Creating weight factors for all datasets per year

In this script, I calculate weight factors for each dataset and each wave. The target values I use as input are derived from Statistics Netherlands. The Excel file with these target values can be found in this replication package. I will weigh for age, gender, urbanity and marital status.

rm(list = ls())
#install.packages("anesrake")
library(anesrake)
library(tidyverse)
library(here)
library(kableExtra) # needed later to make an example table that compares the proportions
set.seed(1)
# I use the here package so that I do not have to load files from my personal documents. However, here does not seem to be compatible with the (d)plyr package. That is why in previous scripts you may have seen that I did load from my local device.
here()
here::i_am("scripts/analysis/single_regression.Rmd")

ISSP

#Create a weight variable for each dataset, which I can later use in the regression analyses to weigh the data with
load(here("./data/all_waves", "issp93sel.RData"))

#Create variables in the environment with the target values
sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(issp93sel$age)
issp93sel$age_cat[issp93sel$age >=15 & issp93sel$age <= 30] <- 1
issp93sel$age_cat[issp93sel$age >=31 & issp93sel$age <= 45] <- 2
issp93sel$age_cat[issp93sel$age >=46 & issp93sel$age <= 60] <- 3
issp93sel$age_cat[issp93sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.25, 0.24, 0.17, 0.17)

#You notice here that I only weigh for sex and age, because information on urbanity and marital status was not yet available for 1993
targets <- list(sex, age_cat)
names(targets) <- c("sex", "age_cat")

issp93sel$caseid <- 1:length(issp93sel$sex)

#To use anesrakefinder, the variables in the original dataset have to be numeric
issp93sel$sex <- as.numeric(issp93sel$sex)

#Anesrakefinder weighs the data and provides a summary of this process
anesrakefinder(targets, issp93sel, choosemethod = "total")

outsave <- anesrake(targets, issp93sel, caseid = issp93sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

#Assign the weights to each person in the new variable weightvec, which can later be used in the regression analysis
issp93sel$weightvec  <- unlist(outsave[1])

save(issp93sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/issp93sel.Rdata")

#Repeat the same for the wave in 2000
load(here("./data/all_waves", "issp00sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(issp00sel$age)
issp00sel$age_cat[issp00sel$age >=15 & issp00sel$age <= 30] <- 1
issp00sel$age_cat[issp00sel$age >=31 & issp00sel$age <= 45] <- 2
issp00sel$age_cat[issp00sel$age >=46 & issp00sel$age <= 60] <- 3
issp00sel$age_cat[issp00sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.21, 0.24, 0.19, 0.17)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.42, 0.17, 0.41)
marstat <- c('1', '2')
marstat <- c(0.59, 0.41)

targets <- list(sex, age_cat, urban, marstat)
names(targets) <- c("sex", "age_cat", "urban", "marstat")

issp00sel$caseid <- 1:length(issp00sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
issp00sel$sex <- as.numeric(issp00sel$sex)
issp00sel$urban <- as.numeric(issp00sel$urban)
issp00sel$marstat <- as.numeric(issp00sel$marstat)

anesrakefinder(targets, issp00sel, choosemethod = "total")

outsave <- anesrake(targets, issp00sel, caseid = issp00sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

issp00sel$weightvec  <- unlist(outsave[1])

save(issp00sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/issp00sel.Rdata")

# Create a table that compares the weighted and unweighted proportions and the target values, using the ISSP 2000 as an example. 

# Create variables
sex_unweight <- wpct(issp00sel$sex)
sex_weight <- wpct(issp00sel$sex, issp00sel$weightvec)

age_unweight <- wpct(issp00sel$age_cat)
age_weight <- wpct(issp00sel$age_cat, issp00sel$weightvec)

urban_unweight <- wpct(issp00sel$urban)
urban_weight <- wpct(issp00sel$urban, issp00sel$weightvec)

marstat_unweight <- wpct(issp00sel$marstat)
marstat_weight <- wpct(issp00sel$marstat, issp00sel$weightvec)

# Put the information in a df per variable
table_sex <- data.frame(Variable = c("Gender", ""),
  Category = c("Male", "Female"),
  Unweighted = paste(round(sex_unweight * 100, 1), "%"),
  Weighted = paste(round(sex_weight * 100, 1), "%")
)

table_mar <- data.frame(Variable = c("Marital Status", ""),
  Category = c("Living together", "Not living together"),
  Unweighted = paste(round(marstat_unweight * 100, 1), "%"),
  Weighted = paste(round(marstat_weight * 100, 1), "%")
)

table_urban <- data.frame(Variable = c("Urbanity", "", ""),
  Category = c("Low urbanity", "Medium urbanity", "High urbanity"),
  Unweighted = paste(round(urban_unweight * 100, 1), "%"),
  Weighted = paste(round(urban_weight * 100, 1), "%")
)

table_age <- data.frame(Variable = c("Age category", "", "", ""),
  Category = c("Under 30", "31 to 45", "46 to 61", "61 and older"),
  Unweighted = paste(round(age_unweight * 100, 1), "%"),
  Weighted = paste(round(age_weight * 100, 1), "%")
)

# Create a df with all target weights
targetweights <- data.frame(Variable = c("Gender", "", "Marital Status", "", "Urbanity", "", "", "Age category", "", "", ""),
Category = c("Male", "Female", "Living together", "Not living together", "Low urbanity", "Medium urbanity", "High urbanity", "Under 30", "31 to 45", "46 to 61", "61 and older"), 
Target = c("49.4%","50.6%", "58.8%", "41.2%", "41.7%", "17.5%", "40.9%", "20.9%", "24.0%", "19.3%", "17.1%"))

# Bind these tables in 2 steps, as targetweights has a different row number than the other dfs. 
table_total <- rbind(table_sex, table_mar, table_urban, table_age)
table_total <- left_join(table_total, targetweights)

