衆議院議員選挙2026年福岡11区の分析
Yutaka
2026-02-23
AONF Project
- RStudioで公開中
# =================================================================
# AONF Project: Standard Graph Format
# =================================================================
# 1. 共通設定（スクリプトの冒頭で一度だけ実行）
library(ggplot2)
# ブログタイトルの定義
# 英語と日本語の間に改行 (\n) あり
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# 2. グラフ作成の基本構成
# 以下をコピーして、データと変数名を書き換えて使用してください
ggplot(data = [データフレーム名], aes(x = [X軸], y = [Y軸], fill = [色分け])) +
geom_bar(stat = "identity") + # または geom_point(), geom_line() など
labs(
title = "ここにタイトルを入力",
subtitle = "ここにサブタイトルを入力",
x = "X軸ラベル",
y = "Y軸ラベル",
fill = "凡例名",
caption = AONF_CAPTION # 共通設定を呼び出し
) +
# 指定のフォントとミニマルテーマを適用
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
# キャプションのスタイルを統一
plot.caption = element_text(size = 8, color = "grey30", lineheight = 1.2, margin = margin(t = 15))
)1. 自治体選管のデータ
行橋市
1
library(ggplot2)
# =================================================================
# AONF Project: Standard Graph Format
# =================================================================
# 1. 共通設定（スクリプトの冒頭で定義）
# 英語と日本語の間に改行 (\n) を入れることで、モバイル表示でも美しく収まります
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# 行橋市の確定データ (2026年 第51回衆院選)
election_data <- data.frame(
candidate = c("村上 智信", "しき 玲子", "武田 良太", "辻 智之", "井上なおき"),
party = c("維新", "社民", "自民", "中道", "参政"),
district_votes = c(11815, 1041, 13915, 4751, 2843), # 小選挙区
pr_votes = c(4202, 839, 13362, 5787, 2799) # 比例代表
)
# COR (候補者超過達成比率) の計算
# $COR = \frac{\text{小選挙区得票}}{\text{比例政党得票}}$
election_data$cor <- election_data$district_votes / election_data$pr_votes
# 2. グラフ作成
ggplot(election_data, aes(x = reorder(candidate, -cor), y = cor, fill = party)) +
geom_bar(stat = "identity", width = 0.7) +
geom_text(aes(label = round(cor, 2)), vjust = -0.5, size = 4) +
geom_hline(yintercept = 1.0, linetype = "dashed", color = "red") +
# 縦軸の範囲を調整
scale_y_continuous(limits = c(0, 3.2)) +
# カスタムカラー設定
scale_fill_manual(values = c("自民" = "#3536C1", "維新" = "#31D925",
"中道" = "#E9EB3B", "社民" = "#EB3B3B", "参政" = "orange")) +
coord_fixed() +
labs(
title = "行橋市における候補者超過達成比率 (COR)",
subtitle = "第51回衆院選 (2026) 確定データに基づく分析\n※赤の破線はCOR=1.0（比例票と小選挙区票が一致）を示す",
x = "候補者",
y = "COR (小選挙区得票 / 比例政党得票)",
fill = "政党名",
caption = AONF_CAPTION # 共通設定を呼び出し
) +
# 指定のフォントとミニマルテーマを適用
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
# キャプションのスタイルを統一
plot.caption = element_text(size = 8, color = "grey30", lineheight = 1.2, margin = margin(t = 15))
)
2
library(ggplot2)
# 1. AONF形式: キャプション変数の定義
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# 行橋市の確定データ (2026年 第51回衆院選)
election_data <- data.frame(
candidate = c("村上 智信", "しき 玲子", "武田 良太", "辻 智之", "井上なおき"),
party = c("維新", "社民", "自民", "中道", "参政"),
district_votes = c(11815, 1041, 13915, 4751, 2843), # 小選挙区
pr_votes = c(4202, 839, 13362, 5787, 2799) # 比例代表
)
# 相関図の作成
ggplot(election_data, aes(x = pr_votes, y = district_votes, color = party)) +
geom_point(size = 4) +
# 凡例の「a」を消去するための設定
geom_text(aes(label = candidate), vjust = -1.5, family = "HiraKakuProN-W3",show.legend = FALSE) +
# COR=1.0 の基準線 (y = x)
geom_abline(intercept = 0, slope = 1, linetype = "dashed",color = "red", alpha = 0.5) +
scale_x_continuous(limits = c(0, 15000)) +
scale_y_continuous(limits = c(0, 15000)) +
scale_color_manual(values = c("自民" = "#3536C1", "維新" = "#31D925",
"中道" = "#E9EB3B", "社民" = "#EB3B3B", "参政" = "orange")) +
coord_fixed() +
# 2. AONF形式: labsでの適用
labs(title = "政党得票数と候補者得票数の相関図（行橋市）",
subtitle = "第51回衆院選 (2026)確定データ\n ※赤い破線は COR = 1.0 (小選挙区票 = 比例票) を示す",
x = "政党得票数（比例代表）",
y = "候補者得票数（小選挙区）",
caption = AONF_CAPTION) +
# 3. AONF形式: themeの一括指定
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(legend.position = "bottom",
plot.caption = element_text(size = 8, hjust = 1, color = "gray30"))
田川市
1
library(ggplot2)
# =================================================================
# AONF Project: Standard Graph Format
# =================================================================
# 1. 共通設定
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# 田川市の確定データ (2026年 第51回衆院選)
election_data <- data.frame(
candidate = c("村上 智信", "しき 玲子", "武田 良太", "辻 智之", "井上なおき"),
party = c("維新", "社民", "自民", "中道", "参政"),
district_votes = c(6109, 1113, 8654, 2981, 1362), # 小選挙区
pr_votes = c(2197, 860, 7322, 3794, 1530) # 比例代表
)
# CORの計算
election_data$cor <- election_data$district_votes / election_data$pr_votes
# 2. グラフ作成
ggplot(election_data, aes(x = reorder(candidate, -cor), y = cor, fill = party)) +
geom_bar(stat = "identity", width = 0.7) +
geom_text(aes(label = round(cor, 2)), vjust = -0.5, size = 4) +
geom_hline(yintercept = 1.0, linetype = "dashed", color = "red") +
# 縦軸の範囲を調整
scale_y_continuous(limits = c(0, 3.2)) +
# カスタムカラー設定
scale_fill_manual(values = c("自民" = "#3536C1", "維新" = "#31D925",
"中道" = "#E9EB3B", "社民" = "#EB3B3B", "参政" = "orange")) +
coord_fixed() +
labs(
title = "田川市における候補者超過達成比率 (COR)",
subtitle = "第51回衆院選 (2026) 確定データに基づく分析\n※赤の破線はCOR=1.0（比例票と小選挙区票が一致）を示す",
x = "候補者",
y = "COR (小選挙区得票 / 比例政党得票)",
fill = "政党名",
caption = AONF_CAPTION # 共通設定を呼び出し
) +
# AONF標準テーマ設定
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
# キャプションのスタイルを統一
plot.caption = element_text(size = 8, color = "grey30", lineheight = 1.2, margin = margin(t = 15))
)
2
library(ggplot2)
# 1. 共通設定
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# 田川市の確定データ (2026年 第51回衆院選)
election_data <- data.frame(
candidate = c("村上 智信", "しき 玲子", "武田 良太", "辻 智之", "井上なおき"),
party = c("維新", "社民", "自民", "中道", "参政"),
district_votes = c(6109, 1113, 8654, 2981, 1362), # 小選挙区
pr_votes = c(2197, 860, 7322, 3794, 1530) # 比例代表
)
# 相関図の作成
ggplot(election_data, aes(x = pr_votes, y = district_votes, color = party)) +
geom_point(size = 4) +
# show.legend = FALSE を追加して凡例の「a」を消去
geom_text(aes(label = candidate), vjust = -1.5, family = "HiraKakuProN-W3", show.legend = FALSE) +
# COR=1.0 の基準線 (y = x)
geom_abline(intercept = 0, slope = 1, linetype = "dashed", color = "red", alpha = 0.5) +
scale_x_continuous(limits = c(0, 9300)) +
scale_y_continuous(limits = c(0, 9300)) +
scale_color_manual(values = c("自民" = "#3536C1", "維新" = "#31D925",
"中道" = "#E9EB3B", "社民" = "#EB3B3B", "参政" = "orange")) +
coord_fixed() +
labs(title = "政党得票数と候補者得票数の相関図（田川市）",
subtitle = "第51回衆院選 (2026) 確定データ\n※ 赤い破線は COR = 1.0 (小選挙区票 = 比例票) を示す",
x = "政党得票数（比例代表）",
y = "候補者得票数（小選挙区）",
caption = AONF_CAPTION) +
# 3. AONF形式: themeの一括指定
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(legend.position = "bottom",
plot.caption = element_text(size = 8, hjust = 1, color = "gray30"))
苅田町
1
library(ggplot2)
# =================================================================
# AONF Project: Standard Graph Format
# =================================================================
# 1. 共通設定
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# 苅田町の確定データ (2026年 第51回衆院選)
election_data <- data.frame(
candidate = c("村上 智信", "しき 玲子", "武田 良太", "辻 智之", "井上なおき"),
party = c("維新", "社民", "自民", "中道", "参政"),
district_votes = c(4818, 470, 6084, 2365, 1541), # 小選挙区
pr_votes = c(1695, 308, 5784, 2702,1414 ) # 比例代表
)
# CORの計算
election_data$cor <- election_data$district_votes / election_data$pr_votes
# 2. グラフ作成
ggplot(election_data, aes(x = reorder(candidate, -cor), y = cor, fill = party)) +
geom_bar(stat = "identity", width = 0.7) +
geom_text(aes(label = round(cor, 2)), vjust = -0.5, size = 4) +
