemp';32='NILF_utw';34='NILF_oth';36='NILF_ret'; else=NA" , as.factor = T),
empstatus=Recode( empstat,recodes="1:12='emp';20:22='ue';0= NA;else='NILF'", as.factor=T),
unemp=Recode( empstat,recodes="1:12='0';20:22='1';0= NA;else='0'", as.factor=T),
nilf=Recode( empstat,recodes="1:12='0';20:22='0';0= NA;else='1'", as.factor=T),
industry=Recode( ind,recodes="770='const';1070:3990='manuf';4070:4590='whsale';4670:5790='retail';6070:6390='trans';570:690='utl';
6470:6672='media_ent';6870:6992='bnk_ins';7071:7072='realest';7860:7890='Teach';7970:8290='Health';8670:8690='rstrnt_svc';8770:8891='repair_maint';8970:8990='perscare';9160:9390='gen_supprt';9470='crimjust'; else=NA", as.factor = T ),
ind.cd =recode.na(ind,0,as.factor = T),
selfemp=Recode( classwkr,recodes="10:14='1';20:29='0';else=NA",as.factor=T ),
uedurwks =Recode( durunem2,recodes="0:3='lt4wks';4:7='4-10wks';8:10='11-22wks';11:15='23-52wks';16='gt52wks'; else=NA", as.factor = T),
whyue =Recode( whyunemp,recodes="1='templayoff';2='jobloss';3='tempwk_end';4='leftjob';5='re_ent';6='new_ent'; else=NA" , as.factor = T),
whypt =Recode( whyptlwk,recodes="60='onlyfindpt';80='ftdown';100:101='health_med';111='persday';120:122='chcare_fam';
123='sch';130='oth'; else=NA" , as.factor = T),
underemp=Recode( wkstat,recodes="12='1'; 13='1'; 21='1';42='1';99=NA; else='0'" , as.factor=T),
sameemp=Recode( empsame,recodes="2='1';1='0';else=NA", as.factor=T ),
offerwrk=Recode( wrkoffer,recodes= "2='0';1='1';else=NA", as.factor=T),
paidhourly=Recode( paidhour,recodes="2='1';1='0';else=NA", as.factor=T),
unioncover=Recode( union,recodes="2:3='1';1='0';else=NA", as.factor=T),
earnerstud=as.factor(x=eligorg),
ottippay=Recode( otpay,recodes="2='1';1='0';else=NA", as.factor=T),
earnweekna=recode.na(earnweek,9999.99,as.numeric = T),
uhrswrk=recode.na(uhrsworkorg,998:999,as.numeric = T),
earnyear=earnweekna*52,
relate.f = as.factor(x=relate),
cpsidp = as.factor(cpsidp)
)
## Recoded 22663 values to NA.
## Recoded variable contains only numeric characters. Consider using as.numeric = TRUE.
## Recoded 65562 values to NA.
## Recoded 76775 values to NA.
earner<-ds2%>%
filter(earnerstud==1 ) %>%
select(date,year,month,earnyear, earnwt, earnweekna, covidint,covid, agegrp, male, raceeth, industry,paidhourly,ottippay)%>%
filter(complete.cases(.))
# outcomes~ earnyear
HHinc<-ds2%>%
filter(earnerstud==0 & relate.f==101) %>%
select(date,cpsidp, wtfinl, hwtfinl,hhinc2,hhincnum, hhinc,covid19unaw,empstatus,uedurwks,whyue,whypt,underemp, covidint, agegrp, male, raceeth, industry,ind.cd, selfemp,sameemp,unemp,nilf)