# Make and save the table as html
weight_table <- kable(table_total, format = "html", align = "l",
      caption = "Appendix B. Proportions weighted and unweighted for ISSP 2000") %>%
  kable_classic_2(full_width = FALSE) %>%
  save_kable("/Users/anuschka/Documents/climatechange/climatechange/output/appendix/weight_table.html")

# And lastly for 2010
load(here("./data/all_waves", "issp10sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(issp10sel$age)
issp10sel$age_cat[issp10sel$age >=15 & issp10sel$age <= 30] <- 1
issp10sel$age_cat[issp10sel$age >=31 & issp10sel$age <= 45] <- 2
issp10sel$age_cat[issp10sel$age >=46 & issp10sel$age <= 60] <- 3
issp10sel$age_cat[issp10sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.19, 0.21, 0.21, 0.20)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.38, 0.18, 0.44)
marstat <- c('1', '2')
marstat <- c(0.55, 0.45)

targets <- list(sex, age_cat, urban, marstat)
names(targets) <- c("sex", "age_cat", "urban", "marstat")

issp10sel$caseid <- 1:length(issp10sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
issp10sel$sex <- as.numeric(issp10sel$sex)
issp10sel$urban <- as.numeric(issp10sel$urban)
issp10sel$marstat <- as.numeric(issp10sel$marstat)

anesrakefinder(targets, issp10sel, choosemethod = "total")

outsave <- anesrake(targets, issp10sel, caseid = issp10sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

issp10sel$weightvec  <- unlist(outsave[1])

save(issp10sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/issp10sel.Rdata")

ESS

# And now for ESS, repeat the same measure
load(here("./data/all_waves", "ess16sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(ess16sel$age)
ess16sel$age_cat[ess16sel$age >=15 & ess16sel$age <= 30] <- 1
ess16sel$age_cat[ess16sel$age >=31 & ess16sel$age <= 45] <- 2
ess16sel$age_cat[ess16sel$age >=46 & ess16sel$age <= 60] <- 3
ess16sel$age_cat[ess16sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.22, 0.23)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.35, 0.17, 0.48)
marstat <- c('1', '2')
marstat <- c(0.53, 0.47)

targets <- list(sex, age_cat, urban)
names(targets) <- c("sex", "age_cat", "urban")

ess16sel$caseid <- 1:length(ess16sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
ess16sel$sex <- as.numeric(ess16sel$sex)
ess16sel$urban <- as.numeric(ess16sel$urban)

anesrakefinder(targets, ess16sel, choosemethod = "total")

outsave <- anesrake(targets, ess16sel, caseid = ess16sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

ess16sel$weightvec  <- unlist(outsave[1])

save(ess16sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/ess16sel.Rdata")

#And for ess2020
load(here("./data/all_waves", "ess20sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(ess20sel$age)
ess20sel$age_cat[ess20sel$age >=15 & ess20sel$age <= 30] <- 1
ess20sel$age_cat[ess20sel$age >=31 & ess20sel$age <= 45] <- 2
ess20sel$age_cat[ess20sel$age >=46 & ess20sel$age <= 60] <- 3
ess20sel$age_cat[ess20sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.21, 0.25)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.34, 0.17, 0.49)

targets <- list(sex, age_cat, urban)
names(targets) <- c("sex", "age_cat", "urban")

ess20sel$caseid <- 1:length(ess20sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
ess20sel$sex <- as.numeric(ess20sel$sex)
ess20sel$urban <- as.numeric(ess20sel$urban)

anesrakefinder(targets, ess20sel, choosemethod = "total")

outsave <- anesrake(targets, ess20sel, caseid = ess20sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

ess20sel$weightvec  <- unlist(outsave[1])

save(ess20sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/ess20sel.Rdata")

EVS

# Then the EVS, which I have to split up because I use the compiled dataset
load(here("./data/final_data", "evssel.RData"))

# First select only the observations of 1990
evs90 <- subset(evssel, surveyyear == 1990)
sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(evs90$age)
evs90$age_cat[evs90$age >=15 & evs90$age <= 30] <- 1
evs90$age_cat[evs90$age >=31 & evs90$age <= 45] <- 2
evs90$age_cat[evs90$age >=46 & evs90$age <= 60] <- 3
evs90$age_cat[evs90$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.25, 0.24, 0.16, 0.16)

targets <- list(sex, age_cat)
names(targets) <- c("sex", "age_cat")

evs90$caseid <- 1:length(evs90$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
evs90$sex <- as.numeric(evs90$sex)

anesrakefinder(targets, evs90, choosemethod = "total")

outsave <- anesrake(targets, evs90, caseid = evs90$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

evs90$weightvec  <- unlist(outsave[1])

# Then those of 1999
evs99 <- subset(evssel, surveyyear == 1999)
sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(evs99$age)
evs99$age_cat[evs99$age >=15 & evs99$age <= 30] <- 1
evs99$age_cat[evs99$age >=31 & evs99$age <= 45] <- 2
evs99$age_cat[evs99$age >=46 & evs99$age <= 60] <- 3
evs99$age_cat[evs99$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.21, 0.24, 0.19, 0.17)

marstat <- c('1', '2')
marstat <- c(0.59, 0.41)

targets <- list(sex, age_cat, marstat)
names(targets) <- c("sex", "age_cat", "marstat")

# To use anesrakefinder, the variables in the original dataset have to be numeric
evs99$sex <- as.numeric(evs99$sex)
evs99$marstat <- as.numeric(evs99$marstat)

evs99$caseid <- 1:length(evs99$sex)

anesrakefinder(targets, evs99, choosemethod = "total")

outsave <- anesrake(targets, evs99, caseid = evs99$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

evs99$weightvec  <- unlist(outsave[1])