geom_hline(yintercept = 1.0, linetype = "dashed", color = "red") +
# 縦軸の範囲を調整
scale_y_continuous(limits = c(0, 3.2)) +
# カスタムカラー設定
scale_fill_manual(values = c("自民" = "#3536C1", "維新" = "#31D925",
"中道" = "#E9EB3B", "社民" = "#EB3B3B", "参政" = "orange")) +
coord_fixed() +
labs(
title = "苅田町における候補者超過達成比率 (COR)",
subtitle = "第51回衆院選 (2026) 確定データに基づく分析\n※赤の破線はCOR=1.0（比例票と小選挙区票が一致）を示す",
x = "候補者",
y = "COR (小選挙区得票 / 比例政党得票)",
fill = "政党名",
caption = AONF_CAPTION # 共通設定を呼び出し
) +
# AONF標準テーマ設定
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
# キャプションのスタイルを統一
plot.caption = element_text(size = 8, color = "grey30", lineheight = 1.2, margin = margin(t = 15))
)
2
library(ggplot2)
# 1. 共通設定
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# 苅田町の確定データ (2026年 第51回衆院選)
election_data <- data.frame(
candidate = c("村上 智信", "しき 玲子", "武田 良太", "辻 智之", "井上なおき"),
party = c("維新", "社民", "自民", "中道", "参政"),
district_votes = c(4818, 470, 6084, 2365, 1541), # 小選挙区
pr_votes = c(1695, 308, 5784, 2702,1414 ) # 比例代表
)
# 相関図の作成
ggplot(election_data, aes(x = pr_votes, y = district_votes, color = party)) +
geom_point(size = 4) +
# show.legend = FALSE を追加して凡例の「a」を消去
geom_text(aes(label = candidate), vjust = -1.5, family = "HiraKakuProN-W3", show.legend = FALSE) +
# COR=1.0 の基準線 (y = x)
geom_abline(intercept = 0, slope = 1, linetype = "dashed", color = "red", alpha = 0.5) +
scale_x_continuous(limits = c(-100, 7000)) +
scale_y_continuous(limits = c(-100, 7000)) +
scale_color_manual(values = c("自民" = "#3536C1", "維新" = "#31D925",
"中道" = "#E9EB3B", "社民" = "#EB3B3B", "参政" = "orange")) +
coord_fixed() +
labs(title = "政党得票数と候補者得票数の相関図（苅田町）",
subtitle = "第51回衆院選 (2026)\n※ 赤い破線は COR = 1.0 (小選挙区票 = 比例票) を示す",
x = "政党得票数（比例代表）",
y = "候補者得票数（小選挙区）",
caption = AONF_CAPTION) +
# 3. AONF形式: themeの一括指定
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(legend.position = "bottom",
plot.caption = element_text(size = 8, hjust = 1, color = "gray30"))
地域比較
# 必要なパッケージの読み込み
# インストールされていない場合は install.packages("ggplot2") を実行してください
library(ggplot2)
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# 1. データの作成
# これまでの会話やソースで確認された中道改革連合候補(辻氏)のCORデータ
df_tsuji_cor <- data.frame(
Municipality = factor(c("行橋市", "田川市", "苅田町"), levels = c("行橋市", "田川市", "苅田町")),
COR = c(0.82, 0.79, 0.88)
)
# 作成日の取得（資料のスクリプトスタイルを踏襲）
today <- as.character(Sys.Date())
# 2. 棒グラフの作成
ggplot(df_tsuji_cor, aes(x = Municipality, y = COR, fill = Municipality)) +
# 棒グラフの描画
geom_bar(stat = "identity", width = 0.5) +
# 棒の上に数値を表示
geom_text(aes(label = COR), vjust = -0.5, size = 5) +
# 基準となるCOR 1.0のラインを赤い点線で追加
geom_hline(yintercept = 1.0, linetype = "dashed", color = "red", size = 1) +
# タイトルとラベルの設定
labs(
title = "中道改革連合候補(辻智之氏)の地域別COR比較",
# subtitle = paste("作成日:", today, " - 点線はCOR 1.0 (政党基礎票と等しいライン)"),
subtitle = "点線はCOR 1.0 (政党基礎票と等しいライン)",
x = "開票区",
y = "COR (候補者超過達成比率)",
caption = AONF_CAPTION) +
# Y軸の範囲を設定 (1.0のラインが見えるように上限を1.1に)
scale_y_continuous(limits = c(0, 1.1)) +
# 棒の色を地域ごとに指定
scale_fill_manual(values = c("行橋市" = "steelblue", "田川市" = "darkorange", "苅田町" = "forestgreen")) +
# AONF標準テーマ設定
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
# キャプションのスタイルを統一
plot.caption = element_text(size = 8, color = "grey30", lineheight = 1.2, margin = margin(t = 15))
)## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
得票率
田川市
library(ggplot2)
library(dplyr)##
## 次のパッケージを付け加えます: 'dplyr'## 以下のオブジェクトは 'package:stats' からマスクされています:
##
## filter, lag## 以下のオブジェクトは 'package:base' からマスクされています:
##
## intersect, setdiff, setequal, unionlibrary(scales)
# AONF形式のキャプション定義
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# データの読み込み（ヘッダーなし、V1:政党名, V2:得票数）
df <- read.csv("田川市.csv", header = FALSE, stringsAsFactors = FALSE)
colnames(df) <- c("Party", "Votes")
# 1. 得票数のクリーニング（カンマの除去）と数値変換
df$Votes <- as.numeric(gsub(",", "", df$Votes))
# 2. 得票率の計算
df <- df %>%
mutate(Percentage = (Votes / sum(Votes)) * 100) %>%
arrange(desc(Percentage)) # 降順にソート
# 3. グラフの作成（AONF/標準フォーマット適用）
p <- ggplot(df, aes(x = reorder(Party, Percentage), y = Percentage)) +
geom_bar(stat = "identity", fill = "steelblue") +
coord_flip() + # 横棒グラフ
geom_text(aes(label = sprintf("%.1f%%", Percentage)), hjust = -0.1) +
scale_y_continuous(expand = expansion(mult = c(0, 0.15))) +
labs(
title = "田川市 比例代表 各党得票率(衆院選2026年)",
x = "政党名",
y = "得票率 (%)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
plot.caption = element_text(size = 8, hjust = 1, color = "gray30", lineheight = 1.2)
)
# グラフの表示
print(p)
# 計算結果の確認（コンソール出力用）
print(df)## Party Votes Percentage
## 1 自民党 7322 36.321246
## 2 中道 3794 18.820378
## 3 維新 2197 10.898358
## 4 参政 1530 7.589662
## 5 民主党 1140 5.655042
## 6 共産党 1054 5.228434
## 7 社民党 860 4.266085
## 8 れいわ 816 4.047820
## 9 みらい 771 3.824594
## 10 保守党 406 2.013989
## 11 ゆうこく連合 269 1.334392行橋市
library(ggplot2)
library(dplyr)
library(scales)
# 1. AONF_CAPTION 変数の定義
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# データの読み込み
# 行橋.tsv は タブ区切り、ヘッダーなし、2列目（得票数）にカンマを含む形式を想定
data <- read.table("行橋.tsv", sep="\t", header=FALSE, col.names=c("party", "votes"), stringsAsFactors=FALSE)
# 文字列から数値を抽出（カンマを除去して数値型に変換）
data$votes_num <- as.numeric(gsub(",", "", data$votes))
# 得票率の計算と並び替え
data <- data %>%
mutate(percentage = (votes_num / sum(votes_num)) * 100) %>%
arrange(desc(percentage))
# 政党名の順序を固定（グラフの並び順に反映）
data$party <- factor(data$party, levels = rev(data$party))
# グラフ作成
ggplot(data, aes(x = party, y = percentage)) +
geom_bar(stat = "identity", fill = "steelblue") +
geom_text(aes(label = sprintf("%.1f%%", percentage)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, max(data$percentage) + 5)) +
# 2. labs(caption = AONF_CAPTION) の使用
labs(title = "行橋市 比例代表 各党得票率(衆院選2026年)",
x = "政党名",
y = "得票率 (%)",
caption = AONF_CAPTION) +
# 3. theme_minimal(base_family = ...) と theme(legend.position = "bottom", plot.caption = ...) の適用
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
plot.caption = element_text(size = 8, hjust = 1, color = "gray30", lineheight = 1.2),
plot.title = element_text(hjust = 0.5, face = "bold"),
axis.text = element_text(size = 10)
)
苅田町
library(ggplot2)
library(dplyr)
library(scales)
library(readr)##
## 次のパッケージを付け加えます: 'readr'## 以下のオブジェクトは 'package:scales' からマスクされています:
##
## col_factor# AONF形式のキャプション定義
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# データの読み込み（ヘッダーなし、V1:政党名, V2:得票数）
d <- read_csv("苅田町.csv")## New names:
## • `` -> `...1`
## • `` -> `...2`## Rows: 11 Columns: 2
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): ...1
## num (1): ...2
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.df <- as.data.frame(d)
colnames(df) <- c("Party", "Votes")
# 1. 得票数のクリーニング（カンマの除去）と数値変換
df$Votes <- as.numeric(gsub(",", "", df$Votes))
# 2. 得票率の計算
df <- df %>%
mutate(Percentage = (Votes / sum(Votes)) * 100) %>%
arrange(desc(Percentage)) # 降順にソート
# 3. グラフの作成（AONF/標準フォーマット適用）
p <- ggplot(df, aes(x = reorder(Party, Percentage), y = Percentage)) +
geom_bar(stat = "identity", fill = "steelblue") +
coord_flip() + # 横棒グラフ
geom_text(aes(label = sprintf("%.1f%%", Percentage)), hjust = -0.1) +
scale_y_continuous(expand = expansion(mult = c(0, 0.15))) +
labs(
title = "苅田町 比例代表 各党得票率(衆院選2026年)",
x = "政党名",
y = "得票率 (%)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
plot.caption = element_text(size = 8, hjust = 1, color = "gray30", lineheight = 1.2)
)
# グラフの表示
print(p)
