Anti-LGBTQ+ Legislation

legisData <- read.csv("~/Downloads/AntiLGBTQData.csv", fileEncoding = "UTF-8")
# Remove the period from the 'ISSUE.' column
colnames(legisData)[colnames(legisData) == "ISSUE."] <- "ISSUE"

# Check for missing values in important columns
anyNA(legisData$State)

[1] FALSE

anyNA(legisData$Bill)

[1] FALSE

anyNA(legisData$ISSUE)

[1] FALSE

anyNA(legisData$Status)

[1] FALSE

anyNA(legisData$Status.Detail)

[1] FALSE

Status Info

set.seed(789)

# Remove duplicate bills based on the Bill column (only keep unique bills)
uniqueLegisData <- legisData %>% distinct(Bill, .keep_all = TRUE)

# Create a table of unique bill statuses
statusTable <- table(uniqueLegisData$Status)

# Generate a pie chart with labels showing both count and percentage for unique bills
pie(statusTable, 
    labels = paste(names(statusTable), "\n", statusTable, "(", round(statusTable/sum(statusTable)*100, 1), "%)", sep = ""),
    main = "Status of Unique Bills",
    col = rainbow(length(statusTable)))

statusTable

      Advancing        Defeated      Introduced Passed into Law 
            327              64               5              23 

Summary Stats: Bills/State Frequency

legisData

stateCounts <- as.numeric(stateTable)

summary(stateCounts)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  1.000   3.000   6.000   9.886  12.000  51.000 

which.max(stateCounts)  

[1] 39

names(stateTable)[which.max(stateCounts)]  

[1] "Texas"

which.min(stateCounts)  

[1] 4

names(stateTable)[which.min(stateCounts)]  

[1] "California"

boxplot(stateCounts, main = "Boxplot of Bills per State", ylab = "Number of Bills")

hist(stateCounts, main = "Histogram of Bills per State", xlab = "Number of Bills", col = "lightblue", breaks = 10)

House v Senate

legisData$Chamber <- ifelse(grepl("\\bAB\\b|\\bHF\\b|\\bHB\\b|\\HSB\\b", legisData$Bill, ignore.case = TRUE), "House", 
                             ifelse(grepl("\\bSF\\b|\\bSB\\b|\\S\\b|\\SSB\\b", legisData$Bill, ignore.case = TRUE), "Senate", "Unknown"))


# Create the table of Senate vs. House bills (no "Unknown" values)
chamberTable <- table(legisData$Chamber)

# Display the table
chamberTable

 House Senate 
   214    221 

# Pie chart for Senate vs. House bills
pie(chamberTable, col = c("lightpink", "lightgreen"),
    main = "Bills by Chamber (Senate vs. House)")

# Assign 'House' or 'Senate' based on bill title (HB, AB, HF for House; SB, SF for Senate)

Issues Stats

unique(legisData$`ISSUE `)


names(legisData)[grepl("ISSUE", names(legisData))]

names(legisData)[grepl("ISSUE", names(legisData))] <- "Issue"
library(dplyr)
library(tidyr)

legisData_clean <- legisData %>%
  separate_rows(Issue, sep = ",|/|;|\\n") %>%  # separate by common delimiters
  mutate(Issue = trimws(Issue))             

head(legisData_clean$Issue)
issueTable <- table(legisData_clean$Issue)
issueTable

issueCounts <- table(legisData_clean$Issue)

names(legisData)


par(mar = c(10, 4, 4, 2))
barplot(issueTable,
        las = 3,  
        col = "lightblue",  
        main = "Frequency of Bill Issues",  
        ylab = "Number of Bills",  
        xlab = "Issue",  
        cex.names = 0.5)  
 
legisData

Dividing States by Party

redStates <- c("Alaska", "Arizona", "Arkansas", "Florida", "Georgia", "Idaho", "Indiana", "Iowa", "Kansas", "Kentucky", "Louisiana", "Mississippi", "Missouri", "Montana", "New Hampshire", "North Carolina", "North Dakota", "Ohio", "Oklahoma", "South Carolina", "South Dakota", "Tennessee", "Texas", "Utah", "West Virginia", "Wyoming")
blueStates <- c("California", "Colorado", "Connecticut", "Hawaii", "Maine", "Maryland", "Massachusetts", "Nevada", "New Jersey", "New Mexico", "Oregon", "Rhode Island", "Virginia", "Washington")
purpleStates <- c("Michigan", "Minnesota", "Nebraska", "Pennsylvania")

legisData$StateColor <- ifelse(legisData$State %in% redStates, "Red",
                         ifelse(legisData$State %in% blueStates, "Blue",
                         ifelse(legisData$State %in% purpleStates, "Purple", "Unknown")))

colorTable <- table(legisData$StateColor)
colorTable 

  Blue Purple    Red 
    47     16    372 

pie(colorTable, col = c("blue", "purple", "red"),
    main = "Bills by State Color")

NA
NA

political affiliation testing

uniqueLegisData <- legisData %>% distinct(Bill, .keep_all = TRUE)
# Check if 'StateColor' exists in the dataset
colnames(uniqueLegisData)

[1] "State"         "Bill"          "ISSUE"         "Status"        "Status.Detail" "Chamber"       "StateColor"   

billsPerState <- table(uniqueLegisData$State)
billsData <- data.frame(State = names(billsPerState), Bills = as.vector(billsPerState))
billsData$StateColor <- uniqueLegisData$StateColor[match(billsData$State, uniqueLegisData$State)]
anovaResult <- aov(Bills ~ StateColor, data = billsData)
summary(anovaResult)

            Df Sum Sq Mean Sq F value Pr(>F)   
StateColor   2   1148   574.0   6.182 0.0045 **
Residuals   41   3807    92.9                  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

tukeyResult <- TukeyHSD(anovaResult)
tukeyResult$StateColor

                  diff        lwr      upr      p adj
Purple-Blue  0.7857143 -12.498669 14.07010 0.98866437
Red-Blue    10.5549451   2.787495 18.32240 0.00551238
Red-Purple   9.7692308  -2.815477 22.35394 0.15511047

plot(tukeyResult)

status v color

library(dplyr)
uniqueLegisData <- legisData %>% distinct(Bill, .keep_all = TRUE)
statusColorTable <- table(uniqueLegisData$StateColor, uniqueLegisData$Status)
statusColorTable

        
         Advancing Defeated Introduced Passed into Law
  Blue          29       14          2               0
  Purple        15        0          1               0
  Red          283       50          2              23

chisq.test(statusColorTable)

Warning in chisq.test(statusColorTable) :
  Chi-squared approximation may be incorrect

    Pearson's Chi-squared test

data:  statusColorTable
X-squared = 24.394, df = 6, p-value = 0.0004419

fisher.test(statusColorTable)

    Fisher's Exact Test for Count Data

data:  statusColorTable
p-value = 0.0007295
alternative hypothesis: two.sided