# And lastly of 2008
evs2008 <- subset(evssel, surveyyear == 2008)
sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(evs2008$age)
evs2008$age_cat[evs2008$age >=15 & evs2008$age <= 30] <- 1
evs2008$age_cat[evs2008$age >=31 & evs2008$age <= 45] <- 2
evs2008$age_cat[evs2008$age >=46 & evs2008$age <= 60] <- 3
evs2008$age_cat[evs2008$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.19, 0.22, 0.21, 0.19)

marstat <- c('1', '2')
marstat <- c(0.56, 0.44)

targets <- list(sex, age_cat, marstat)
names(targets) <- c("sex", "age_cat", "marstat")

# To use anesrakefinder, the variables in the original dataset have to be numeric
evs2008$sex <- as.numeric(evs2008$sex)
evs2008$marstat <- as.numeric(evs2008$marstat)

evs2008$caseid <- 1:length(evs2008$sex)

anesrakefinder(targets, evs2008, choosemethod = "total")

outsave <- anesrake(targets, evs2008, caseid = evs2008$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

evs2008$weightvec  <- unlist(outsave[1])

#Turn marital status back to original because it is not a numeric variable
evs90$marstat <- as.factor(evs90$marstat)
evs99$marstat <- as.factor(evs99$marstat)
evs2008$marstat <- as.factor(evs2008$marstat)

# Bind the three together
evssel <- dplyr::bind_rows(evs90, evs99, evs2008)

#Save the total to which now the weights are added per year
save(evssel, file="/Users/anuschka/Documents/climatechange/climatechange//data/final_data/evssel.Rdata")

I&O research

# I & O Research
load(here("./data/all_waves", "io2019sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(io2019sel$age)
io2019sel$age_cat[io2019sel$age >=15 & io2019sel$age <= 30] <- 1
io2019sel$age_cat[io2019sel$age >=31 & io2019sel$age <= 45] <- 2
io2019sel$age_cat[io2019sel$age >=46 & io2019sel$age <= 60] <- 3
io2019sel$age_cat[io2019sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.22, 0.24)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.34, 0.17, 0.49)
marstat <- c('1', '2')
marstat <- c(0.52, 0.48)

targets <- list(sex, age_cat, urban, marstat)
names(targets) <- c("sex", "age_cat", "urban", "marstat")

io2019sel$caseid <- 1:length(io2019sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
io2019sel$sex <- as.numeric(io2019sel$sex)
io2019sel$urban <- as.numeric(io2019sel$urban)
io2019sel$marstat <- as.numeric(io2019sel$marstat)

anesrakefinder(targets, io2019sel, choosemethod = "total")

outsave <- anesrake(targets, io2019sel, caseid = io2019sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

io2019sel$weightvec  <- unlist(outsave[1])

save(io2019sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/io2019sel.Rdata")

# Io2020

load(here("./data/all_waves", "io2020sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(io2020sel$age)
io2020sel$age_cat[io2020sel$age >=15 & io2020sel$age <= 30] <- 1
io2020sel$age_cat[io2020sel$age >=31 & io2020sel$age <= 45] <- 2
io2020sel$age_cat[io2020sel$age >=46 & io2020sel$age <= 60] <- 3
io2020sel$age_cat[io2020sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.21, 0.25)

marstat <- c('1', '2')
marstat <- c(0.52, 0.48)

targets <- list(sex, age_cat, marstat)
names(targets) <- c("sex", "age_cat", "marstat")

io2020sel$caseid <- 1:length(io2020sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
io2020sel$sex <- as.numeric(io2020sel$sex)
io2020sel$marstat <- as.numeric(io2020sel$marstat)

anesrakefinder(targets, io2020sel, choosemethod = "total")

outsave <- anesrake(targets, io2020sel, caseid = io2020sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

io2020sel$weightvec  <- unlist(outsave[1])

save(io2020sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/io2020sel.Rdata")

#Io 2022
load(here("./data/all_waves", "io2022sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(io2022sel$age)
io2022sel$age_cat[io2022sel$age >=15 & io2022sel$age <= 30] <- 1
io2022sel$age_cat[io2022sel$age >=31 & io2022sel$age <= 45] <- 2
io2022sel$age_cat[io2022sel$age >=46 & io2022sel$age <= 60] <- 3
io2022sel$age_cat[io2022sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.21, 0.25)

marstat <- c('1', '2')
marstat <- c(0.51, 0.49)

targets <- list(sex, age_cat, marstat)
names(targets) <- c("sex", "age_cat", "marstat")

io2022sel$caseid <- 1:length(io2022sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
io2022sel$sex <- as.numeric(io2022sel$sex)
io2022sel$marstat <- as.numeric(io2022sel$marstat)

anesrakefinder(targets, io2022sel, choosemethod = "total")

outsave <- anesrake(targets, io2022sel, caseid = io2022sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

io2022sel$weightvec  <- unlist(outsave[1])

save(io2022sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/io2022sel.Rdata")

EB

#Lastly Eurobarometer, which has a lot of waves and thus a lot of repetition. Still it is difficult to write a loop, as targetweights differ each year. So Im just repeating the script many times. 
load(here("./data/all_waves", "eb1986sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(eb1986sel$age)
eb1986sel$age_cat[eb1986sel$age >=15 & eb1986sel$age <= 30] <- 1
eb1986sel$age_cat[eb1986sel$age >=31 & eb1986sel$age <= 45] <- 2
eb1986sel$age_cat[eb1986sel$age >=46 & eb1986sel$age <= 60] <- 3
eb1986sel$age_cat[eb1986sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.25, 0.24, 0.16, 0.16)

targets <- list(sex, age_cat)
names(targets) <- c("sex", "age_cat")

eb1986sel$caseid <- 1:length(eb1986sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb1986sel$sex <- as.numeric(eb1986sel$sex)

anesrakefinder(targets, eb1986sel, choosemethod = "total")

outsave <- anesrake(targets, eb1986sel, caseid = eb1986sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb1986sel$weightvec  <- unlist(outsave[1])

save(eb1986sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb1986sel.Rdata")