# 計算結果の確認（コンソール出力用）
print(df)## Party Votes Percentage
## 1 自由民主党 5784 37.702888
## 2 中道改革連合 2702 17.612933
## 3 日本維新の会 1695 11.048823
## 4 参政党 1414 9.217131
## 5 国民民主党 1252 8.161137
## 6 チームみらい 754 4.914934
## 7 れいわ新選組 443 2.887687
## 8 日本保守党 414 2.698651
## 9 日本共産党 386 2.516133
## 10 社会民主党 308 2.007692
## 11 減税日本・ゆうこく連合 189 1.2319932. 県選管データ
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ forcats 1.0.1 ✔ stringr 1.6.0
## ✔ lubridate 1.9.4 ✔ tibble 3.3.0
## ✔ purrr 1.2.0 ✔ tidyr 1.3.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ readr::col_factor() masks scales::col_factor()
## ✖ purrr::discard() masks scales::discard()
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(scales)
# AONF形式の定義
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# データの読み込み
df <- read.csv("fukuoka11ku2026.csv", check.names = FALSE)
# データの加工：カンマ削除、数値変換、得票率の計算
df_long <- df %>%
mutate(across(-Municipality, ~as.numeric(gsub(",", "", .)))) %>%
pivot_longer(-Municipality, names_to = "Candidate", values_to = "Votes") %>%
group_by(Municipality) %>%
mutate(Share = Votes / sum(Votes)) %>%
ungroup() %>%
# 自治体名の空白をトリミング（必要に応じて）
mutate(Municipality = trimws(Municipality))
# グラフ作成
p <- ggplot(df_long, aes(x = Municipality, y = Share, fill = Candidate)) +
geom_bar(stat = "identity", position = "fill") +
scale_y_continuous(labels = percent) +
labs(
title = "福岡11区 自治体別候補者得票率",
subtitle = "衆議院議員選挙2026年",
x = "自治体",
y = "得票率",
fill = "候補者",
caption = AONF_CAPTION
) +
# AONF標準形式のテーマ適用
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1),
axis.text.x = element_text(angle = 45, hjust = 1)
)
# グラフの保存
ggsave("fukuoka11ku_voteshare.png", p, width = 10, height = 7, dpi = 300)library(tidyverse)
library(scales)
# AONF形式の定義
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# データの読み込み
df <- read.csv("fukuoka11ku2026.csv", check.names = FALSE)
# データの加工
df_clean <- df %>%
# カンマを削除して数値化
mutate(across(-Municipality, ~as.numeric(gsub(",", "", .)))) %>%
# 自治体ごとの総得票数を計算
mutate(Total_Votes = rowSums(select(., -Municipality))) %>%
# 自治体名の空白をトリミング
mutate(Municipality = trimws(Municipality))
# プロット用にロング形式に変換し、総得票数で並べ替え
df_long <- df_clean %>%
pivot_longer(cols = c(Tsuji, Takeda, Shiki, Inoue, Murakami),
names_to = "Candidate", values_to = "Votes") %>%
mutate(Share = Votes / Total_Votes) %>%
# MunicipalityをTotal_Votesの降順で因子(factor)化する
mutate(Municipality = fct_reorder(Municipality, Total_Votes, .desc = TRUE))
# グラフ作成
p <- ggplot(df_long, aes(x = Municipality, y = Share, fill = Candidate)) +
geom_bar(stat = "identity", position = "fill") +
scale_y_continuous(labels = percent) +
labs(
title = "2026年衆議院議員選挙福岡11区 自治体別候補者得票率",
subtitle = "左から総得票数が多い順に自治体を並べています",
x = "自治体",
y = "得票率",
fill = "候補者",
caption = AONF_CAPTION
) +
# AONF標準形式のテーマ適用
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1),
axis.text.x = element_text(angle = 45, hjust = 1)
)
# グラフの保存
ggsave("fukuoka11ku_voteshare_sorted.png", p, width = 10, height = 7, dpi = 300)候補者と政党のデータを合併
library(tidyverse)
# 1. データの読み込み
# Note: read_csvはカンマ入りの数値を文字列として読み込む場合があります
df_candidates <- read_csv("fukuoka11ku2026.csv")## Rows: 15 Columns: 6
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): Municipality
## num (5): Tsuji, Takeda, Shiki, Inoue, Murakami
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.df_parties <- read_csv("福岡11区2026年政党.csv")## Rows: 15 Columns: 12
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): municipality
## dbl (4): れいわ新選組, 日本保守党, 減税日本・ゆうこく連合, 社会民主党
## num (7): 自由民主党, 中道改革連合, チームみらい, 日本維新の会, 参政党, 国民民主党, 日本共産党...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.# 2. 市町村名のクリーニング
# fukuoka11ku2026.csv の地域名に含まれる全角・半角スペースを削除します
df_candidates <- df_candidates %>%
mutate(Municipality = str_trim(Municipality))
# 3. 合併（マージ）
# 列名が 'Municipality' と 'municipality' で異なるため明示的に結合キーを指定します
df_merged <- df_candidates %>%
inner_join(df_parties, by = c("Municipality" = "municipality"))
# 4. 数値データの整形（カンマの除去と数値型への変換）
# 地域名以外の列で、カンマが含まれる文字列を数値に変換します
df_merged <- df_merged %>%
mutate(across(-Municipality, ~ {
if(is.character(.)) {
as.numeric(str_remove_all(., ","))
} else {
.
}
}))
# 5. 結果の書き出し
write_csv(df_merged, "fukuoka11ku2026_merged.csv")
# データの確認
print(head(df_merged))## # A tibble: 6 × 17
## Municipality Tsuji Takeda Shiki Inoue Murakami れいわ新選組 日本保守党
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 田川市 2981 8654 1113 1362 6109 816 406
## 2 行橋市 4751 13915 1041 2843 11815 965 822
## 3 豊前市 1528 5506 440 820 3774 381 271
## 4 香春町 631 2667 201 309 1302 182 97
## 5 添田町 576 2127 148 217 1230 148 78
## 6 糸田町 642 1939 191 255 863 180 65
## # ℹ 9 more variables: 自由民主党 <dbl>, 中道改革連合 <dbl>,
## # `減税日本・ゆうこく連合` <dbl>, チームみらい <dbl>, 社会民主党 <dbl>,
## # 日本維新の会 <dbl>, 参政党 <dbl>, 国民民主党 <dbl>, 日本共産党 <dbl>knitr::kable(df_merged)| Municipality | Tsuji | Takeda | Shiki | Inoue | Murakami | れいわ新選組 | 日本保守党 | 自由民主党 | 中道改革連合 | 減税日本・ゆうこく連合 | チームみらい | 社会民主党 | 日本維新の会 | 参政党 | 国民民主党 | 日本共産党 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 田川市 | 2981 | 8654 | 1113 | 1362 | 6109 | 816 | 406 | 7322 | 3794 | 269 | 771 | 860 | 2197 | 1530 | 1140 | 1054 |
| 行橋市 | 4751 | 13915 | 1041 | 2843 | 11815 | 965 | 822 | 13362 | 5787 | 461 | 1553 | 839 | 4202 | 2799 | 2515 | 963 |
| 豊前市 | 1528 | 5506 | 440 | 820 | 3774 | 381 | 271 | 4813 | 2051 | 175 | 542 | 413 | 1403 | 949 | 650 | 326 |
| 香春町 | 631 | 2667 | 201 | 309 | 1302 | 182 | 97 | 2041 | 1065 | 62 | 176 | 209 | 481 | 346 | 233 | 199 |
| 添田町 | 576 | 2127 | 148 | 217 | 1230 | 148 | 78 | 1732 | 909 | 38 | 141 | 135 | 467 | 262 | 184 | 184 |
| 糸田町 | 642 | 1939 | 191 | 255 | 863 | 180 | 65 | 1400 | 975 | 67 | 115 | 167 | 330 | 263 | 197 | 124 |
| 川崎町 | 935 | 3348 | 283 | 349 | 1516 | 260 | 88 | 2473 | 1653 | 84 | 174 | 216 | 555 | 402 | 258 | 258 |
| 大任町 | 175 | 1828 | 57 | 106 | 415 | 119 | 36 | 1291 | 464 | 22 | 58 | 68 | 177 | 159 | 117 | 44 |
| 赤村 | 133 | 878 | 59 | 57 | 378 | 42 | 25 | 646 | 278 | 16 | 70 | 66 | 123 | 88 | 77 | 35 |
| 福智町 | 1361 | 4955 | 472 | 655 | 2223 | 410 | 203 | 3709 | 1949 | 149 | 308 | 429 | 824 | 717 | 499 | 390 |
| 苅田町 | 2365 | 6084 | 470 | 1541 | 4818 | 443 | 414 | 5784 | 2702 | 189 | 754 | 308 | 1695 | 1414 | 1252 | 386 |
| みやこ町 | 1281 | 4181 | 271 | 566 | 2917 | 261 | 160 | 3743 | 1706 | 124 | 378 | 264 | 1083 | 621 | 531 | 284 |
| 吉富町 | 436 | 1336 | 107 | 249 | 878 | 87 | 68 | 1146 | 580 | 45 | 136 | 72 | 262 | 264 | 193 | 156 |
| 上毛町 | 459 | 1855 | 124 | 311 | 1112 | 142 | 93 | 1632 | 602 | 47 | 183 | 123 | 380 | 299 | 232 | 131 |
| 築上町 | 757 | 3146 | 210 | 430 | 3652 | 204 | 167 | 3339 | 1229 | 113 | 331 | 236 | 1306 | 578 | 442 | 212 |
2.1 対応分析等
福岡11区：候補者と政党の得票構造（ハブ分析）
library(readr)
library(ggplot2)
library(FactoMineR)
library(patchwork)
library(vegan)## 要求されたパッケージ permute をロード中ですlibrary(igraph)##
## 次のパッケージを付け加えます: 'igraph'## 以下のオブジェクトは 'package:vegan' からマスクされています:
##
## diversity## 以下のオブジェクトは 'package:permute' からマスクされています:
##
## permute## 以下のオブジェクトは 'package:lubridate' からマスクされています:
##
## %--%, union## 以下のオブジェクトは 'package:purrr' からマスクされています:
##
## compose, simplify## 以下のオブジェクトは 'package:tidyr' からマスクされています:
##
## crossing## 以下のオブジェクトは 'package:tibble' からマスクされています:
##
## as_data_frame## 以下のオブジェクトは 'package:dplyr' からマスクされています:
##
## as_data_frame, groups, union## 以下のオブジェクトは 'package:stats' からマスクされています:
##
## decompose, spectrum## 以下のオブジェクトは 'package:base' からマスクされています:
##
## unionlibrary(stringr)
library(dplyr)
