library(dplyr)
library(readr)
library(tidyr)
library(plotly)
library(geosphere)
library(readxl)
library(rvec)
central_lat = -34.9211
central_long = 138.5964Adding in Referendum Results
pp <- read_csv("data/GeneralPollingPlacesDownload-29581.csv", skip=1)
pp_primary <- read_csv("data/ReferendumPollingPlaceResultsByStateDownload-29581-SA.csv", skip=1) %>%
left_join(pp, by=c("PollingPlaceId" = "PollingPlaceID")) %>%
mutate(
Latitude = as.numeric(Latitude),
Longitude = as.numeric(Longitude),
) %>%
filter(!is.na(Latitude)) %>%
filter(!is.na(Longitude)) %>% rowwise() %>% mutate(
dist = distm(c(central_long, central_lat), c(Longitude, Latitude), fun=distHaversine)[,1]/1000
) %>% select(
c(DivisionName, PollingPlaceNm.x, PremisesSuburb, YesVotes, YesPercentage, NoVotes, NoPercentage, dist)
) %>% pivot_longer(
cols = c(YesVotes, YesPercentage, NoVotes, NoPercentage),
names_to = c("Response", ".value"),
names_pattern = "(Yes|No)(Votes|Percentage)"
) %>% rename(
PollingPlaceName = PollingPlaceNm.x,
Suburb=PremisesSuburb,
Party = "Response"
) %>% mutate(
election = "23Voice"
)Adding federal 25
pp <- read_csv("data/GeneralPollingPlacesDownload-31496.csv", skip=1)
pp_to_add <- read_csv("data/HouseStateFirstPrefsByPollingPlaceDownload-31496-SA.csv", skip=1) %>%
filter(StateAb == "SA") %>%
filter(CandidateID != 999) %>%
filter(PartyAb %in% c("ALP", "LP", "ON", "GRN")) %>%
group_by(DivisionID, PollingPlaceID) %>%
mutate(
pp_percent = OrdinaryVotes*100/sum(OrdinaryVotes)
) %>% ungroup() %>%
left_join(pp, by=c("PollingPlaceID"="PollingPlaceID")) %>%
mutate(
Latitude = as.numeric(Latitude),
Longitude = as.numeric(Longitude),
) %>%
filter(!is.na(Latitude)) %>%
filter(!is.na(Longitude)) %>% rowwise() %>% mutate(
dist = distm(c(central_long, central_lat), c(Longitude, Latitude), fun=distHaversine)[,1]/1000
) %>% select(DivisionNm.x, PollingPlace, PremisesSuburb, dist, PartyAb, OrdinaryVotes, pp_percent) %>%
rename(
DivisionName = DivisionNm.x,
PollingPlaceName = PollingPlace,
Suburb=PremisesSuburb,
Party = PartyAb,
Votes = OrdinaryVotes,
Percentage = pp_percent
) %>% mutate(
election = "25Fed"
)
pp_primary <- pp_primary %>% rbind(pp_to_add)Adding SA26
pp <- read_csv("data/SA-2026-Pollingplaces.xlsx - Sheet1.csv")
pp_to_add <- read_csv("data/SA-2026-HA-Primary-PollingPlace.xlsx - Sheet1.csv") %>%
filter(candidate_id != 999) %>%
filter(party_code %in% c("ALP", "LIB", "ON", "GRN")) %>%
group_by(district_id, pp_id) %>%
mutate(
pp_percent = votes*100/sum(votes)
) %>% ungroup() %>%
left_join(pp, by = c("pp_id" = "pp_id")) %>%
mutate(
latitude = as.numeric(latitude),
longitude = as.numeric(longitude),
) %>%
filter(!is.na(latitude)) %>%
filter(!is.na(longitude)) %>% rowwise() %>% mutate(
dist = distm(c(central_long, central_lat), c(longitude, latitude), fun=distHaversine)[,1]/1000
) %>% select(district_name.x, pp_name.x, locality, dist, party_code, votes, pp_percent) %>%
rename(
DivisionName = district_name.x,
PollingPlaceName = pp_name.x,
Suburb=locality,
Party = party_code,
Votes = votes,
Percentage = pp_percent
) %>% mutate(
election = "26SA",
Party = recode(Party, "LIB" = "LP")
)
pp_primary <- pp_primary %>% rbind(pp_to_add)pp_primary$Party = factor(pp_primary$Party)house_prices <- read_excel("data/lsg_stats_2026_q1.xlsx") %>%