# Eb 1992
load(here("./data/all_waves", "eb1992sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(eb1992sel$age)
eb1992sel$age_cat[eb1992sel$age >=15 & eb1992sel$age <= 30] <- 1
eb1992sel$age_cat[eb1992sel$age >=31 & eb1992sel$age <= 45] <- 2
eb1992sel$age_cat[eb1992sel$age >=46 & eb1992sel$age <= 60] <- 3
eb1992sel$age_cat[eb1992sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.25, 0.24, 0.24, 0.16)

targets <- list(sex, age_cat)
names(targets) <- c("sex", "age_cat")

eb1992sel$caseid <- 1:length(eb1992sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb1992sel$sex <- as.numeric(eb1992sel$sex)

anesrakefinder(targets, eb1992sel, choosemethod = "total")

outsave <- anesrake(targets, eb1992sel, caseid = eb1992sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb1992sel$weightvec  <- unlist(outsave[1])

save(eb1992sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb1992sel.Rdata")

# Eb 1995
load(here("./data/all_waves", "eb1995sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(eb1995sel$age)
eb1995sel$age_cat[eb1995sel$age >=15 & eb1995sel$age <= 30] <- 1
eb1995sel$age_cat[eb1995sel$age >=31 & eb1995sel$age <= 45] <- 2
eb1995sel$age_cat[eb1995sel$age >=46 & eb1995sel$age <= 60] <- 3
eb1995sel$age_cat[eb1995sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.24, 0.24, 0.18, 0.17)

marstat <- c('1', '2')
marstat <- c(0.6, 0.4)

targets <- list(sex, age_cat, marstat)
names(targets) <- c("sex", "age_cat", "marstat")

eb1995sel$caseid <- 1:length(eb1995sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb1995sel$sex <- as.numeric(eb1995sel$sex)
eb1995sel$marstat <- as.numeric(eb1995sel$marstat)

anesrakefinder(targets, eb1995sel, choosemethod = "total")

outsave <- anesrake(targets, eb1995sel, caseid = eb1995sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb1995sel$weightvec  <- unlist(outsave[1])

save(eb1995sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb1995sel.Rdata")

# 2004
load(here("./data/all_waves", "eb2004sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(eb2004sel$age)
eb2004sel$age_cat[eb2004sel$age >=15 & eb2004sel$age <= 30] <- 1
eb2004sel$age_cat[eb2004sel$age >=31 & eb2004sel$age <= 45] <- 2
eb2004sel$age_cat[eb2004sel$age >=46 & eb2004sel$age <= 60] <- 3
eb2004sel$age_cat[eb2004sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.24, 0.20, 0.18)

marstat <- c('1', '2')
marstat <- c(0.58, 0.42)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.41, 0.18, 0.42)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2004sel$caseid <- 1:length(eb2004sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2004sel$sex <- as.numeric(eb2004sel$sex)
eb2004sel$marstat <- as.numeric(eb2004sel$marstat)
eb2004sel$urban <- as.numeric(eb2004sel$urban)

anesrakefinder(targets, eb2004sel, choosemethod = "total")

outsave <- anesrake(targets, eb2004sel, caseid = eb2004sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2004sel$weightvec  <- unlist(outsave[1])

save(eb2004sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2004sel.Rdata")

# 2007
load(here("./data/all_waves", "eb2007sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(eb2007sel$age)
eb2007sel$age_cat[eb2007sel$age >=15 & eb2007sel$age <= 30] <- 1
eb2007sel$age_cat[eb2007sel$age >=31 & eb2007sel$age <= 45] <- 2
eb2007sel$age_cat[eb2007sel$age >=46 & eb2007sel$age <= 60] <- 3
eb2007sel$age_cat[eb2007sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.19, 0.23, 0.21, 0.19)

marstat <- c('1', '2')
marstat <- c(0.56, 0.43)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.40, 0.18, 0.42)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2007sel$caseid <- 1:length(eb2007sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2007sel$sex <- as.numeric(eb2007sel$sex)
eb2007sel$marstat <- as.numeric(eb2007sel$marstat)
eb2007sel$urban <- as.numeric(eb2007sel$urban)

anesrakefinder(targets, eb2007sel, choosemethod = "total")

outsave <- anesrake(targets, eb2007sel, caseid = eb2007sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2007sel$weightvec  <- unlist(outsave[1])

save(eb2007sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2007sel.Rdata")

# 2008
load(here("./data/all_waves", "eb2008sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(eb2008sel$age)
eb2008sel$age_cat[eb2008sel$age >=15 & eb2008sel$age <= 30] <- 1
eb2008sel$age_cat[eb2008sel$age >=31 & eb2008sel$age <= 45] <- 2
eb2008sel$age_cat[eb2008sel$age >=46 & eb2008sel$age <= 60] <- 3
eb2008sel$age_cat[eb2008sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.19, 0.22, 0.21, 0.19)

marstat <- c('1', '2')
marstat <- c(0.56, 0.44)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.39, 0.18, 0.42)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2008sel$caseid <- 1:length(eb2008sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2008sel$sex <- as.numeric(eb2008sel$sex)
eb2008sel$marstat <- as.numeric(eb2008sel$marstat)
eb2008sel$urban <- as.numeric(eb2008sel$urban)

anesrakefinder(targets, eb2008sel, choosemethod = "total")

outsave <- anesrake(targets, eb2008sel, caseid = eb2008sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2008sel$weightvec  <- unlist(outsave[1])

save(eb2008sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2008sel.Rdata")