# --- 定数の定義 (AONF format) ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# --- データの準備 ---
data <- read_csv("fukuoka11ku2026_merged_a.csv")## Rows: 15 Columns: 17## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): Municipality
## dbl (16): Tsuji, Takeda, Shiki, Inoue, Murakami, れいわ, 保守党, 自民党, 中道, ゆうこく連合, ...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.data_clean <- data %>%
mutate(across(-Municipality, ~ as.numeric(str_remove_all(as.character(.), ","))))
voting_data <- as.data.frame(data_clean[,-1])
rownames(voting_data) <- data_clean$Municipality
# --- 1. 対応分析 (CA) とクラスタリング ---
ca_res <- CA(voting_data, graph = FALSE)
ca_full_coord <- ca_res$col$coord
ca_hc <- hclust(dist(ca_full_coord), method = "ward.D2")
ca_cluster <- factor(cutree(ca_hc, k = 3))
ca_df <- data.frame(
Dim1 = ca_full_coord[,1],
Dim2 = ca_full_coord[,2],
Name = rownames(ca_full_coord),
Cluster = ca_cluster
)
# --- 2. MST (最小全域木) データ作成 ---
get_mst_segments <- function(df, dist_obj) {
mst <- spantree(dist_obj)
n <- nrow(df)
data.frame(
x = df$Dim1[2:n],
y = df$Dim2[2:n],
xend = df$Dim1[mst$kid],
yend = df$Dim2[mst$kid]
)
}
d_ca <- dist(ca_full_coord)
mst_ca <- spantree(d_ca)
n <- nrow(ca_df)
edges <- cbind(mst_ca$kid, 2:n)
g_mst <- graph_from_edgelist(edges, directed = FALSE)
ca_df$Degree <- degree(g_mst)
# --- 3. 可視化 (Cluster凡例の非表示とスケール固定) ---
p_hub <- ggplot(ca_df, aes(x = Dim1, y = Dim2)) +
# MSTのネットワーク線
geom_segment(data = get_mst_segments(ca_df, d_ca),
aes(x = x, y = y, xend = xend, yend = yend),
color = "gray85", linetype = "solid") +
# 次数の大きさを点のサイズに反映
geom_point(aes(color = Cluster, size = Degree), alpha = 0.7) +
# ラベルの描画
geom_text(aes(label = Name, color = Cluster),
vjust = -1.8, fontface = "bold",
family = "HiraKakuProN-W3", show.legend = FALSE) +
scale_size_continuous(range = c(4, 10)) +
scale_color_brewer(palette = "Set1") +
# ★重要：Clusterの凡例（color）を非表示にする
guides(color = "none") +
# 縦横比を1:1に固定
coord_fixed(ratio = 1, clip = "off") +
expand_limits(x = range(ca_df$Dim1)*1.5, y = range(ca_df$Dim2)*1.5) +
# グラフタイトルとAONF形式のキャプション設定
labs(title = "福岡11区：候補者と政党の得票構造（ハブ分析）",
subtitle = "Correspondence Analysis & MST: Hub Identification",
x = "Dimension 1",
y = "Dimension 2",
size = "次数 (Hub Strength)",
caption = AONF_CAPTION) +
# AONF標準テーマ
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(legend.position = "bottom",
plot.caption = element_text(size = 8, hjust = 1, color = "gray30", lineheight = 1.2))
# グラフの表示
print(p_hub)
Greenacre’s Symbiplot
# [Greenacre's Symbiplot]
# AONF_CAPTION の定義
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
#####################################
### symbiplot(Greenacre)
#####################################
if (!require("ca")) install.packages("ca")## 要求されたパッケージ ca をロード中です##
## 次のパッケージを付け加えます: 'ca'## 以下のオブジェクトは 'package:vegan' からマスクされています:
##
## caif (!require("ape")) install.packages("ape")## 要求されたパッケージ ape をロード中です##
## 次のパッケージを付け加えます: 'ape'## 以下のオブジェクトは 'package:igraph' からマスクされています:
##
## degree, edges, mst, ring## 以下のオブジェクトは 'package:dplyr' からマスクされています:
##
## whereif (!require("ggrepel")) install.packages("ggrepel")## 要求されたパッケージ ggrepel をロード中ですlibrary(ca)
library(ape)
library(ggplot2)
library(ggrepel)
# 1. データの読み込み
voting_data <- read.csv("fukuoka11ku2026_merged_a.csv", row.names = 1)
# 【重要】データ構造に合わせて修正
counts <- voting_data[, 1:ncol(voting_data)]
# 2. 対応分析 (CA) の実行
res_ca <- ca(counts)
# 3. Symbiplot用の座標計算
sv_sqrt <- diag(sqrt(res_ca$sv))
rows_coord <- as.data.frame(res_ca$rowcoord[, 1:2] %*% sv_sqrt[1:2, 1:2])
cols_coord <- as.data.frame(res_ca$colcoord[, 1:2] %*% sv_sqrt[1:2, 1:2])
colnames(rows_coord) <- c("Dim1", "Dim2")
rows_coord$Name <- rownames(rows_coord)
rows_coord$Group <- "City"
colnames(cols_coord) <- c("Dim1", "Dim2")
cols_coord$Name <- colnames(counts)
cols_coord$Group <- "Party/Candidate"
# 4. 最小全域木 (MST) の計算
full_principal_coords <- res_ca$colcoord %*% diag(res_ca$sv)
party_dist_chi2 <- dist(full_principal_coords)
st_tree <- mst(party_dist_chi2)
edges <- which(as.matrix(st_tree) == 1, arr.ind = TRUE)
edges <- edges[edges[,1] < edges[,2], ]
edge_df <- data.frame(
x = cols_coord$Dim1[edges[,1]], y = cols_coord$Dim2[edges[,1]],
xend = cols_coord$Dim1[edges[,2]], yend = cols_coord$Dim2[edges[,2]]
)
# 5. 可視化
ggplot() +
geom_segment(data = edge_df, aes(x=x, y=y, xend=xend, yend=yend),
color = "gray60", linetype = "dotted", linewidth = 0.5) +
geom_point(data = rows_coord, aes(x=Dim1, y=Dim2), color = "blue", alpha = 0.3, size = 1.5) +
geom_text_repel(data = rows_coord, aes(x=Dim1, y=Dim2, label = Name),
color = "blue", size = 2.5, alpha = 0.6, family="HiraKakuProN-W3") +
geom_point(data = cols_coord, aes(x=Dim1, y=Dim2), color = "red", size = 3) +
geom_text_repel(data = cols_coord, aes(x=Dim1, y=Dim2, label = Name),
color = "red", size = 3.5, fontface = "bold", family="HiraKakuProN-W3",
box.padding = 0.5, max.overlaps = Inf) +
geom_vline(xintercept = 0, linetype = "dashed", color = "gray80") +
geom_hline(yintercept = 0, linetype = "dashed", color = "gray80") +
coord_fixed() +
theme_minimal(base_family="HiraKakuProN-W3") +
theme(legend.position = "bottom", plot.caption = element_text(size = 7, color = "gray30")) +
labs(title = "Greenacre's Symbiplot with MST",
subtitle = "Scaling: Standard Coordinates * sqrt(Singular Values)",
x = paste0("Dimension 1 (", round((res_ca$sv[1]^2/sum(res_ca$sv^2))*100, 1), "%)"),
y = paste0("Dimension 2 (", round((res_ca$sv[2]^2/sum(res_ca$sv^2))*100, 1), "%)"),
caption = AONF_CAPTION)
対応分析のみ(2026年データ)
library(ggrepel) # ラベル重なり防止用
# --- 定数の定義 (AONF format) ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# --- 1. 対応分析 (CA) とデータ準備 ---
# （計算部分は前回の内容を継承）
ca_res <- FactoMineR::CA(voting_data, graph = FALSE)
ca_full_coord <- ca_res$col$coord
ca_hc <- hclust(dist(ca_full_coord), method = "ward.D2")
ca_cluster <- factor(cutree(ca_hc, k = 3))
ca_df <- data.frame(
Dim1 = ca_full_coord[,1],
Dim2 = ca_full_coord[,2],
Name = rownames(ca_full_coord),
Cluster = ca_cluster
)
# --- 2. グラフ作成 (ggrepel適用版) ---
p_ca_repel <- ggplot(ca_df, aes(x = Dim1, y = Dim2, color = Cluster)) +
# 原点を通る補助線
geom_hline(yintercept = 0, linetype = "dashed", color = "gray85") +
geom_vline(xintercept = 0, linetype = "dashed", color = "gray85") +
# 点のプロット
geom_point(size = 3.5, alpha = 0.8) +
# ggrepelによるラベル描画
geom_text_repel(aes(label = Name),
family = "HiraKakuProN-W3",
fontface = "bold",
box.padding = 0.5, # ラベル周りの余白
point.padding = 0.3, # 点とラベルの間の余白
segment.color = "gray50", # ラベルから点へ引く線の色
size = 4,
show.legend = FALSE) +
# スケール・色の設定
scale_color_brewer(palette = "Set1") +
# 重要：縦横比を1:1に固定
coord_fixed(ratio = 1) +
# タイトルとAONF形式のキャプション設定
labs(title = "福岡11区：対応分析による得票構造の可視化",
subtitle = "衆院選2026年福岡11区",
x = paste0("Dimension 1 (", round(ca_res$eig[1, 2], 1), "%)"),
y = paste0("Dimension 2 (", round(ca_res$eig[2, 2], 1), "%)"),
color = "Cluster",
caption = AONF_CAPTION) +
# AONF標準テーマ
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(legend.position = "bottom",
plot.caption = element_text(size = 8, hjust = 1, color = "gray30", lineheight = 1.2))
# グラフの表示
print(p_ca_repel)
対応分析のみ(2024年データ)
library(readr)
library(ggplot2)
library(FactoMineR)
library(ggrepel)
library(dplyr)
# --- 1. 定数の定義 (AONF format) ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
# --- 2. データの準備 ---
data <- read_csv("data11ku_PC.csv")## New names:
## Rows: 15 Columns: 14
## ── Column specification
## ────────────────────────────────────────────────────────
## Delimiter: "," chr (2): ...1, Var.1 dbl (12): JCP, RS, CDP, DPP, JIP, SDP, LDP,
## NKP, SAN, Shiki, Takeda, Murakami
## ℹ Use `spec()` to retrieve the full column specification for this data. ℹ
## Specify the column types or set `show_col_types = FALSE` to quiet this message.
## • `` -> `...1`voting_data <- as.data.frame(data[,-1])
rownames(voting_data) <- data[[1]]
# 数値データのみを抽出
voting_numeric <- voting_data[, 2:ncol(voting_data)]
# --- 3. 対応分析 (CA) とクラスタリング ---
ca_res <- CA(voting_numeric, graph = FALSE)
# 列変数（政党・候補者等）の座標を取得
ca_full_coord <- ca_res$col$coord
# 階層的クラスター分析 (Ward法、3グループ)
ca_hc <- hclust(dist(ca_full_coord), method = "ward.D2")
ca_cluster <- factor(cutree(ca_hc, k = 3))
# 可視化用データフレーム
ca_df <- data.frame(
Dim1 = ca_full_coord[,1],
Dim2 = ca_full_coord[,2],
Name = rownames(ca_full_coord),
Cluster = ca_cluster
)
# --- 4. 可視化 (CA + ggrepel + AONF format) ---
p_ca_clean <- ggplot(ca_df, aes(x = Dim1, y = Dim2, color = Cluster)) +
# 原点を通る補助線
geom_hline(yintercept = 0, linetype = "dashed", color = "gray85") +
geom_vline(xintercept = 0, linetype = "dashed", color = "gray85") +
# データのプロット
geom_point(size = 3.5, alpha = 0.8) +
# ggrepelによるラベル描画
geom_text_repel(aes(label = Name),
family = "HiraKakuProN-W3",
fontface = "bold",
box.padding = 0.5,
point.padding = 0.3,
segment.color = "gray50",
show.legend = FALSE) +
# スケール・色の設定
scale_color_brewer(palette = "Set1") +
# 重要：縦横比を1:1に固定
coord_fixed(ratio = 1) +
# タイトルとAONF形式のキャプション設定
labs(title = "福岡11区：対応分析による得票構造の可視化", subtitle = "衆院選2024年福岡11区",
x = paste0("Dimension 1 (", round(ca_res$eig[1, 2], 1), "%)"),
y = paste0("Dimension 2 (", round(ca_res$eig[2, 2], 1), "%)"),
color = "Cluster",
caption = AONF_CAPTION) +
# AONF標準テーマ
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(legend.position = "bottom",
plot.caption = element_text(size = 8, hjust = 1, color = "gray30", lineheight = 1.2))
# --- 5. 実行 ---
print(p_ca_clean)
キャンプのswing
# --- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ca)
library(ggplot2)
library(ggrepel)
library(dplyr)
library(tidyr)
# 1. データの読み込み
d24 <- read.csv("data11ku_PC_fixed.csv")
d26 <- read.csv("fukuoka11ku2026_merged_a.csv")
# 2. 陣営（キャンプ）の完全統合
# 2024年
df24_all_camp <- d24 %>%
mutate(
自民キャンプ = LDP + Takeda,
中道キャンプ = CDP + NKP,
維新キャンプ = JIP + Murakami, # ← ここで統合
社民キャンプ = SDP + Shiki,
参政キャンプ = SAN,
民主 = DPP, 共産 = JCP, れいわ = RS
) %>%
select(Municipality = 市町村, 自民キャンプ, 中道キャンプ, 維新キャンプ,
社民キャンプ, 参政キャンプ, 民主, 共産, れいわ) %>%
mutate(Year = "2024")
# 2026年
df26_all_camp <- d26 %>%
mutate(
自民キャンプ = 自民党 + Takeda,
中道キャンプ = 中道 + Tsuji,
維新キャンプ = 維新 + Murakami, # ← ここで統合
社民キャンプ = 社民党 + Shiki,
参政キャンプ = 参政 + Inoue,
民主 = 民主党, 共産 = 共産党,
保守党 = 保守党, ゆうこく連合 = ゆうこく連合, みらい = みらい
) %>%
select(Municipality, 自民キャンプ, 中道キャンプ, 維新キャンプ,
社民キャンプ, 参政キャンプ, 民主, 共産, れいわ,
保守党, ゆうこく連合, みらい) %>%
mutate(Year = "2026")
# 3. 共通行列の作成とCA実行
combined <- bind_rows(df24_all_camp, df26_all_camp) %>% replace(is.na(.), 0)
row_labels <- paste0(combined$Municipality, "_", combined$Year)
counts_matrix <- combined %>% select(-Municipality, -Year)
rownames(counts_matrix) <- row_labels
res_ca <- ca(counts_matrix)
sv_sqrt <- diag(sqrt(res_ca$sv))
rows_coord <- as.data.frame(res_ca$rowcoord[, 1:2] %*% sv_sqrt[1:2, 1:2])
cols_coord <- as.data.frame(res_ca$colcoord[, 1:2] %*% sv_sqrt[1:2, 1:2])
colnames(rows_coord) <- c("Dim1", "Dim2")
rows_coord$FullName <- rownames(counts_matrix)
rows_coord <- rows_coord %>% separate(FullName, into = c("City", "Year"), sep = "_")
colnames(cols_coord) <- c("Dim1", "Dim2")
cols_coord$Name <- colnames(counts_matrix)
# 4. スイングデータの抽出
swing_data <- rows_coord %>%
pivot_wider(names_from = Year, values_from = c(Dim1, Dim2))
# 5. 可視化 (AONF標準形式)
ggplot() +
geom_segment(data = swing_data,
aes(x = Dim1_2024, y = Dim2_2024, xend = Dim1_2026, yend = Dim2_2026),
arrow = arrow(length = unit(0.2, "cm")), color = "blue", alpha = 0.4) +
geom_point(data = cols_coord, aes(x = Dim1, y = Dim2), color = "red", size = 3.5) +
geom_text_repel(data = cols_coord, aes(x = Dim1, y = Dim2, label = Name),
color = "red", size = 4, fontface = "bold", family="HiraKakuProN-W3") +
geom_text_repel(data = rows_coord %>% filter(Year == "2026"),
aes(x = Dim1, y = Dim2, label = City),
color = "blue", size = 2.8, alpha = 0.7, family="HiraKakuProN-W3") +
coord_fixed() +
theme_minimal(base_family="HiraKakuProN-W3") +
theme(legend.position = "bottom", plot.caption = element_text(size = 8, color = "gray30")) +
labs(title = "Fukuoka 11th District: 2024-2026 Full-Camp Swing",
subtitle = "All major forces aggregated: (Candidate + Party) for true structural comparison",
x = paste0("Dimension 1 (", round((res_ca$sv[1]^2/sum(res_ca$sv^2))*100, 1), "%)"),
y = paste0("Dimension 2 (", round((res_ca$sv[2]^2/sum(res_ca$sv^2))*100, 1), "%)"),
caption = AONF_CAPTION)
政党と候補者の分離版
# --- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ca)
library(dplyr)
library(tidyr)
# 1. データの読み込み
d24 <- read.csv("data11ku_PC_fixed.csv")
d26 <- read.csv("fukuoka11ku2026_merged_a.csv")
# 2. 候補者と政党を分離したまま名寄せ
# 2024年 (立憲+公明を「中道」とする)
df24_sep <- d24 %>%
mutate(中道 = CDP + NKP) %>%
select(Municipality = 市町村, 自民 = LDP, 維新 = JIP, 民主 = DPP,
社民 = SDP, 共産 = JCP, れいわ = RS, 参政 = SAN,
中道, Takeda, Shiki, Murakami) %>%
mutate(Year = "2024")
# 2026年 (候補者と政党を別々に保持)
df26_sep <- d26 %>%
select(Municipality, 自民 = 自民党, 維新, 民主 = 民主党,
社民 = 社民党, 共産 = 共産党, れいわ, 参政,
中道, Takeda, Shiki, Murakami, Tsuji, Inoue) %>%
mutate(Year = "2026")
# 共通行列の作成
combined_sep <- bind_rows(df24_sep, df26_sep) %>% replace(is.na(.), 0)
row_labels <- paste0(combined_sep$Municipality, "_", combined_sep$Year)
counts_matrix <- combined_sep %>% select(-Municipality, -Year)
rownames(counts_matrix) <- row_labels
# 3. 対応分析 (CA)
res_ca_sep <- ca(counts_matrix)
# 4. 座標（列＝政党・候補者）の抽出
sv_sqrt <- diag(sqrt(res_ca_sep$sv))
cols_coord <- as.data.frame(res_ca_sep$colcoord[, 1:2] %*% sv_sqrt[1:2, 1:2])
# --- 列名の割り当て
colnames(cols_coord) <- c("Dim1", "Dim2")
# ------------------------------------
cols_coord$Name <- colnames(counts_matrix)
# 5. 座標値を表示（Dim1で昇順ソート）
print("--- 候補者・政党 分離版 Dim1座標 (右ほど保守/地域回帰傾向) ---")## [1] "--- 候補者・政党 分離版 Dim1座標 (右ほど保守/地域回帰傾向) ---"print(cols_coord %>% select(Name, Dim1) %>% arrange(Dim1))## Name Dim1
## Shiki Shiki -0.91726650
## れいわ れいわ -0.74299085
## 中道 中道 -0.48117766
## 社民 社民 -0.39978629
## 共産 共産 -0.32430642
## Murakami Murakami -0.29669661
## 維新 維新 -0.07824198
## 民主 民主 -0.03956648
## Takeda Takeda 0.05465939
## 自民 自民 0.41192466
## 参政 参政 0.72341776
## Tsuji Tsuji 1.75627578
## Inoue Inoue 1.78939222# --- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ca)
library(ggplot2)
library(ggrepel)
library(dplyr)
library(tidyr)
# 1. データの読み込み
d24 <- read.csv("data11ku_PC_fixed.csv")
d26 <- read.csv("fukuoka11ku2026_merged_a.csv")
# 2. 候補者と政党を「分離」したまま名寄せ
df24_sep <- d24 %>%
mutate(中道 = CDP + NKP) %>%
select(Municipality = 市町村, 自民党 = LDP, 維新 = JIP, 民主党 = DPP,
社民党 = SDP, 共産党 = JCP, れいわ = RS, 参政 = SAN,
中道, Takeda, Shiki, Murakami) %>%
mutate(Year = "2024")
df26_sep <- d26 %>%
select(Municipality, 自民党, 維新, 民主党,
社民党, 共産党, れいわ, 参政,
中道, Takeda, Shiki, Murakami, Tsuji, Inoue) %>%
mutate(Year = "2026")
# 共通行列の作成
combined_sep <- bind_rows(df24_sep, df26_sep) %>% replace(is.na(.), 0)
row_labels <- paste0(combined_sep$Municipality, "_", combined_sep$Year)
counts_matrix <- combined_sep %>% select(-Municipality, -Year)
rownames(counts_matrix) <- row_labels
# 3. 対応分析 (CA)
res_ca_sep <- ca(counts_matrix)
sv_sqrt <- diag(sqrt(res_ca_sep$sv))
rows_coord <- as.data.frame(res_ca_sep$rowcoord[, 1:2] %*% sv_sqrt[1:2, 1:2])
cols_coord <- as.data.frame(res_ca_sep$colcoord[, 1:2] %*% sv_sqrt[1:2, 1:2])
