print(paste("Started at", Sys.time()))
## [1] "Started at 2020-07-18 11:02:15"
surv_ver <- "20200707"

dt_resp <- data.table(readRDS(paste0(dir_clean, "survey_worker/homebase_worker_survey_hours_",surv_ver,".rds")))[row_sel==1 & user_base==1 & user_nfirme==1 & firm_base_2wk==1 & st_sel_2wk==1,]
dt_surv <- data.table(readRDS(paste0(dir_clean, "survey_worker/homebase_worker_survey_raw_",surv_ver,".rds")))[row_sel==1,]

dt_resp[,user_sexch:=interaction(user_gender,user_child)]
print(levels(dt_resp$user_sexch))
## [1] "Female.Yes"     "Male.Yes"       "Non-binary.Yes" "Female.No"     
## [5] "Male.No"        "Non-binary.No"
setattr(dt_resp$user_sexch,"levels",c("Female w/ Child","Male w/ Child","Non-binary w/ Child",
                                      "Female w/o Child","Male w/o Child","Non-binary w/o Child"))
table(dt_resp$user_sexch)
## 
##      Female w/ Child        Male w/ Child  Non-binary w/ Child 
##                27175                10831                    7 
##     Female w/o Child       Male w/o Child Non-binary w/o Child 
##                45327                21557                 2013
dt_resp[user_sexch %in% c("Non-binary w/ Child","Non-binary w/o Child"), user_sexch:=NA]
dt_resp[,user_sexch:=droplevels(user_sexch)]
table(dt_resp$user_sexch)
## 
##  Female w/ Child    Male w/ Child Female w/o Child   Male w/o Child 
##            27175            10831            45327            21557
# dt_resp[,user_level:=level]
dt_dms <- readRDS(paste0(dir_clean,"survey_worker/worker_userid_var_sel.rds"))
dt_resp <- merge(dt_resp, dt_dms[,c("userid","user_level")], by=c("userid"), all.x=T)

dt_resp[, wkd:=as.POSIXlt(date)$wday]
lvar <- names(dt_resp)
lvar <- lvar[grepl("userid|user_",lvar) ] #& !grepl("user_race_",lvar)
dt_demogr <- unique(dt_resp[,.SD,.SDcols=lvar])
dt_demogr[, nobs:=1]

setcolorder(dt_demogr,c("userid","user_surv","user_base",
                        "user_gender","user_age","user_race","user_race2","user_marital","user_child","user_child18","user_educ","user_income","user_wage","user_level","user_nfirm","user_sexch","nobs"))
tbl_summary(dt_demogr[, -c("userid","user_surv","user_base","user_race2","user_sexch","user_nfirm")],missing = "no",type = list("user_child"~"categorical"))
Characteristic N = 16881
user_gender
Female 1127 (67%)
Male 514 (31%)
Non-binary 40 (2.4%)
user_age
18-25 611 (36%)
26-37 459 (27%)
38-49 273 (16%)
50-64 259 (15%)
65 or above 77 (4.6%)
user_race
White 1124 (67%)
Black 152 (9.1%)
Hispanic 271 (16%)
Asian 96 (5.8%)
Native American 15 (0.9%)
Pacific Islander 9 (0.5%)
user_marital
Single 928 (55%)
Married 445 (26%)
Living with partner 169 (10%)
Separated 25 (1.5%)
Divorced 97 (5.8%)
Widowed 20 (1.2%)
user_child
Yes 533 (32%)
No 1152 (68%)
user_child18
1 176 (49%)
2 103 (29%)
3 54 (15%)
4 19 (5.3%)
More than 4 5 (1.4%)
user_educ
Some high school 116 (6.9%)
High school graduate 483 (29%)
Two-year degree/some college 569 (34%)
Bachelor's degree 396 (24%)
Master's degree or more 119 (7.1%)
user_income
Less than $15,000 352 (21%)
$15,000-$24,999 363 (22%)
$25,000-$34,999 249 (15%)
$35,000-$44,999 144 (8.7%)
$45,000-$54,999 118 (7.1%)
$55,000-$64,999 102 (6.2%)
$65,000-$74,999 76 (4.6%)
$75,000-$84,999 45 (2.7%)
More than $85,000 205 (12%)
user_wage
$5-$7.49 118 (7.0%)
$7.50-$9.99 252 (15%)
$10-$12.49 449 (27%)
$12.50-$14.99 350 (21%)
$15-$17.49 265 (16%)
$17.50-$19.99 81 (4.8%)
$20 -$22.49 68 (4.0%)
$22.50-$24.99 30 (1.8%)
$25 or higher 69 (4.1%)
user_level
Employee 1589 (94%)
Manager 78 (4.6%)
General Manager 21 (1.2%)
nobs 1688 (100%)
user_race_h 273 (100%)
user_race_i 20 (100%)
user_race_a 99 (100%)
user_race_p 10 (100%)
user_race_b 152 (100%)
user_race_w 1134 (100%)
user_nfirme 1688 (100%)

