-
Notifications
You must be signed in to change notification settings - Fork 7
/
0108a-nzes-prep.R
283 lines (231 loc) · 10.4 KB
/
0108a-nzes-prep.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
# This script is prep for the shiny app in the ./_working/0108/ folder
# See previous blog posts for sourcing the original data
# App is deployed to https://ellisp.shinyapps.io/nzes2014_x_by_party/
library(foreign)
library(tidyverse)
library(forcats)
library(DT)
library(survey)
library(rsconnect)
library(frs)
maori <- "M\U0101ori"
#----------------------2014----------------------
# Import the New Zealand Election Study data:
nzes14_orig <- read.spss("NZES2014GeneralReleaseApril16.sav",
to.data.frame = TRUE, trim.factor.names = TRUE, reencode = FALSE)
# full names for all the questions
varlab14 <- cbind(attributes(nzes14_orig)$variable.labels)
# subset which variables we want
vars14 <- varlab14[c(10:22, 94:99, 178:212, 261, 270:288, 308:324, 331, 341:343,
351:358, 362:381, 385:403), ]
vars14 <- gsub("M.ori", maori, vars14)
# View(data.frame(as.character(vars14)))
vc14 <- enc2utf8(as.character(vars14))
vars14_list <- list(
"Political engagement" = vc14[c(1:8, 14:19, 56:66)],
"Views on spending" = vc14[20:30],
"Other attitudes" = vc14[31:55],
"Views on governance" = vc14[c(67:84, 87, 88)],
"About you" = vc14[c(85, 86, 89:142, 9:13)]
)
# tidy up the data
nzes14 <- nzes14_orig %>%
# Fix M?ori in any levels of factors:
map_df(function(x){
if(is.factor(x)){
levels(x) <- gsub("M.ori", maori, levels(x))
x
} else {
x
}}) %>%
# Fix non-useful encoding of binary variables when some are NA
map_df(function(x){
if(is.factor(x) & length(levels(x)) == 1){
lx <- tolower(as.character(levels(x)))
x <- as.character(x)
x[is.na(x)] <- paste("Not", lx)
return(x)
} else {
return(x)
}
}) %>%
# Fix non-useful encoding of binary variables when some are 0
map_df(function(x){
if(is.factor(x) & 0 %in% levels(x)){
lx <- as.character(levels(x)[2])
x <- as.character(x)
x[x == 0] <- paste("Not", lx)
return(x)
} else {
return(x)
}
}) %>%
# update gender classification to match the 2015 Stats NZ form of words at
# http://www.stats.govt.nz/tools_and_services/media-centre/media-releases-2015/gender-identity-17-july-15.aspx
mutate(dsex = ifelse(dsex == "Transsexual or transgender", "Gender-diverse", as.character(dsex)),
dsex = fct_infreq(dsex)) %>%
# lump up party vote:
mutate(partyvote = ifelse(is.na(dpartyvote), "Did not vote", as.character(dpartyvote)),
partyvote = gsub("net.Man", "net-Man", partyvote),
partyvote = fct_lump(partyvote, 10, other_level = "Another party"),
partyvote = fct_infreq(partyvote),
# This magic constant, 3140417 was the size of the electoral roll at the time of election;
# see http://www.electionresults.govt.nz/electionresults_2014/e9/html/e9_part9_1.html
dwtfin = dwtfin * 3140.417 / sum(dwtfin) )
nzes14 <- nzes14[ ,c(names(vars14), "partyvote", "dwtfin")]
party_numbers <- nzes14 %>%
group_by(partyvote) %>%
summarise(sample_size = n())
nzes14 <- nzes14 %>%
left_join(party_numbers, by = "partyvote") %>%
mutate(partyvote_n = paste0(partyvote, ", n = ", sample_size),
partyvote_n = fct_infreq(partyvote_n))