# 2009
load(here("./data/all_waves", "eb2009asel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(eb2009asel$age)
eb2009asel$age_cat[eb2009asel$age >=15 & eb2009asel$age <= 30] <- 1
eb2009asel$age_cat[eb2009asel$age >=31 & eb2009asel$age <= 45] <- 2
eb2009asel$age_cat[eb2009asel$age >=46 & eb2009asel$age <= 60] <- 3
eb2009asel$age_cat[eb2009asel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.19, 0.22, 0.21, 0.20)

marstat <- c('1', '2')
marstat <- c(0.56, 0.44)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.39, 0.18, 0.43)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2009asel$caseid <- 1:length(eb2009asel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2009asel$sex <- as.numeric(eb2009asel$sex)
eb2009asel$marstat <- as.numeric(eb2009asel$marstat)
eb2009asel$urban <- as.numeric(eb2009asel$urban)

anesrakefinder(targets, eb2009asel, choosemethod = "total")

outsave <- anesrake(targets, eb2009asel, caseid = eb2009asel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2009asel$weightvec  <- unlist(outsave[1])

save(eb2009asel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2009asel.Rdata")

# 2009b
load(here("./data/all_waves", "eb2009bsel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(eb2009bsel$age)
eb2009bsel$age_cat[eb2009bsel$age >=15 & eb2009bsel$age <= 30] <- 1
eb2009bsel$age_cat[eb2009bsel$age >=31 & eb2009bsel$age <= 45] <- 2
eb2009bsel$age_cat[eb2009bsel$age >=46 & eb2009bsel$age <= 60] <- 3
eb2009bsel$age_cat[eb2009bsel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.19, 0.22, 0.21, 0.20)

marstat <- c('1', '2')
marstat <- c(0.56, 0.44)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.39, 0.18, 0.43)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2009bsel$caseid <- 1:length(eb2009bsel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2009bsel$sex <- as.numeric(eb2009bsel$sex)
eb2009bsel$marstat <- as.numeric(eb2009bsel$marstat)
eb2009bsel$urban <- as.numeric(eb2009bsel$urban)

anesrakefinder(targets, eb2009bsel, choosemethod = "total")

outsave <- anesrake(targets, eb2009bsel, caseid = eb2009bsel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2009bsel$weightvec  <- unlist(outsave[1])

save(eb2009bsel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2009bsel.Rdata")

# 2011
load(here("./data/all_waves", "eb2011asel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(eb2011asel$age)
eb2011asel$age_cat[eb2011asel$age >=15 & eb2011asel$age <= 30] <- 1
eb2011asel$age_cat[eb2011asel$age >=31 & eb2011asel$age <= 45] <- 2
eb2011asel$age_cat[eb2011asel$age >=46 & eb2011asel$age <= 60] <- 3
eb2011asel$age_cat[eb2011asel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.21, 0.21, 0.21)

marstat <- c('1', '2')
marstat <- c(0.55, 0.45)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.38, 0.18, 0.44)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2011asel$caseid <- 1:length(eb2011asel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2011asel$sex <- as.numeric(eb2011asel$sex)
eb2011asel$marstat <- as.numeric(eb2011asel$marstat)
eb2011asel$urban <- as.numeric(eb2011asel$urban)

anesrakefinder(targets, eb2011asel, choosemethod = "total")

outsave <- anesrake(targets, eb2011asel, caseid = eb2011asel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2011asel$weightvec  <- unlist(outsave[1])

save(eb2011asel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2011asel.Rdata")

#2011b
load(here("./data/all_waves", "eb2011bsel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.49, 0.51)

attributes(eb2011bsel$age)
eb2011bsel$age_cat[eb2011bsel$age >=15 & eb2011bsel$age <= 30] <- 1
eb2011bsel$age_cat[eb2011bsel$age >=31 & eb2011bsel$age <= 45] <- 2
eb2011bsel$age_cat[eb2011bsel$age >=46 & eb2011bsel$age <= 60] <- 3
eb2011bsel$age_cat[eb2011bsel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.21, 0.21, 0.21)

marstat <- c('1', '2')
marstat <- c(0.55, 0.45)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.38, 0.18, 0.44)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2011bsel$caseid <- 1:length(eb2011bsel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2011bsel$sex <- as.numeric(eb2011bsel$sex)
eb2011bsel$marstat <- as.numeric(eb2011bsel$marstat)
eb2011bsel$urban <- as.numeric(eb2011bsel$urban)

anesrakefinder(targets, eb2011bsel, choosemethod = "total")

outsave <- anesrake(targets, eb2011bsel, caseid = eb2011bsel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2011bsel$weightvec  <- unlist(outsave[1])

save(eb2011bsel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2011bsel.Rdata")

# 2013
load(here("./data/all_waves", "eb2013sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(eb2013sel$age)
eb2013sel$age_cat[eb2013sel$age >=15 & eb2013sel$age <= 30] <- 1
eb2013sel$age_cat[eb2013sel$age >=31 & eb2013sel$age <= 45] <- 2
eb2013sel$age_cat[eb2013sel$age >=46 & eb2013sel$age <= 60] <- 3
eb2013sel$age_cat[eb2013sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.20, 0.21, 0.22)

marstat <- c('1', '2')
marstat <- c(0.54, 0.46)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.37, 0.18, 0.45)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2013sel$caseid <- 1:length(eb2013sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2013sel$sex <- as.numeric(eb2013sel$sex)
eb2013sel$marstat <- as.numeric(eb2013sel$marstat)
eb2013sel$urban <- as.numeric(eb2013sel$urban)

anesrakefinder(targets, eb2013sel, choosemethod = "total")

outsave <- anesrake(targets, eb2013sel, caseid = eb2013sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2013sel$weightvec  <- unlist(outsave[1])

save(eb2013sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2013sel.Rdata")