select(!"Median Change") %>% select(!c("Sales 1Q 2025", "Median 1Q 2025")) %>%
mutate(
Year = 2026,
Quarter = 1
) %>% rename(
Sales = "Sales 1Q 2026",
Median = "Median 1Q 2026"
)
house_prices = rbind(
read_excel("data/lsg_stats_2026_q1.xlsx") %>%
select(!"Median Change") %>% select(!c("Sales 1Q 2025", "Median 1Q 2025")) %>%
mutate(
Year = 2026,
Quarter = 1
) %>% rename(
Sales = "Sales 1Q 2026",
Median = "Median 1Q 2026"
),
read_excel("data/lsg_stats_2025_q4.xlsx") %>%
select(!"Median Change") %>% select(!c("Sales 4Q 2024", "Median 4Q 2024")) %>%
mutate(
Year = 2025,
Quarter = 4
) %>% rename(
Sales = "Sales 4Q 2025",
Median = "Median 4Q 2025"
),
read_excel("data/lsg_stats_2025_q3.xlsx") %>%
select(!"Median Change") %>% select(!c("Sales 3Q 2024", "Median 3Q 2024")) %>%
mutate(
Year = 2025,
Quarter = 3
) %>% rename(
Sales = "Sales 3Q 2025",
Median = "Median 3Q 2025"
),
read_excel("data/lsg_stats_2025_q2.xlsx") %>%
select(!"Median Change") %>% select(!c("Sales 2Q 2024", "Median 2Q 2024")) %>%
mutate(
Year = 2025,
Quarter = 2
) %>% rename(
Sales = "Sales 2Q 2025",
Median = "Median 2Q 2025"
)
) %>% group_by(City,Suburb) %>%
replace_na(list(Sales = 0, Median=0)) %>%
summarise(
Median = weighted_median(Median, wt=Sales),
Sales = sum(Sales),
)party_pal <- c("#d8352a", "#505aaf", "orange", "#29a37a", "#b81f14", "#29339b")
party_pal <- setNames(party_pal, c("ALP", "LP", "ON", "GRN", "Yes", "No"))
election_pal <- c("#b81f14", "#29339b", "#1f7a5c")
election_pal <- setNames(election_pal, c("23Voice", "25Fed", "26SA"))
pp_primary %>%
filter(Party %in% c("ALP", "LP", "ON", "GRN")) %>%
filter(!grepl('PPVC', PollingPlaceName)) %>%
filter(!grepl('Early Voting', PollingPlaceName)) %>%
arrange(dist) %>%
group_by(Party) %>%
group_map(~plot_ly(data=.,
x=~dist, y=~Percentage, color=~Party,
text = ~paste("District: ", DivisionName, '<br>Polling Place:', PollingPlaceName, "<br>Distance:", round(dist, digits=3), "km<br>Percent:", round(Percentage, digits=2)),
size=~Votes,
colors=party_pal,
frame = ~election
) %>% layout(xaxis=list(type="log")), .keep=TRUE) %>%
subplot(nrows = 2, shareX = TRUE, shareY=TRUE, titleX = FALSE, titleY = FALSE) %>%
layout(
annotations=list(
list(
text = "Distance from Adelaide (log km)",
x = 0.5, y = -0.1,
xref = "paper", yref = "paper",
showarrow = FALSE
),
list(
text = "Polling Place Primary Result (%)",
x = -0.08, y = 0.5,
xref = "paper", yref = "paper",
textangle = -90,
showarrow = FALSE
)
),
width=600
)pp_primary %>%
filter(Party %in% c("ALP", "LP", "ON", "GRN")) %>%
filter(!grepl('PPVC', PollingPlaceName)) %>%
filter(!grepl('Early Voting', PollingPlaceName)) %>%
left_join(house_prices, by = c("Suburb" = "Suburb")) %>%
filter(!is.na(Median)) %>%
arrange(Median) %>%
group_by(Party) %>%
group_map(~plot_ly(data=.,
x=~Median, y=~Percentage, color=~Party,
text = ~paste("District: ", DivisionName, '<br>Polling Place:', PollingPlaceName, "<br>Price:", paste0("$", Median), "<br>Percent:", round(Percentage, digits=2)),
size=~Votes,
colors=party_pal,
frame = ~election
) %>% layout(xaxis=list(type="log")), .keep=TRUE) %>%
subplot(nrows = 2, shareX = TRUE, shareY=TRUE, titleX = FALSE, titleY = FALSE) %>%