colnames(rows_coord) <- c("Dim1", "Dim2")
rows_coord$FullName <- rownames(counts_matrix)
rows_coord <- rows_coord %>% separate(FullName, into = c("City", "Year"), sep = "_")
colnames(cols_coord) <- c("Dim1", "Dim2")
cols_coord$Name <- colnames(counts_matrix)
# 4. 可視化
ggplot() +
# 2024 -> 2026 のスイング
geom_segment(data = rows_coord %>% pivot_wider(names_from = Year, values_from = c(Dim1, Dim2)),
aes(x = Dim1_2024, y = Dim2_2024, xend = Dim1_2026, yend = Dim2_2026),
arrow = arrow(length = unit(0.2, "cm")), color = "blue", alpha = 0.2) +
# 政党・候補者の点
geom_point(data = cols_coord, aes(x = Dim1, y = Dim2), color = "red", size = 3) +
geom_text_repel(data = cols_coord, aes(x = Dim1, y = Dim2, label = Name),
color = "red", size = 3.5, fontface = "bold", family="HiraKakuProN-W3") +
# 2026年の自治体名
geom_text(data = rows_coord %>% filter(Year == "2026"),
aes(x = Dim1, y = Dim2, label = City),
color = "blue", size = 2.5, alpha = 0.5, family="HiraKakuProN-W3") +
theme_minimal(base_family="HiraKakuProN-W3") +
theme(legend.position = "bottom", plot.caption = element_text(size = 8)) +
labs(title = "Fukuoka 11th District: 2024-2026 Split Analysis",
subtitle = "Separated: 'Tsuji' (Candidate) and '中道' (Party) shows huge strategic gap",
caption = AONF_CAPTION)
2.2 最大の動員力
# --- AONF定数の定義 (Saved Informationに基づく) ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
#####################################
### 最大動員力（Mobilization Power）分析 - 軸調整版
#####################################
library(ggplot2)
library(dplyr)
library(tidyr)
# 1. データの読み込み
df26 <- read.csv("fukuoka11ku2026_merged_a.csv")
# 2. 陣営（キャンプ）ごとの「候補者票」と「政党票」の整理
camp_data <- df26 %>%
summarise(
自民_Candidate = sum(Takeda), 自民_Party = sum(自民党),
中道_Candidate = sum(Tsuji), 中道_Party = sum(中道),
維新_Candidate = sum(Murakami), 維新_Party = sum(維新),
参政_Candidate = sum(Inoue), 参政_Party = sum(参政),
社民_Candidate = sum(Shiki), 社民_Party = sum(社民党)
) %>%
pivot_longer(everything(), names_to = c("Camp", "Type"), names_sep = "_", values_to = "Votes")
# 3. 可視化：軸の範囲を15%拡張して数字の欠けを防止
ggplot(camp_data, aes(x = reorder(Camp, -Votes, sum), y = Votes, fill = Type)) +
geom_bar(stat = "identity", position = "stack", width = 0.7) +
# 合計値（最大動員力）を表示
stat_summary(fun = sum, aes(label = scales::comma(after_stat(y)), group = Camp),
geom = "text", vjust = -0.8, size = 3.5, fontface = "bold") +
# 軸の拡張設定（上方向に15%の余裕を持たせる）
scale_y_continuous(labels = scales::comma, expand = expansion(mult = c(0, 0.15))) +
scale_fill_manual(values = c("Candidate" = "#ff7f0e", "Party" = "#1f77b4"),
labels = c("Candidate" = "候補者票 (個人力)", "Party" = "政党票 (組織力)")) +
# AONF標準テーマ (Saved Informationに基づく)
theme_minimal(base_family="HiraKakuProN-W3") +
theme(legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)) +
labs(title = "福岡11区 2026：陣営別「最大動員力」の構造",
subtitle = "Candidate票とParty票の合算：陣営が有権者にリーチした延べ総数",
x = "陣営 (Camp)", y = "総得票（延べ数）", fill = "票の内訳",
caption = AONF_CAPTION)
2.3 得票転移
# --- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
#####################################
### 得票転移（Vote Transfer）推計分析
#####################################
library(ggplot2)
library(dplyr)
library(tidyr)
# 1. データの読み込み
d24 <- read.csv("data11ku_PC_fixed.csv")
d26 <- read.csv("fukuoka11ku2026_merged_a.csv")
# 2. 前処理：得票率（Share）の算出
# 2024年の「中道（立憲+公明）」の合計得票率を算出
d24_prep <- d24 %>%
mutate(Total_24 = CDP + NKP + LDP + JIP + DPP + SDP + JCP + RS + SAN,
Chudo_Share_24 = (CDP + NKP) / Total_24) %>%
select(Municipality = 市町村, Chudo_Share_24)
# 2026年の主要候補者の得票率を算出
d26_prep <- d26 %>%
mutate(Total_26 = Tsuji + Takeda + Shiki + Inoue + Murakami +
れいわ + 保守党 + 自民党 + 中道 + ゆうこく連合 + みらい +
社民党 + 維新 + 参政 + 民主党 + 共産党) %>%
mutate(Tsuji_S = Tsuji / Total_26,
Murakami_S = Murakami / Total_26,
Takeda_S = Takeda / Total_26,
Inoue_S = Inoue / Total_26) %>%
select(Municipality, Tsuji_S, Murakami_S, Takeda_S, Inoue_S)
# データの結合
transfer_df <- inner_join(d24_prep, d26_prep, by = "Municipality")
# 3. グッドマン回帰による転移係数の算出
# 「2024年中道票」が「2026年各候補」にどれだけ寄与したかを回帰係数(Slope)で推定
models <- list(
Tsuji = lm(Tsuji_S ~ Chudo_Share_24, data = transfer_df),
Murakami = lm(Murakami_S ~ Chudo_Share_24, data = transfer_df),
Takeda = lm(Takeda_S ~ Chudo_Share_24, data = transfer_df),
Inoue = lm(Inoue_S ~ Chudo_Share_24, data = transfer_df)
)
# 係数の抽出
transfer_rates <- data.frame(
Candidate = names(models),
Rate = sapply(models, function(m) coef(m)[2]) # 回帰係数（傾き）
)
# 4. 可視化：2024年中道支持層の「吸着度」
ggplot(transfer_rates, aes(x = reorder(Candidate, -Rate), y = Rate, fill = Candidate)) +
geom_bar(stat = "identity", width = 0.6) +
geom_hline(yintercept = 0, color = "gray50") +
theme_minimal(base_family="HiraKakuProN-W3") +
theme(legend.position = "none",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)) +
labs(title = "2024年「中道支持層」はどこへ流れたか（推計）",
subtitle = "2024年立憲・公明得票率との相関係数に基づく支持の継承度",
x = "2026年 候補者", y = "支持継承の強さ (Regression Coefficient)",
caption = AONF_CAPTION)
4つのグラフ
# --- AONF定数の定義 (Saved Informationに基づく) ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
library(dplyr)
library(tidyr)
library(ggrepel)
# 1. データの読み込み
d24 <- read.csv("data11ku_PC_fixed.csv")
d26 <- read.csv("fukuoka11ku2026_merged_a.csv")
# 2. 得票率の算出
d24_prep <- d24 %>%
mutate(Total_24 = CDP + NKP + LDP + JIP + DPP + SDP + JCP + RS + SAN,
Chudo_Share_24 = (CDP + NKP) / Total_24) %>%
select(Municipality = 市町村, Chudo_Share_24)
d26_prep <- d26 %>%
mutate(Total_26 = Tsuji + Takeda + Shiki + Inoue + Murakami +
れいわ + 保守党 + 自民党 + 中道 + ゆうこく連合 + みらい +
社民党 + 維新 + 参政 + 民主党 + 共産党) %>%
mutate(Tsuji = Tsuji / Total_26,
Murakami = Murakami / Total_26,
Takeda = Takeda / Total_26,
Inoue = Inoue / Total_26) %>%
select(Municipality, Tsuji, Murakami, Takeda, Inoue)
# データの結合と縦長形式への変換（Facet用）
plot_df <- inner_join(d24_prep, d26_prep, by = "Municipality") %>%
pivot_longer(cols = c(Tsuji, Murakami, Takeda, Inoue),
names_to = "Candidate", values_to = "Share_26")
# 3. 4つの散布図の可視化
ggplot(plot_df, aes(x = Chudo_Share_24, y = Share_26)) +
# 回帰直線の描画
geom_smooth(method = "lm", se = FALSE, color = "gray60", linetype = "dashed", linewidth = 0.5) +
# 自治体ポイント
geom_point(aes(color = Candidate), size = 2.5, alpha = 0.8) +
# 自治体名のラベル（Saved Informationに基づき日本語フォント指定）
geom_text_repel(aes(label = Municipality), size = 2.5, family = "HiraKakuProN-W3") +
# 候補者ごとにグラフを分割
facet_wrap(~ Candidate, scales = "free_y") +
# AONF標準テーマ (Saved Informationに基づく)
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(legend.position = "none",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1),
strip.background = element_rect(fill = "gray95", color = "white"),
strip.text = element_text(face = "bold")) +
labs(title = "2024年「中道支持層」の流動性：自治体別散布図",
subtitle = "X軸：2024年立憲・公明得票率 / Y軸：2026年各候補者得票率",
x = "2024年 中道票シェア（立憲+公明）",
y = "2026年 候補者個人得票率",
caption = AONF_CAPTION)## `geom_smooth()` using formula = 'y ~ x'
3. 政治的影響力（ハブの強さ）と実得票数の相関分析
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
library(ggrepel)
# 資料29Pの市町村別得票数から算出した正確な合計値
df_exact <- data.frame(
Name = c("自民党", "武田", "ゆうこく連合", "中道", "Tsuji", "Inoue",
"民主党", "参政", "保守党", "みらい", "維新", "Murakami", "共産党", "Shiki"),
HubStrength = c(2.5, 2.5, 3.0, 2.5, 1.0, 1.5, 2.0, 1.5, 1.5, 1.5, 1.0, 1.0, 1.0, 1.0),
TotalVotes = c(54433, 62419, 1861, 25744, 19011, 10070, 8540, 10691, 2993, 5690, 15485, 43002, 4746, 5187)
)
# 可視化の実行
ggplot(df_exact, aes(x = HubStrength, y = TotalVotes, label = Name)) +
# 点の描画（サイズを得票数に比例）
geom_point(aes(size = TotalVotes), color = "steelblue", alpha = 0.7) +
# ラベルの重なり防止
geom_text_repel(family = "HiraKakuProN-W3", box.padding = 0.6, max.overlaps = Inf) +
# サイズの範囲調整
scale_size_continuous(range = c(3, 15), guide = "none") +
# 縦軸の範囲設定（円が欠けないように75,000まで拡張）
scale_y_continuous(limits = c(0, 75000), labels = scales::comma) +