#------------2014 actual outcomes----------------
# http://www.elections.org.nz/news-media/new-zealand-2014-general-election-official-results
# This is calibration of the larger votes to their actual votes. Recommended for our purposes,
# as we want the reported votes to be actual numbers sometimes.
actual_vote_2014 <- tibble(
partyvote2014 = c("National", "Labour", "Green", "NZ First", "Other", "Did not vote"),
Freq = c(1131501, 604534, 257356, 208300,
31850 + 16689 + 5286 + 95598 + 34095 + 10961 + 5113 + 1730 + 1096 + 872 + 639,
NA)
)
# calculate the did not vote, from the 77.9 percent turnout
actual_vote_2014[6, 2] <- (100 / 77.9 - 1) * sum(actual_vote_2014[1:5, 2])
nzes14_vote_totals <- nzes14 %>%
ungroup() %>%
filter(partyvote != "Don't know") %>%
mutate(partyvote2014 = fct_other(partyvote, keep = actual_vote_2014$partyvote2014)) %>%
group_by(partyvote2014) %>%
summarise(original_weight = sum(dwtfin)) %>%
mutate(partyvote2014 = as.character(partyvote2014)) %>%
left_join(actual_vote_2014, by = "partyvote2014") %>%
mutate(multiplier = Freq / original_weight / 1000) %>%
select(partyvote2014, multiplier) %>%
rbind(tibble(partyvote2014 = "Don't know", multiplier = 0))
nzes14 <- nzes14 %>%
mutate(partyvote2014 = fct_other(partyvote, keep = c(actual_vote_2014$partyvote2014, "Don't know"))) %>%
mutate(partyvote2014 = as.character(partyvote2014)) %>%
left_join(nzes14_vote_totals, by = "partyvote2014") %>%
mutate(calibrated_weight = dwtfin * multiplier) %>%
mutate(partyvote = fct_relevel(partyvote, "Did not vote"),
partyvote_n = fct_relevel(partyvote_n, levels(partyvote_n)[grepl("Did not vote", levels(partyvote_n))]))
#========================2017===============================
# The R version doesn't load, you get
# > load("NZES2017R/NZES2017Release14-07-19.rdata")
# Error in load("NZES2017R/NZES2017Release14-07-19.rdata") :
# Value of SET_STRING_ELT() must be a 'CHARSXP' not a 'integer'
# unzip("NZES2017SPSS.zip")
nzes17_orig <- read.spss("NZES2017SPSS/NZES2017Release14-07-19.sav",
to.data.frame = TRUE, trim.factor.names = TRUE, reencode = FALSE)
# full names for all the questions
varlab17 <- attributes(nzes17_orig)$variable.labels
# check that we have all the variables we used in 2014 (we don't)
# stopifnot(sum(!names(vars) %in% names(varlab)) == 0)
# stopifnot(sum(!vc %in% as.character(varlab)) == 0)
# So we will have to identify these by hand
# subset which variables we want
vars17 <- varlab17[c(18:42, 91:96, 98:155, 192:245,297:341, 342, 344, 356,
357:360, 361, 365:377, 379:398, 406:440,466:485)]
vars17 <- gsub("M.ori", maori, vars17)
vars17 <- gsub("MÄ\u0081ori", maori, vars17)
vars17 <- gsub("2014–2017", "2014-2017", vars17)
vc17 <- enc2utf8(as.character(vars17))
vars17_list <- list(
"Political engagement" = vc17[c(284, 1:2, 8:31, 150:161)],
"Views on spending" = vc17[c(98:108)],
"Other economic views" = vc17[c(54:71, 109:112, 136:143)],
"Other attitudes" = vc17[c(94:97, 113:134, 164:172)],
"Views on governance" = vc17[c(32:53, 72:91, 135, 144:149, 162:163, 173:188)],
"About you" = vc17[c(189:208, 210:261, 263:264, 273:274,276:283, 3:7)]
)
# tidy up the data
nzes17 <- nzes17_orig %>%
# Fix M?ori in any levels of factors:
map_df(function(x){
if(is.factor(x)){
levels(x) <- gsub("M.ori", maori, levels(x))
x
} else {
x
}}) %>%
# Fix non-useful encoding of binary variables when some are NA
map_df(function(x){
if(is.factor(x) & length(levels(x)) == 1){
lx <- tolower(as.character(levels(x)))
x <- as.character(x)