# 2014
load(here("./data/all_waves", "eb2014sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(eb2014sel$age)
eb2014sel$age_cat[eb2014sel$age >=15 & eb2014sel$age <= 30] <- 1
eb2014sel$age_cat[eb2014sel$age >=31 & eb2014sel$age <= 45] <- 2
eb2014sel$age_cat[eb2014sel$age >=46 & eb2014sel$age <= 60] <- 3
eb2014sel$age_cat[eb2014sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.20, 0.22, 0.22)

marstat <- c('1', '2')
marstat <- c(0.54, 0.46)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.37, 0.18, 0.45)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2014sel$caseid <- 1:length(eb2014sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2014sel$sex <- as.numeric(eb2014sel$sex)
eb2014sel$marstat <- as.numeric(eb2014sel$marstat)
eb2014sel$urban <- as.numeric(eb2014sel$urban)

anesrakefinder(targets, eb2014sel, choosemethod = "total")

outsave <- anesrake(targets, eb2014sel, caseid = eb2014sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2014sel$weightvec  <- unlist(outsave[1])

save(eb2014sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2014sel.Rdata")

# 2015
load(here("./data/all_waves", "eb2015sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(eb2015sel$age)
eb2015sel$age_cat[eb2015sel$age >=15 & eb2015sel$age <= 30] <- 1
eb2015sel$age_cat[eb2015sel$age >=31 & eb2015sel$age <= 45] <- 2
eb2015sel$age_cat[eb2015sel$age >=46 & eb2015sel$age <= 60] <- 3
eb2015sel$age_cat[eb2015sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.20, 0.22, 0.22)

marstat <- c('1', '2')
marstat <- c(0.54, 0.46)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.37, 0.18, 0.45)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2015sel$caseid <- 1:length(eb2015sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2015sel$sex <- as.numeric(eb2015sel$sex)
eb2015sel$marstat <- as.numeric(eb2015sel$marstat)
eb2015sel$urban <- as.numeric(eb2015sel$urban)

anesrakefinder(targets, eb2015sel, choosemethod = "total")

outsave <- anesrake(targets, eb2015sel, caseid = eb2015sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2015sel$weightvec  <- unlist(outsave[1])

save(eb2015sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2015sel.Rdata")

#2017a
load(here("./data/all_waves", "eb2017asel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(eb2017asel$age)
eb2017asel$age_cat[eb2017asel$age >=15 & eb2017asel$age <= 30] <- 1
eb2017asel$age_cat[eb2017asel$age >=31 & eb2017asel$age <= 45] <- 2
eb2017asel$age_cat[eb2017asel$age >=46 & eb2017asel$age <= 60] <- 3
eb2017asel$age_cat[eb2017asel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.19, 0.22, 0.23)

marstat <- c('1', '2')
marstat <- c(0.53, 0.47)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.35, 0.17, 0.48)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2017asel$caseid <- 1:length(eb2017asel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2017asel$sex <- as.numeric(eb2017asel$sex)
eb2017asel$marstat <- as.numeric(eb2017asel$marstat)
eb2017asel$urban <- as.numeric(eb2017asel$urban)

anesrakefinder(targets, eb2017asel, choosemethod = "total")

outsave <- anesrake(targets, eb2017asel, caseid = eb2017asel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2017asel$weightvec  <- unlist(outsave[1])

save(eb2017asel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2017asel.Rdata")

#2017b
load(here("./data/all_waves", "eb2017bsel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(eb2017bsel$age)
eb2017bsel$age_cat[eb2017bsel$age >=15 & eb2017bsel$age <= 30] <- 1
eb2017bsel$age_cat[eb2017bsel$age >=31 & eb2017bsel$age <= 45] <- 2
eb2017bsel$age_cat[eb2017bsel$age >=46 & eb2017bsel$age <= 60] <- 3
eb2017bsel$age_cat[eb2017bsel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.19, 0.22, 0.23)

marstat <- c('1', '2')
marstat <- c(0.53, 0.47)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.35, 0.17, 0.48)

targets <- list(sex, age_cat, marstat, urban)
names(targets) <- c("sex", "age_cat", "marstat", "urban")

eb2017bsel$caseid <- 1:length(eb2017bsel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2017bsel$sex <- as.numeric(eb2017bsel$sex)
eb2017bsel$marstat <- as.numeric(eb2017bsel$marstat)
eb2017bsel$urban <- as.numeric(eb2017bsel$urban)

anesrakefinder(targets, eb2017bsel, choosemethod = "total")

outsave <- anesrake(targets, eb2017bsel, caseid = eb2017bsel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2017bsel$weightvec  <- unlist(outsave[1])

save(eb2017bsel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2017bsel.Rdata")

#2021
load(here("./data/all_waves", "eb2021sel.RData"))

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(eb2021sel$age)
eb2021sel$age_cat[eb2021sel$age >=15 & eb2021sel$age <= 30] <- 1
eb2021sel$age_cat[eb2021sel$age >=31 & eb2021sel$age <= 45] <- 2
eb2021sel$age_cat[eb2021sel$age >=46 & eb2021sel$age <= 60] <- 3
eb2021sel$age_cat[eb2021sel$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.21, 0.25)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.34, 0.17, 0.49)

targets <- list(sex, age_cat, urban)
names(targets) <- c("sex", "age_cat", "urban")

eb2021sel$caseid <- 1:length(eb2021sel$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
eb2021sel$sex <- as.numeric(eb2021sel$sex)
eb2021sel$urban <- as.numeric(eb2021sel$urban)

anesrakefinder(targets, eb2021sel, choosemethod = "total")

outsave <- anesrake(targets, eb2021sel, caseid = eb2021sel$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

eb2021sel$weightvec  <- unlist(outsave[1])

save(eb2021sel, file="/Users/anuschka/Documents/climatechange/climatechange/data/all_waves/eb2021sel.Rdata")