# AONF標準設定
labs(
title = "Hub Strength vs. Actual Votes (Fukuoka 11th District 2026)",
subtitle = "政治的影響力（ハブの強さ）と実得票数の相関分析",
x = "ハブの強さ (Hub Strength / 次数)",
y = "実得票数 (Total Votes)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
plot.caption = element_text(size = 8, hjust = 1, color = "gray30")
)
4. COR(候補者超過達成比率)
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
df_cor_fixed <- data.frame(
Candidate = c("Murakami", "Shiki", "Takeda", "Inoue", "Tsuji"),
COR = c(2.78, 1.18, 1.15, 0.94, 0.74)
)
# 可視化
ggplot(df_cor_fixed, aes(x = reorder(Candidate, -COR), y = COR, fill = COR > 1)) +
geom_bar(stat = "identity", width = 0.7) +
geom_hline(yintercept = 1.0, linetype = "dashed", color = "red") +
geom_text(aes(label = sprintf("%.2f", COR)), vjust = -0.5, family = "HiraKakuProN-W3") +
# 軸の拡張設定（上方向に15%の余裕を持たせる）
scale_y_continuous(labels = scales::comma, expand = expansion(mult = c(0, 0.15))) +
scale_fill_manual(values = c("TRUE" = "steelblue", "FALSE" = "indianred")) +
labs(
title = "Candidate Overperformance Ratio (Fukuoka 11th District 2026)",
subtitle = "個人得票力指数（COR）",
x = "候補者名",
y = "COR (候補者得票 / 政党比例得票)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 自治体別CORデータ（資料29Pに基づく）
df_tsuji_cor <- data.frame(
City = c("苅田町", "行橋市", "田川市", "上毛町", "吉富町", "みやこ町",
"豊前市", "福智町", "糸田町", "添田町", "築上町", "香春町", "川崎町", "赤村", "大任町"),
COR = c(0.88, 0.82, 0.79, 0.76, 0.75, 0.75, 0.75, 0.70, 0.66, 0.63, 0.62, 0.59, 0.57, 0.48, 0.38)
)
# 可視化
ggplot(df_tsuji_cor, aes(x = reorder(City, COR), y = COR)) +
geom_bar(stat = "identity", fill = "indianred", alpha = 0.8) +
geom_hline(yintercept = 1.0, linetype = "dashed", color = "gray50") +
geom_hline(yintercept = 0.74, linetype = "dotted", color = "blue") + # 全体平均
geom_text(aes(label = sprintf("%.2f", COR)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 1.1)) +
labs(
title = "Municipal COR for Candidate Tsuji (Fukuoka 11th District 2026)",
subtitle = "辻候補の自治体別COR：青い点線は全体平均(0.74)",
x = "自治体名",
y = "COR (個人票 / 政党票)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 資料29Pの正確な集計値に基づくデータ
df_murakami_cor <- data.frame(
City = c("吉富町", "赤村", "上毛町", "苅田町", "行橋市", "築上町", "田川市",
"川崎町", "香春町", "福智町", "みやこ町", "豊前市", "添田町", "糸田町", "大任町"),
COR = c(3.35, 3.07, 2.93, 2.84, 2.81, 2.80, 2.78, 2.73, 2.71, 2.70, 2.69, 2.69, 2.63, 2.62, 2.34)
)
# 可視化：タイトルに \n を入れて改行
ggplot(df_murakami_cor, aes(x = reorder(City, COR), y = COR)) +
geom_bar(stat = "identity", fill = "steelblue", alpha = 0.8) +
geom_hline(yintercept = 1.0, linetype = "dashed", color = "red") +
geom_hline(yintercept = 2.78, linetype = "dotted", color = "blue") +
geom_text(aes(label = sprintf("%.2f", COR)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 4.0)) +
labs(
# \n を使ってタイトルを2行に分割
title = "Municipal COR for Candidate Murakami\n(Fukuoka 11th District 2026)",
subtitle = "村上候補の自治体別COR：青い点線は全体平均(2.78)",
x = "自治体名",
y = "COR (個人票 / 政党票)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2), # 行間の調整
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
2候補の比較
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
library(tidyr)
library(dplyr)
# 資料29Pに基づく正確なCORデータ
df_compare <- data.frame(
City = c("吉富町", "赤村", "上毛町", "苅田町", "行橋市", "築上町", "田川市",
"川崎町", "香春町", "福智町", "みやこ町", "豊前市", "添田町", "糸田町", "大任町"),
Murakami = c(3.35, 3.07, 2.93, 2.84, 2.81, 2.80, 2.78, 2.73, 2.71, 2.70, 2.69, 2.69, 2.63, 2.62, 2.34),
Tsuji = c(0.75, 0.48, 0.76, 0.88, 0.82, 0.62, 0.79, 0.57, 0.59, 0.70, 0.75, 0.75, 0.63, 0.66, 0.38)
)
# 比較用にデータをロング形式に変換
df_long <- df_compare %>%
pivot_longer(cols = c(Murakami, Tsuji), names_to = "Candidate", values_to = "COR")
# 可視化
ggplot(df_long, aes(x = reorder(City, COR), y = COR, fill = Candidate)) +
# サイド・バイ・サイドの棒グラフ
geom_bar(stat = "identity", position = position_dodge(width = 0.7), width = 0.6) +
# 政党票ライン (COR=1.0)
geom_hline(yintercept = 1.0, linetype = "dashed", color = "red", alpha = 0.5) +
# 数値ラベル
geom_text(aes(label = sprintf("%.2f", COR)),
position = position_dodge(width = 0.7),
hjust = -0.2, size = 2.5, family = "HiraKakuProN-W3") +
# 軸を反転させて読みやすく
coord_flip() +
scale_y_continuous(limits = c(0, 4.2)) +
scale_fill_manual(values = c("Murakami" = "steelblue", "Tsuji" = "indianred"),
labels = c("Murakami" = "村上候補", "Tsuji" = "辻候補")) +
labs(
title = "Municipal COR Comparison: Murakami vs. Tsuji\n(Fukuoka 11th District 2026)",
subtitle = "自治体別個人得票力指数の比較：赤の破線(1.0)は政党票相当",
x = "自治体名",
y = "COR (個人得票 / 政党比例票)",
fill = "候補者名",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
武田候補(当選者)のCOR
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 武田候補の自治体別CORデータ（資料29Pに基づく正確な算出値）
df_takeda_cor <- data.frame(
City = c("大任町", "糸田町", "赤村", "川崎町", "福智町", "香春町", "添田町",
"田川市", "吉富町", "豊前市", "上毛町", "みやこ町", "苅田町", "行橋市", "築上町"),
COR = c(1.42, 1.39, 1.36, 1.35, 1.34, 1.31, 1.23, 1.18, 1.17, 1.14, 1.14, 1.12, 1.05, 1.04, 0.94)
)
# 可視化
ggplot(df_takeda_cor, aes(x = reorder(City, COR), y = COR)) +
geom_bar(stat = "identity", fill = "steelblue", alpha = 0.8) +
geom_hline(yintercept = 1.0, linetype = "dashed", color = "red") + # 政党票ライン
geom_hline(yintercept = 1.15, linetype = "dotted", color = "blue") + # 全体平均
geom_text(aes(label = sprintf("%.2f", COR)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 1.8)) +
labs(
title = "Municipal COR for Candidate Takeda\n(Fukuoka 11th District 2026)",
subtitle = "武田候補の自治体別COR：青い点線は全体平均(1.15)",
x = "自治体名",
y = "COR (個人票 / 政党票)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
井上候補のCOR
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 井上候補の自治体別CORデータ
df_inoue_cor <- data.frame(
City = c("苅田町", "上毛町", "行橋市", "糸田町", "吉富町", "福智町", "みやこ町",
"香春町", "田川市", "川崎町", "豊前市", "添田町", "築上町", "大任町", "赤村"),
COR = c(1.09, 1.04, 1.02, 0.97, 0.94, 0.91, 0.91, 0.89, 0.89, 0.87, 0.86, 0.83, 0.74, 0.67, 0.65)
)
# 可視化
ggplot(df_inoue_cor, aes(x = reorder(City, COR), y = COR, fill = COR > 1)) +
geom_bar(stat = "identity", width = 0.7, alpha = 0.8) +
geom_hline(yintercept = 1.0, linetype = "dashed", color = "red") +
geom_hline(yintercept = 0.94, linetype = "dotted", color = "blue") + # 全体平均
geom_text(aes(label = sprintf("%.2f", COR)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 1.4)) +
scale_fill_manual(values = c("TRUE" = "indianred", "FALSE" = "indianred")) +
labs(
title = "Municipal COR for Candidate Inoue\n(Fukuoka 11th District 2026)",
subtitle = "井上候補の自治体別COR：青い点線は全体平均(0.94)",
x = "自治体名",
y = "COR (個人得票 / 政党比例票)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "none",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
志岐候補のCOR
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 志岐候補の自治体別CORデータ（CSV集計値に基づく正確な算出値）
df_shiki_cor <- data.frame(
City = c("苅田町", "吉富町", "川崎町", "田川市", "行橋市", "糸田町", "福智町",
"添田町", "豊前市", "みやこ町", "上毛町", "香春町", "赤村", "築上町", "大任町"),
COR = c(1.53, 1.49, 1.31, 1.29, 1.24, 1.14, 1.10, 1.10, 1.07, 1.03, 1.01, 0.96, 0.89, 0.89, 0.84)
)
# 可視化：主要カラーに鮭色（社民党イメージに近い暖色）を採用
ggplot(df_shiki_cor, aes(x = reorder(City, COR), y = COR, fill = COR > 1)) +
geom_bar(stat = "identity", width = 0.7, alpha = 0.8) +
geom_hline(yintercept = 1.0, linetype = "dashed", color = "red") +
geom_hline(yintercept = 1.18, linetype = "dotted", color = "blue") + # 全体平均
geom_text(aes(label = sprintf("%.2f", COR)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 1.8)) +
scale_fill_manual(values = c("TRUE" = "steelblue", "FALSE" = "steelblue")) +
labs(
title = "Municipal COR for Candidate Shiki\n(Fukuoka 11th District 2026)",
subtitle = "志岐候補の自治体別COR：青い点線は全体平均(1.18)",
x = "自治体名",
y = "COR (個人得票 / 政党比例票)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "none",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
5. 政党得票率(福岡11区全体)
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 福岡11区全体での政党得票率データ