# %>% filter(complete.cases(.))
# outcomes~ hhinc;empstatus;uedurwks;underemp
emp<-ds2%>%
filter(earnerstud==0) %>%
select(date,cpsidp, wtfinl, hwtfinl,hhinc2,hhincnum, hhinc,covid19unaw,empstatus,uedurwks,whyue,whypt,underemp, covidint, agegrp, male, raceeth, industry,ind.cd, selfemp,sameemp,unemp,nilf)
#summary(earner)
#summary(HHinc)
#summary(emp)
plot(density(earner$earnweekna))
options(survey.lonely.psu = "adjust")
earner.des<-svydesign(ids=~1, weights=~earnwt, data =earner )
HHinc.des<-svydesign(ids=~1, weights=~hwtfinl, data =HHinc )
emp.des<-svydesign(ids=~1, weights=~wtfinl, data =emp )
fit.logit1<-svyglm(earnweekna~covidint,design= earner.des, family=Gamma(link = "log"))
fit.logit2<-svyglm(earnweekna~covidint+raceeth,design= earner.des, family=Gamma(link = "log"))
fit.logit3<-svyglm(earnweekna~covidint+raceeth+agegrp,design= earner.des, family=Gamma(link = "log"))
fit.logit4<-svyglm(earnweekna~covidint+raceeth+agegrp+male,design= earner.des, family=Gamma(link = "log"))
fit.logit5<-svyglm(earnweekna~covidint+raceeth+agegrp+male+industry,design= earner.des, family=Gamma(link = "log"))
fit.logit6<-svyglm(earnweekna~covidint+raceeth+agegrp+male+industry+paidhourly,design= earner.des, family=Gamma(link = "log"))
fit.logit7<-svyglm(earnweekna~covidint+raceeth+agegrp+male+industry+paidhourly+ottippay,design= earner.des, family=Gamma(link = "log"))
# summary(fit.logit1)
# summary(fit.logit2)
# summary(fit.logit3)
# summary(fit.logit4)
# summary(fit.logit5)
# summary(fit.logit6)
summary(fit.logit7)
##
## Call:
## svyglm(formula = earnweekna ~ covidint + raceeth + agegrp + male +
## industry + paidhourly + ottippay, design = earner.des, family = Gamma(link = "log"))
##
## Survey design:
## svydesign(ids = ~1, weights = ~earnwt, data = earner)
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.376429 0.019859 371.445 < 2e-16 ***
## covidint1precov_Q32019 0.002844 0.016646 0.171 0.864346
## covidint2precov_Q42019 -0.001840 0.015953 -0.115 0.908184
## covidint4earlycov_Q22020 0.031869 0.017242 1.848 0.064591 .
## covidint5midcov_Q32020 0.052610 0.017258 3.048 0.002307 **
## covidint6latecov_Q42020 0.008796 0.016048 0.548 0.583615
## covidint7latercov_Q12021 0.004886 0.019262 0.254 0.799779
## raceethHisp -0.105268 0.010643 -9.891 < 2e-16 ***
## raceethNH_blk -0.069139 0.015322 -4.512 6.49e-06 ***
## raceethNH_other -0.004061 0.018967 -0.214 0.830476
## agegrp(16,24] -0.497231 0.019721 -25.214 < 2e-16 ***
## agegrp(24,34] -0.094043 0.012570 -7.481 8.01e-14 ***
## agegrp(34,44] -0.017770 0.012144 -1.463 0.143420
## agegrp(54,64] -0.039183 0.013130 -2.984 0.002850 **
## male0 -0.138469 0.010431 -13.275 < 2e-16 ***
## industryconst -0.129966 0.019987 -6.502 8.31e-11 ***
## industrycrimjust -0.025630 0.027018 -0.949 0.342835
## industrygen_supprt -0.225474 0.030370 -7.424 1.23e-13 ***
## industryHealth -0.068176 0.017678 -3.856 0.000116 ***
## industrymanuf -0.070687 0.017492 -4.041 5.36e-05 ***
## industrymedia_ent -0.122911 0.052895 -2.324 0.020163 *
## industryperscare -0.544242 0.066503 -8.184 3.11e-16 ***
## industryrealest -0.140771 0.048362 -2.911 0.003614 **
## industryrepair_maint -0.170715 0.034781 -4.908 9.34e-07 ***
## industryretail -0.289303 0.019047 -15.189 < 2e-16 ***
## industryrstrnt_svc -0.512016 0.026336 -19.442 < 2e-16 ***
## industryTeach -0.193188 0.017873 -10.809 < 2e-16 ***
## industrytrans -0.132454 0.022021 -6.015 1.87e-09 ***
## industryutl 0.029384 0.028545 1.029 0.303328
## industrywhsale -0.166577 0.024146 -6.899 5.59e-12 ***
## paidhourly1 -0.371951 0.010017 -37.130 < 2e-16 ***
## ottippay1 0.230499 0.012949 17.801 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for Gamma family taken to be 0.1875096)
##
## Number of Fisher Scoring iterations: 5
library(broom)
fit.logit7%>%
tidy()%>%
mutate(OR = exp(estimate))%>%
knitr::kable(digits = 3)
| (Intercept) |
7.376 |
0.020 |
371.445 |
0.000 |
1597.874 |
| covidint1precov_Q32019 |
0.003 |
0.017 |
0.171 |
0.864 |
1.003 |
| covidint2precov_Q42019 |
-0.002 |
0.016 |
-0.115 |
0.908 |
0.998 |
| covidint4earlycov_Q22020 |
0.032 |
0.017 |
1.848 |
0.065 |
1.032 |
| covidint5midcov_Q32020 |
0.053 |
0.017 |
3.048 |
0.002 |
1.054 |
| covidint6latecov_Q42020 |
0.009 |
0.016 |
0.548 |
0.584 |
1.009 |
| covidint7latercov_Q12021 |
0.005 |
0.019 |
0.254 |
0.800 |
1.005 |
| raceethHisp |
-0.105 |
0.011 |
-9.891 |
0.000 |
0.900 |
| raceethNH_blk |
-0.069 |
0.015 |
-4.512 |
0.000 |
0.933 |
| raceethNH_other |
-0.004 |
0.019 |
-0.214 |
0.830 |
0.996 |
| agegrp(16,24] |
-0.497 |
0.020 |
-25.214 |
0.000 |
0.608 |
| agegrp(24,34] |
-0.094 |
0.013 |
-7.481 |
0.000 |
0.910 |
| agegrp(34,44] |
-0.018 |
0.012 |
-1.463 |
0.143 |
0.982 |
| agegrp(54,64] |
-0.039 |
0.013 |
-2.984 |
0.003 |
0.962 |
| male0 |
-0.138 |
0.010 |
-13.275 |
0.000 |
0.871 |
| industryconst |
-0.130 |
0.020 |
-6.502 |
0.000 |
0.878 |
| industrycrimjust |
-0.026 |
0.027 |
-0.949 |
0.343 |
0.975 |
| industrygen_supprt |
-0.225 |
0.030 |
-7.424 |
0.000 |
0.798 |
| industryHealth |
-0.068 |
0.018 |
-3.856 |
0.000 |
0.934 |
| industrymanuf |
-0.071 |
0.017 |
-4.041 |
0.000 |
0.932 |
| industrymedia_ent |
-0.123 |
0.053 |
-2.324 |
0.020 |
0.884 |
| industryperscare |
-0.544 |
0.067 |
-8.184 |
0.000 |
0.580 |
| industryrealest |
-0.141 |
0.048 |
-2.911 |
0.004 |
0.869 |
| industryrepair_maint |
-0.171 |
0.035 |
-4.908 |
0.000 |
0.843 |
| industryretail |
-0.289 |
0.019 |
-15.189 |
0.000 |
0.749 |
| industryrstrnt_svc |
-0.512 |
0.026 |
-19.442 |
0.000 |
0.599 |
| industryTeach |
-0.193 |
0.018 |
-10.809 |
0.000 |
0.824 |
| industrytrans |
-0.132 |
0.022 |
-6.015 |
0.000 |
0.876 |
| industryutl |
0.029 |
0.029 |
1.029 |
0.303 |
1.030 |
| industrywhsale |
-0.167 |
0.024 |
-6.899 |
0.000 |
0.847 |
| paidhourly1 |
-0.372 |
0.010 |
-37.130 |
0.000 |
0.689 |
| ottippay1 |
0.230 |
0.013 |
17.801 |
0.000 |
1.259 |
AIC(fit.logit1)
## eff.p AIC deltabar
## 1.5838471 3572.3839365 0.2639745
AIC(fit.logit2)
## eff.p AIC deltabar
## 2.3538283 3446.4652372 0.2615365
AIC(fit.logit3)
## eff.p AIC deltabar
## 3.2611366 2997.3715314 0.2508567
AIC(fit.logit4)
## eff.p AIC deltabar
## 3.4640094 2933.5394101 0.2474292
AIC(fit.logit5)
## eff.p AIC deltabar
## 6.596457 2752.426991 0.227464
AIC(fit.logit6)
## eff.p AIC deltabar
## 6.3122839 2515.8024000 0.2104095
AIC(fit.logit7) #************* best model with lowest AIC
## eff.p AIC deltabar
## 6.2626055 2449.8917072 0.2020195