Motivaction

load("./data/all_waves/mot2019.Rdata")

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(mot2019$age)
mot2019$age_cat[mot2019$age >=15 & mot2019$age <= 30] <- 1
mot2019$age_cat[mot2019$age >=31 & mot2019$age <= 45] <- 2
mot2019$age_cat[mot2019$age >=46 & mot2019$age <= 60] <- 3
mot2019$age_cat[mot2019$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.22, 0.24)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.34, 0.17, 0.49)
marstat <- c('1', '2')
marstat <- c(0.52, 0.48)

targets <- list(sex, age_cat, urban, marstat)
names(targets) <- c("sex", "age_cat", "urban", "marstat")

mot2019$caseid <- 1:length(mot2019$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
mot2019$sex <- as.numeric(mot2019$sex)
mot2019$urban <- as.numeric(mot2019$urban)
mot2019$marstat <- as.numeric(mot2019$marstat)

anesrakefinder(targets, mot2019, choosemethod = "total")

outsave <- anesrake(targets, mot2019, caseid = mot2019$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

mot2019$weightvec  <- unlist(outsave[1])

utils::View(mot2019)

save(mot2019, file="./data/all_waves/mot2019.Rdata")

# 2021
load("./data/all_waves/mot2021.Rdata")

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(mot2021$age)
mot2021$age_cat[mot2021$age >=15 & mot2021$age <= 30] <- 1
mot2021$age_cat[mot2021$age >=31 & mot2021$age <= 45] <- 2
mot2021$age_cat[mot2021$age >=46 & mot2021$age <= 60] <- 3
mot2021$age_cat[mot2021$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.21, 0.25)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.34, 0.17, 0.49)

marstat <- c('1', '2')
marstat <- c(0.51, 0.49)

targets <- list(sex, age_cat, urban, marstat)
names(targets) <- c("sex", "age_cat", "urban", "marstat")

mot2021$caseid <- 1:length(mot2021$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
mot2021$sex <- as.numeric(mot2021$sex)
mot2021$urban <- as.numeric(mot2021$urban)
mot2021$marstat <- as.numeric(mot2021$marstat)

anesrakefinder(targets, mot2021, choosemethod = "total")

outsave <- anesrake(targets, mot2021, caseid = mot2021$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

mot2021$weightvec  <- unlist(outsave[1])

save(mot2021, file="./data/all_waves/mot2021.Rdata")

LISS data

rm(list=ls())
load("./data/all_waves/liss2019.Rdata")

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(liss2019$age)
liss2019$age_cat[liss2019$age >=15 & liss2019$age <= 30] <- 1
liss2019$age_cat[liss2019$age >=31 & liss2019$age <= 45] <- 2
liss2019$age_cat[liss2019$age >=46 & liss2019$age <= 60] <- 3
liss2019$age_cat[liss2019$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.22, 0.24)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.34, 0.17, 0.49)
marstat <- c('1', '2')
marstat <- c(0.52, 0.48)

targets <- list(sex, age_cat, urban, marstat)
names(targets) <- c("sex", "age_cat", "urban", "marstat")

liss2019$caseid <- 1:length(liss2019$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
liss2019$sex <- as.numeric(liss2019$sex)
liss2019$urban <- as.numeric(liss2019$urban)
liss2019$marstat <- as.numeric(liss2019$marstat)

anesrakefinder(targets, liss2019, choosemethod = "total")

outsave <- anesrake(targets, liss2019, caseid = liss2019$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

liss2019$weightvec  <- unlist(outsave[1])

utils::View(liss2019)

save(liss2019, file="./data/all_waves/liss2019.Rdata")

load("./data/all_waves/liss2019.Rdata")

# 2020 (x2)
load("./data/all_waves/liss2020a.Rdata")

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(liss2020a$age)
liss2020a$age_cat[liss2020a$age >=15 & liss2020a$age <= 30] <- 1
liss2020a$age_cat[liss2020a$age >=31 & liss2020a$age <= 45] <- 2
liss2020a$age_cat[liss2020a$age >=46 & liss2020a$age <= 60] <- 3
liss2020a$age_cat[liss2020a$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.21, 0.25)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.34, 0.17, 0.49)
marstat <- c('1', '2')
marstat <- c(0.52, 0.48)

targets <- list(sex, age_cat, urban, marstat)
names(targets) <- c("sex", "age_cat", "urban", "marstat")

liss2020a$caseid <- 1:length(liss2020a$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
liss2020a$sex <- as.numeric(liss2020a$sex)
liss2020a$urban <- as.numeric(liss2020a$urban)
liss2020a$marstat <- as.numeric(liss2020a$marstat)

anesrakefinder(targets, liss2020a, choosemethod = "total")

outsave <- anesrake(targets, liss2020a, caseid = liss2020a$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

liss2020a$weightvec  <- unlist(outsave[1])
save(liss2020a, file="./data/all_waves/liss2020a.Rdata")

load("./data/all_waves/liss2020b.Rdata")

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(liss2020b$age)
liss2020b$age_cat[liss2020b$age >=15 & liss2020b$age <= 30] <- 1
liss2020b$age_cat[liss2020b$age >=31 & liss2020b$age <= 45] <- 2
liss2020b$age_cat[liss2020b$age >=46 & liss2020b$age <= 60] <- 3
liss2020b$age_cat[liss2020b$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.21, 0.25)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.34, 0.17, 0.49)
marstat <- c('1', '2')
marstat <- c(0.52, 0.48)

targets <- list(sex, age_cat, urban, marstat)
names(targets) <- c("sex", "age_cat", "urban", "marstat")

liss2020b$caseid <- 1:length(liss2020b$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
liss2020b$sex <- as.numeric(liss2020b$sex)
liss2020b$urban <- as.numeric(liss2020b$urban)
liss2020b$marstat <- as.numeric(liss2020b$marstat)

anesrakefinder(targets, liss2020b, choosemethod = "total")

outsave <- anesrake(targets, liss2020b, caseid = liss2020b$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

liss2020b$weightvec  <- unlist(outsave[1])
save(liss2020b, file="./data/all_waves/liss2020b.Rdata") 