df_party_total <- data.frame(
Party = c("自民党", "中道", "維新", "参政", "民主党", "みらい",
"共産党", "れいわ", "社民党", "保守党", "ゆうこく連合"),
Rate = c(39.10, 18.49, 11.12, 7.68, 6.12, 4.09, 3.41, 3.33, 3.16, 2.15, 1.34)
)
# 可視化
ggplot(df_party_total, aes(x = reorder(Party, Rate), y = Rate, fill = Party)) +
geom_bar(stat = "identity", alpha = 0.8) +
geom_text(aes(label = sprintf("%.2f%%", Rate)), hjust = -0.1, size = 3.5, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 45), breaks = seq(0, 45, 10)) +
# 視認性向上のためのカラーパレット（任意で調整可能）
scale_fill_manual(values = c(
"自民党" = "dodgerblue4", "中道" = "indianred", "維新" = "turquoise4",
"参政" = "#E69F00", "民主党" = "darkcyan", "みらい" = "gray50",
"共産党" = "firebrick", "れいわ" = "deeppink3", "社民党" = "salmon",
"保守党" = "darkblue", "ゆうこく連合" = "darkgreen"
)) +
labs(
title = "PR Vote Share by Party (Fukuoka 11th District Total 2026)",
subtitle = "各政党の比例代表得票率（福岡11区全体：有効投票数 139,208票）",
x = "政党名",
y = "得票率 (%)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(size = 14, face = "bold"),
legend.position = "none", # 棒グラフ自体に色がついているため凡例は省略
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
自民党の得票率
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 自民党の自治体別比例得票率データ（正確な算出値）
df_ldp_pr <- data.frame(
City = c("大任町", "赤村", "上毛町", "築上町", "みやこ町", "添田町", "豊前市",
"香春町", "行橋市", "福智町", "川崎町", "吉富町", "苅田町", "田川市", "糸田町"),
Rate = c(50.53, 44.07, 42.24, 40.93, 40.88, 40.49, 40.20, 40.09, 38.99, 38.69, 38.51, 38.09, 37.70, 36.32, 36.05)
)
# 可視化
ggplot(df_ldp_pr, aes(x = reorder(City, Rate), y = Rate)) +
geom_bar(stat = "identity", fill = "dodgerblue4", alpha = 0.8) + # 自民党のイメージカラー
geom_hline(yintercept = 39.1, linetype = "dashed", color = "red") + # 全体平均
geom_text(aes(label = sprintf("%.1f%%", Rate)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 60), breaks = seq(0, 60, 10)) +
labs(
title = "LDP PR Vote Share by Municipality (Fukuoka 11th District 2026)",
subtitle = "自民党の自治体別比例得票率：赤の破線は全体平均(39.1%)",
x = "自治体名",
y = "比例得票率 (%)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
中道の得票率
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 中道の自治体別比例得票率データ
df_chudo_pr <- data.frame(
City = c("川崎町", "糸田町", "添田町", "香春町", "福智町", "吉富町", "赤村",
"田川市", "みやこ町", "大任町", "苅田町", "豊前市", "行橋市", "上毛町", "築上町"),
Rate = c(25.74, 25.11, 21.25, 20.92, 20.33, 19.28, 18.96, 18.82, 18.63, 18.16, 17.61, 17.13, 16.89, 15.58, 15.07)
)
# 可視化
ggplot(df_chudo_pr, aes(x = reorder(City, Rate), y = Rate)) +
geom_bar(stat = "identity", fill = "dodgerblue4", alpha = 0.8) +
geom_hline(yintercept = 18.5, linetype = "dashed", color = "red") + # 全体平均
geom_text(aes(label = sprintf("%.1f%%", Rate)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 30), breaks = seq(0, 30, 5)) +
labs(
title = "Chudo PR Vote Share by Municipality\n (Fukuoka 11th District 2026)",
subtitle = "中道改革連合の自治体別比例得票率：赤の破線は全体平均(18.5%)",
x = "自治体名",
y = "比例得票率 (%)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
参政党の得票率
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 参政党の自治体別比例得票率データ
df_sanseito <- data.frame(
City = c("苅田町", "吉富町", "行橋市", "豊前市", "上毛町", "田川市", "福智町",
"築上町", "香春町", "みやこ町", "糸田町", "川崎町", "大任町", "添田町", "赤村"),
Rate = c(9.22, 8.77, 8.17, 7.93, 7.74, 7.59, 7.48, 7.09, 6.80, 6.78, 6.77, 6.26, 6.22, 6.12, 6.00)
)
# 可視化：参政党のイメージカラーに近いオレンジ系（#E69F00）を使用
ggplot(df_sanseito, aes(x = reorder(City, Rate), y = Rate)) +
geom_bar(stat = "identity", fill = "dodgerblue4", alpha = 0.8) +
geom_hline(yintercept = 7.68, linetype = "dashed", color = "red") + # 全体平均
geom_text(aes(label = sprintf("%.2f%%", Rate)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 12), breaks = seq(0, 12, 2)) +
labs(
title = "Sanseito PR Vote Share by Municipality \n(Fukuoka 11th District 2026)",
subtitle = "参政党の自治体別比例得票率：赤の破線は全体平均(7.68%)",
x = "自治体名",
y = "比例得票率 (%)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
維新の得票率
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 維新の自治体別比例得票率データ（正確な算出値）
df_ishin_pr <- data.frame(
City = c("築上町", "行橋市", "みやこ町", "豊前市", "苅田町", "添田町", "田川市",
"上毛町", "香春町", "吉富町", "川崎町", "福智町", "糸田町", "赤村", "大任町"),
Rate = c(16.01, 12.26, 11.83, 11.72, 11.05, 10.92, 10.90, 9.83, 9.45, 8.71, 8.64, 8.59, 8.50, 8.39, 6.93)
)
# 可視化：維新のイメージカラーに近いターコイズ系を採用
ggplot(df_ishin_pr, aes(x = reorder(City, Rate), y = Rate)) +
geom_bar(stat = "identity", fill = "dodgerblue4", alpha = 0.8) +
geom_hline(yintercept = 11.1, linetype = "dashed", color = "red") + # 全体平均
geom_text(aes(label = sprintf("%.1f%%", Rate)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 20), breaks = seq(0, 20, 5)) +
labs(
title = "Ishin PR Vote Share by Municipality (Fukuoka 11th District 2026)",
subtitle = "日本維新の会の自治体別比例得票率：赤の破線は全体平均(11.1%)",
x = "自治体名",
y = "比例得票率 (%)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
国民民主党の得票率
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 国民民主党の自治体別比例得票率データ
df_dpp_pr <- data.frame(
City = c("苅田町", "行橋市", "吉富町", "上毛町", "みやこ町", "田川市", "豊前市",
"築上町", "赤村", "福智町", "糸田町", "大任町", "香春町", "添田町", "川崎町"),
Rate = c(8.16, 7.34, 6.41, 6.00, 5.80, 5.66, 5.43, 5.42, 5.25, 5.20, 5.07, 4.58, 4.58, 4.30, 4.02)
)
# 可視化
ggplot(df_dpp_pr, aes(x = reorder(City, Rate), y = Rate)) +
geom_bar(stat = "identity", fill = "dodgerblue4", alpha = 0.8) +
geom_hline(yintercept = 6.12, linetype = "dashed", color = "red") + # 全体平均
geom_text(aes(label = sprintf("%.2f%%", Rate)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 10), breaks = seq(0, 10, 2)) +
labs(
title = "DPP PR Vote Share by Municipality (Fukuoka 11th District 2026)",
subtitle = "国民民主党の自治体別比例得票率：赤の破線は全体平均(6.12%)",
x = "自治体名",
y = "比例得票率 (%)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
社民党の得票率
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 社民党の自治体別比例得票率データ
df_sdp_pr <- data.frame(
City = c("赤村", "福智町", "糸田町", "田川市", "香春町", "豊前市", "川崎町",
"上毛町", "添田町", "築上町", "みやこ町", "大任町", "行橋市", "吉富町", "苅田町"),
Rate = c(4.50, 4.47, 4.30, 4.27, 4.11, 3.45, 3.36, 3.18, 3.16, 2.89, 2.88, 2.66, 2.45, 2.39, 2.01)
)
# 可視化
ggplot(df_sdp_pr, aes(x = reorder(City, Rate), y = Rate)) +
geom_bar(stat = "identity", fill = "dodgerblue4", alpha = 0.8) +
geom_hline(yintercept = 3.16, linetype = "dashed", color = "red") + # 全体平均
geom_text(aes(label = sprintf("%.2f%%", Rate)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 6), breaks = seq(0, 6, 1)) +
labs(
title = "SDP PR Vote Share by Municipality (Fukuoka 11th District 2026)",
subtitle = "社会民主党の自治体別比例得票率：赤の破線は全体平均(3.16%)",
x = "自治体名",
y = "比例得票率 (%)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)
共産党の得票率
#--- AONF定数の定義 ---
AONF_CAPTION <- "Archives from 2026 Onward and Notes for the Future\n2026年以降の記録 | https://ab.cocolog-nifty.com/note/"
library(ggplot2)
# 共産党の自治体別比例得票率データ（正確な算出値）
df_jcp_pr <- data.frame(
City = c("田川市", "吉富町", "添田町", "福智町", "川崎町", "香春町", "上毛町",
"糸田町", "みやこ町", "行橋市", "豊前市", "築上町", "苅田町", "赤村", "大任町"),
Rate = c(5.23, 5.18, 4.30, 4.07, 4.02, 3.91, 3.39, 3.19, 3.10, 2.81, 2.72, 2.60, 2.52, 2.39, 1.72)
)
# 可視化
ggplot(df_jcp_pr, aes(x = reorder(City, Rate), y = Rate)) +
geom_bar(stat = "identity", fill = "dodgerblue4", alpha = 0.8) +
geom_hline(yintercept = 3.41, linetype = "dashed", color = "red") + # 全体平均
geom_text(aes(label = sprintf("%.2f%%", Rate)), hjust = -0.2, size = 3, family = "HiraKakuProN-W3") +
coord_flip() +
scale_y_continuous(limits = c(0, 7), breaks = seq(0, 7, 1)) +
labs(
title = "JCP PR Vote Share by Municipality (Fukuoka 11th District 2026)",
subtitle = "日本共産党の自治体別比例得票率：赤の破線は全体平均(3.41%)",
x = "自治体名",
y = "比例得票率 (%)",
caption = AONF_CAPTION
) +
theme_minimal(base_family = "HiraKakuProN-W3") +
theme(
plot.title = element_text(lineheight = 1.2),
legend.position = "bottom",
plot.caption = element_text(size = 8, color = "gray30", hjust = 1)
)