layout(
annotations=list(
list(
text = "Median House Price (Log) (Q1 2025-Q1 2026)",
x = 0.5, y = -0.15,
xref = "paper", yref = "paper",
showarrow = FALSE
),
list(
text = "Polling Place Primary Result (%)",
x = -0.08, y = 0.5,
xref = "paper", yref = "paper",
textangle = -90,
showarrow = FALSE
)
),
width=600
)pp_primary %>%
filter(Party %in% c("Yes", "No")) %>%
filter(!grepl('PPVC', PollingPlaceName)) %>%
filter(!grepl('Early Voting', PollingPlaceName)) %>%
left_join(house_prices, by = c("Suburb" = "Suburb")) %>%
filter(!is.na(Median)) %>%
arrange(Median) %>%
group_by(Party) %>%
plot_ly(data=.,
x=~Median, y=~Percentage, color=~Party,
text = ~paste("District: ", DivisionName, '<br>Polling Place:', PollingPlaceName, "<br>Price:", paste0("$", Median), "<br>Percent:", round(Percentage, digits=2)),
size=~Votes,
colors=party_pal
) %>%
layout(
xaxis=list(type="log", title="Median House Price (Log) (Q1 2025-Q1 2026)"),
yaxis=list(title="Polling Place Primary Result (%)"),
width=600
)pp_primary %>%
filter(Party %in% c("Yes", "No")) %>%
filter(!grepl('PPVC', PollingPlaceName)) %>%
filter(!grepl('Early Voting', PollingPlaceName)) %>%
left_join(house_prices, by = c("Suburb" = "Suburb")) %>%
filter(!is.na(dist)) %>%
arrange(dist) %>%
group_by(Party) %>%
plot_ly(data=.,
x=~dist, y=~Percentage, color=~Party,
text = ~paste("District: ", DivisionName, '<br>Polling Place:', PollingPlaceName, "<br>Distance:", round(dist, digits=3), "km<br>Percent:", round(Percentage, digits=2)),
size=~Votes,
colors=party_pal
) %>% layout(
xaxis=list(type="log", title="Distance from Adelaide (log km)"),
yaxis=list(title="Polling Place Primary Result (%)"),
width=600
)pp_primary %>% select(Suburb, election, dist, Party, Votes) %>%
group_by(Suburb, election, Party) %>%
summarise(
Votes = sum(Votes)
) %>% ungroup() %>%
group_by(Suburb, election) %>%
mutate(
suburb_percent = Votes*100/sum(Votes)
) %>% filter(Party %in% c("No", "ON")) %>%
mutate(
Party = recode(Party, "No" = "ON")
) %>% ungroup() %>% pivot_wider(
names_from = election,
values_from = c(Votes, suburb_percent)
) %>%
plot_ly(
data=.,
x=~suburb_percent_23Voice
) %>%
add_markers(y=~suburb_percent_25Fed, name="Federal 25") %>%
add_markers(y=~suburb_percent_26SA, name="South Australia 26") %>%
layout(yaxis = list(title = "One Nation Suburb Result (%)"),
xaxis = list(title = "No Voice Referendum Suburb Result (%)"), width=600, legend = list(x = 0.1, y = 0.9))pp_primary %>% select(Suburb, election, dist, Party, Votes) %>%
group_by(Suburb, election, Party) %>%
summarise(
Votes = sum(Votes)
) %>% ungroup() %>%
group_by(Suburb, election) %>%
mutate(
suburb_percent = Votes*100/sum(Votes)
) %>% filter(Party %in% c("No", "LP")) %>%
mutate(
Party = recode(Party, "No" = "LP")
) %>% ungroup() %>% pivot_wider(
names_from = election,
values_from = c(Votes, suburb_percent)
) %>%
plot_ly(
data=.,
x=~suburb_percent_23Voice
) %>%
add_markers(y=~suburb_percent_25Fed, name="Federal 25") %>%
add_markers(y=~suburb_percent_26SA, name="South Australia 26") %>%
layout(yaxis = list(title = "Liberal Suburb Result (%)"),
xaxis = list(title = "No Voice Referendum Suburb Result (%)"), width=600, legend = list(x = 0.1, y = 0.9))