x[is.na(x)] <- paste("Not", lx)
return(x)
} else {
return(x)
}
}) %>%
# Fix non-useful encoding of binary variables when some are 0
map_df(function(x){
if(is.factor(x) & 0 %in% levels(x)){
lx <- as.character(levels(x)[2])
x <- as.character(x)
x[x == 0] <- paste("Not", lx)
return(x)
} else {
return(x)
}
}) %>%
# update gender classification to match the 2015 Stats NZ form of words at
# http://www.stats.govt.nz/tools_and_services/media-centre/media-releases-2015/gender-identity-17-july-15.aspx
mutate(rsex = ifelse(rsex == "Transsexual or transgender", "Gender-diverse", as.character(rsex)),
rsex = fct_infreq(rsex)) %>%
# tidy party names and lump up party vote:
mutate(partyvote = ifelse(is.na(rpartyvote), "Did not vote", as.character(rpartyvote)),
partyvote = gsub("[0-9]+\\. ", "", partyvote),
partyvote = gsub("net.Man", "net-Man", partyvote),
partyvote = fct_lump(partyvote, 10, other_level = "Another party"),
partyvote = fct_infreq(partyvote),
# This magic constant, 3298009 was the size of the electoral roll at the time of election;
# see https://www.electionresults.govt.nz/electionresults_2017/statistics/index.html
dwtfin = rwt * 3298.009 / sum(rwt) )
nzes17 <- nzes17[ ,c(names(vars17), "partyvote", "dwtfin")]
party_numbers17 <- nzes17 %>%
group_by(partyvote) %>%
summarise(sample_size = n())
nzes17 <- nzes17 %>%
left_join(party_numbers17, by = "partyvote") %>%
mutate(partyvote_n = paste0(partyvote, ", n = ", sample_size),
partyvote_n = fct_infreq(partyvote_n))
#------------2017 actual outcomes----------------
# http://www.elections.org.nz/news-media/new-zealand-2017-general-election-official-results
actual_vote_2017 <- tibble(
partyvote2017 = c("National", "Labour", "Green", "NZ First", "Other", "Did not vote"),
Freq = c(1152075, 956184, 162443, 186706,
13075 + 63261 + 30580 + 8075 + 6253 + 3642 + 3005 + 1890 + 1782 + 1620 + 806 + 499,
NA)
)
# check adds up correct and I didn't have a typo"
stopifnot(sum(actual_vote_2017$Freq, na.rm = TRUE) == 2591896)
# calculate the did not vote, from the 79.75 percent turnout
actual_vote_2017[6, 2] <- (100 / 79.75 - 1) * sum(actual_vote_2017[1:5, 2])
nzes17_vote_totals <- nzes17 %>%
ungroup() %>%
filter(partyvote != "Don't know") %>%
mutate(partyvote2017 = as.character(fct_other(partyvote, keep = actual_vote_2017$partyvote2017))) %>%
group_by(partyvote2017) %>%
summarise(original_weight = sum(dwtfin)) %>%
left_join(actual_vote_2017, by = "partyvote2017") %>%
mutate(multiplier = Freq / original_weight / 1000) %>%
select(partyvote2017, multiplier) %>%
rbind(data_frame(partyvote2017 = "Don't know", multiplier = 0))
nzes17 <- nzes17 %>%
mutate(partyvote2017 = fct_other(partyvote, keep = c(actual_vote_2017$partyvote2017, "Don't know"))) %>%
mutate(partyvote2017 = as.character(partyvote2017)) %>%
left_join(nzes17_vote_totals, by = "partyvote2017") %>%
mutate(calibrated_weight = dwtfin * multiplier) %>%
mutate(partyvote = fct_relevel(partyvote, "Did not vote"),
partyvote_n = fct_relevel(partyvote_n, levels(partyvote_n)[grepl("Did not vote", levels(partyvote_n))]))
#==================Combine and save==================
save(vars14_list, file = "0108a/data/vars14_list.rda")
save(vars14, file = "0108a/data/vars14.rda")
save(nzes14, file = "0108a/data/nzes14.rda")
save(vars17_list, file = "0108a/data/vars17_list.rda")
save(vars17, file = "0108a/data/vars17.rda")
save(nzes17, file = "0108a/data/nzes17.rda")
# rsconnect::deployApp("0108a", appName = "nzes_by_party", account = "ellisp")