1 Statistics presented: n (%) 

dt_info <- data.table(var=c("gender","age","marital","child","child18","race","income","wage","educ","level"),
                      vars=c("Gender","Age","Marital Status","Have Children","Number of Children (if any)","Race","Household Income","Wage in Jan 2020","Education","Job Title"))

# Alternative Race and Ethnicity, and Manager
dt_info <- rbind(dt_info, data.table(var=c("racee","ethn","manager"),vars=c("Race","Ethnicity","Job Title")))

lrace <- factor(c(1:4))
levels(lrace) <- c("White","Black","Asian or Pacific Islander","American Indian or Alaskan Native")
dt_demogr[user_race_w==1 & is.na(user_race_b) & is.na(user_race_i) & is.na(user_race_a) & is.na(user_race_p),user_racee:=lrace[1]]
dt_demogr[is.na(user_racee) & user_race_b==1,user_racee:=lrace[2]]
dt_demogr[is.na(user_racee) & user_race_a==1 | user_race_p==1,user_racee:=lrace[3]]
dt_demogr[is.na(user_racee) & user_race_i==1,user_racee:=lrace[4]]

lethn <- factor(c(1:2))
levels(lethn) <- c("Hispanic","Non-Hispanic")
dt_demogr[user_race_h==1 ,user_ethn:=lethn[1]]
dt_demogr[is.na(user_ethn) & (!is.na(user_race_w) | !is.na(user_race_b) | !is.na(user_race_i) | !is.na(user_race_a) | !is.na(user_race_p)),user_ethn:=lethn[2]]

lpos <- factor(c(1:2))
levels(lpos) <- c("Employee","Manager")
dt_demogr[user_level %in% c("Employee"),user_pos:=lpos[1]]
dt_demogr[user_level %in% c("Manager","General Manager"),user_pos:=lpos[2]]

vvar <- "gender"
dt_tabr <- data.table()
for (vvar in c("gender","age","racee","ethn","marital","child","child18","educ","income","wage","pos")) {
  vvar_by <- paste0("user_",vvar)
  dt_sum <- dt_demogr[,.(nobs=.N),by=c(vvar_by)]
  setorderv(dt_sum, c(vvar_by))
  setattr(dt_sum[[vvar_by]],"levels",c(levels(dt_sum[[vvar_by]]),"NA"))
  setnames(dt_sum,vvar_by,"val")
  dt_sum[is.na(val),val:="NA"]
  dt_sum <- dt_sum[val!="NA",] # Remove NA from table
  dt_sum[, sobs:=percent(nobs/sum(nobs),accuracy = 0.1)]
  dt_sum <- dt_sum[,c("val","sobs")]
  dt_sum <- dt_sum[, lapply(.SD, as.character)]
  dt_tabr <- rbind(dt_tabr,data.table(val=dt_info[var==vvar,vars]),dt_sum[,var:=vvar],fill=T)
}

ibreak <- 34
dt_tab <- merge(dt_tabr[c(1:ibreak),c("val","sobs")][,row:=.I],dt_tabr[c((ibreak+1):nrow(dt_tabr)),c("val","sobs")][,row:=.I], by="row", all=T)
fwrite(dt_tab, paste0(dir_otab,"P2_surv_demogr_sum.csv"))
print(paste("Ended at", Sys.time()))
## [1] "Ended at 2020-07-18 11:02:27"
# End of R script