#2021
load("./data/all_waves/liss2021.Rdata")

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(liss2021$age)
liss2021$age_cat[liss2021$age >=15 & liss2021$age <= 30] <- 1
liss2021$age_cat[liss2021$age >=31 & liss2021$age <= 45] <- 2
liss2021$age_cat[liss2021$age >=46 & liss2021$age <= 60] <- 3
liss2021$age_cat[liss2021$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.21, 0.25)

urban <- c('Low urbanity', 'Medium urbanity', 'High urbanity')
urban <- c(0.34, 0.17, 0.49)

marstat <- c('1', '2')
marstat <- c(0.51, 0.49)

targets <- list(sex, age_cat, urban, marstat)
names(targets) <- c("sex", "age_cat", "urban", "marstat")

liss2021$caseid <- 1:length(liss2021$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
liss2021$sex <- as.numeric(liss2021$sex)
liss2021$urban <- as.numeric(liss2021$urban)
liss2021$marstat <- as.numeric(liss2021$marstat)

anesrakefinder(targets, liss2021, choosemethod = "total")

outsave <- anesrake(targets, liss2021, caseid = liss2021$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

liss2021$weightvec  <- unlist(outsave[1])
save(liss2021, file="./data/all_waves/liss2021.Rdata")

SOCON data

# Doesn't have urbanity
rm(list=ls())
load("./data/all_waves/socon2020.Rdata")

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(socon2020$age)
socon2020$age_cat[socon2020$age >=15 & socon2020$age <= 30] <- 1
socon2020$age_cat[socon2020$age >=31 & socon2020$age <= 45] <- 2
socon2020$age_cat[socon2020$age >=46 & socon2020$age <= 60] <- 3
socon2020$age_cat[socon2020$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.21, 0.25)

marstat <- c('1', '2')
marstat <- c(0.52, 0.48)

targets <- list(sex, age_cat, marstat)
names(targets) <- c("sex", "age_cat", "marstat")

socon2020$caseid <- 1:length(socon2020$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
socon2020$sex <- as.numeric(socon2020$sex)
socon2020$marstat <- as.numeric(socon2020$marstat)

anesrakefinder(targets, socon2020, choosemethod = "total")

outsave <- anesrake(targets, socon2020, caseid = socon2020$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

socon2020$weightvec  <- unlist(outsave[1])
save(socon2020, file="./data/all_waves/socon2020.Rdata")

load("./data/all_waves/socon2021.Rdata")

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(socon2021$age)
#In 2021, a very old sample as there is nobody younger than 32 years. The first category thus can't be used (produces an error in the anesrake). Also, anesrake can only work with variables starting with score 1. This means that only for this year of socon, age_cat means something different than the other years. So I will recode it afterwards. 
#socon2021$age_cat[socon2021$age >=15 & socon2021$age <= 30] <- 1
socon2021$age_cat[socon2021$age >=31 & socon2021$age <= 45] <- 1
socon2021$age_cat[socon2021$age >=46 & socon2021$age <= 60] <- 2
socon2021$age_cat[socon2021$age >=61] <- 3

table(socon2021$age, useNA = "always")

age_cat <- c( '3145', '4661', '61o')
age_cat <- c( 0.18, 0.21, 0.25)

marstat <- c('1', '2')
marstat <- c(0.51, 0.49)

targets <- list(sex, age_cat, marstat)
names(targets) <- c("sex", "age_cat", "marstat")

socon2021$caseid <- 1:length(socon2021$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
socon2021$sex <- as.numeric(socon2021$sex)
socon2021$marstat <- as.numeric(socon2021$marstat)

utils::View(socon2021)

anesrakefinder(targets, socon2021, choosemethod = "total")

outsave <- anesrake(targets, socon2021, caseid = socon2021$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

socon2021$weightvec  <- unlist(outsave[1])
save(socon2021, file="./data/all_waves/socon2021.Rdata")

#Socon2022
load("./data/all_waves/socon2022.Rdata")

sex <- c('Male', 'Female')
sex <- c(0.50, 0.50)

attributes(socon2022$age)
socon2022$age_cat[socon2022$age >=15 & socon2022$age <= 30] <- 1
socon2022$age_cat[socon2022$age >=31 & socon2022$age <= 45] <- 2
socon2022$age_cat[socon2022$age >=46 & socon2022$age <= 60] <- 3
socon2022$age_cat[socon2022$age >=61] <- 4

age_cat <- c('u30', '3145', '4661', '61o')
age_cat <- c(0.20, 0.18, 0.21, 0.25)

marstat <- c('1', '2')
marstat <- c(0.51, 0.49)

targets <- list(sex, age_cat, marstat)
names(targets) <- c("sex", "age_cat", "marstat")

socon2022$caseid <- 1:length(socon2022$sex)

# To use anesrakefinder, the variables in the original dataset have to be numeric
socon2022$sex <- as.numeric(socon2022$sex)
socon2022$marstat <- as.numeric(socon2022$marstat)

anesrakefinder(targets, socon2022, choosemethod = "total")

outsave <- anesrake(targets, socon2022, caseid = socon2022$caseid,
  verbose= FALSE, cap = 5, choosemethod = "total",
  type = "pctlim", pctlim = .05 , nlim = 5,
  iterate = TRUE , force1 = TRUE)
summary(outsave)

socon2022$weightvec  <- unlist(outsave[1])
save(socon2022, file="./data/all_waves/socon2022.Rdata")