acbass49/0_1_older_singles.r

## 0_1_older_singles.r
# Title: Exploring Singles
# Author; Alex Bass
# Date: 2025-03-08

# load packages
library(tidyverse)
library(haven)
library(sjlabelled)

# load data
data_07 <- read_sav('./data/rls/rls2007.sav')
data_14 <- read_sav('./data/rls/rls2014.sav')
data_24 <- read_sav('./data/rls/rls2024.sav')

# Step 1: Get variables to align as best as possible

vars_to_keep <- c(
    'disaffiliated',
    'active',
    'inactive',
    'inactive_or_disaffiliated',
    'still_id',
    'age_rc',
    'marital_rc',
    'year',
    'female'
)

# 2007
data_07$age_rc <- factor(car::recode(
    as.numeric(data_07$age),
    "1:35='18-35';36:45='36-45';46:65='46-65';66:97='66+';99=NA", as.numeric = FALSE),
    levels = c("18-35", "36-45", "46-65", "66+")
    )
data_07$marital_rc <- as.numeric(car::recode(data_07$marital, "4:5=2;6=4;9=NA"))
data_07$female <- as.numeric(car::recode(data_07$sex, "1=0;2=1;9=NA"))
data_07$attend <- as.numeric(car::recode(data_07$q20, "1:2=1;else=0"))
data_07$mormon <- ifelse(data_07$reltrad == 20000, 1, 0)

data_07_lds <- data_07 |>
    filter(q50 == 3) |>
    mutate(
        disaffiliated = ifelse(q50 == 3 & reltrad != 20000, 1, 0),
        active = ifelse(q50 == 3 & reltrad == 20000 & q20 %in% c(1,2), 1, 0),
        inactive = ifelse(q50 == 3 & reltrad == 20000 & !q20 %in% c(1,2), 1, 0),
        inactive_or_disaffiliated = ifelse(disaffiliated == 1 | inactive == 1, 1, 0),
        still_id = ifelse(q50 == 3 & reltrad == 20000, 1, 0),
        year = 2007
    ) |>
    select(all_of(vars_to_keep))

# 2014
data_14$age_rc <- factor(car::recode(as.numeric(data_14$agerec),
    "1:3='18-35';4:5='36-45';6:9='46-65';10:15='66+';99=NA", as.numeric = FALSE),
    levels = c("18-35", "36-45", "46-65", "66+"))
data_14$marital_rc <- as.numeric(car::recode(data_14$marital, "4:5=2;6=4;9=NA"))
data_14$female <- as.numeric(car::recode(data_14$SEX, "1=0;2=1"))
data_14$attend <- as.numeric(car::recode(data_14$attend, "1:2=1;else=0"))
data_14$mormon <- ifelse(data_14$RELTRAD == 20000, 1, 0)

data_14_lds <- data_14 |>
    filter(qj1 == 3) |>
    mutate(
        disaffiliated = ifelse(qj1 == 3 & RELTRAD != 20000, 1, 0),
        active = ifelse(qj1 == 3 & RELTRAD == 20000 & attend %in% c(1,2), 1, 0),
        inactive = ifelse(qj1 == 3 & RELTRAD == 20000 & !attend %in% c(1,2), 1, 0),
        inactive_or_disaffiliated = ifelse(disaffiliated == 1 | inactive == 1, 1, 0),
        still_id = ifelse(qj1 == 3 & RELTRAD == 20000, 1, 0),
        year = 2014
    ) |>
    select(all_of(vars_to_keep))

# 2024
data_24$age_rc <- factor(car::recode(
    as.numeric(data_24$BIRTHDECADE),
    "6:7='18-35';5='36-45';3:4='46-65';1:2='66+';99=NA", as.numeric = FALSE),
    levels = c(
        "18-35", "36-45", "46-65", "66+")
    )
data_24$marital_rc <- as.numeric(car::recode(data_24$MARITAL, "4:5=2;6=4;99=NA"))
data_24$female <- as.numeric(car::recode(data_24$GENDER, "1=0;2=1;3=NA;99=NA"))
data_24$attend <- as.numeric(car::recode(data_24$ATTNDPERRLS, "1:2=1;else=0"))
data_24$mormon <- ifelse(data_24$RELTRAD == 20000, 1, 0)

data_24_lds <- data_24 |>
    filter(FRMREL == 20000) |>
    mutate(
        disaffiliated = ifelse(FRMREL == 20000 & RELTRAD != 20000, 1, 0),
        active = ifelse(FRMREL == 20000 & RELTRAD == 20000 & ATTNDPERRLS %in% c(1,2), 1, 0),
        inactive = ifelse(FRMREL == 20000 & RELTRAD == 20000 & !ATTNDPERRLS %in% c(1,2), 1, 0),
        inactive_or_disaffiliated = ifelse(disaffiliated == 1 | inactive == 1, 1, 0),
        still_id = ifelse(FRMREL == 20000 & RELTRAD == 20000, 1, 0),
        year = 2024
    ) |>
    select(all_of(vars_to_keep))

# combine data
data <- bind_rows(data_07_lds, data_14_lds, data_24_lds) |>
    drop_na() |>
    mutate(year = as.factor(year))

# singles by year
data |>
    filter(marital_rc == 4) |>
    group_by(year) |>
    count(still_id)
# This is really not very large Nsizes 😡

data |>
    filter(marital_rc == 4) |>
    count(still_id)

sum(data$marital_rc == 4) # total of 248 singles

# Visualize age distribution by inactive_or_disaffiliated
data |>
    group_by(still_id) |>
    count(age_rc) |>
    mutate(prop = n / sum(n)) |>
    ggplot(aes(x = age_rc, y=prop, group=still_id, fill=still_id)) +
    geom_col(position = 'dodge') +
    labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

# Visualize single age distribution by inactive_or_disaffiliated
data |>
    filter(marital_rc %in% c(4)) |>
    group_by(still_id, marital_rc) |>
    count(age_rc) |>
    mutate(prop = n / sum(n)) |>
    ggplot(aes(x = age_rc, y=prop, fill=still_id, group=still_id)) +
    geom_col(position = 'dodge') +
    labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

# Visualize age distribution by inactive_or_disaffiliated - stacked
data |>
    filter(marital_rc %in% c(4) & still_id == 1) |>
    group_by(female, marital_rc) |>
    count(age_rc) |>
    mutate(prop = n / sum(n)) |>
    ggplot(aes(x = age_rc, y=prop, fill=female, group=female)) +
    geom_col(position = 'dodge') +
    labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

data |>
    filter(marital_rc %in% c(4) & still_id == 0) |>
    group_by(female, marital_rc) |>
    count(age_rc) |>
    mutate(prop = n / sum(n)) |>
    ggplot(aes(x = age_rc, y=prop, fill=female, group=female)) +
    geom_col(position = 'dodge') +
    labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

# Checking to see the age distribution of never married singles
vars_to_keep <- c(
    'age_rc',
    'marital_rc',
    'female',
    'attend',
    'mormon'
)
all_data <- data_07 |>
    select(all_of(vars_to_keep)) |>
    bind_rows(data_14 |> select(all_of(vars_to_keep))) |>
    bind_rows(data_24 |> select(all_of(vars_to_keep))) |>
    drop_na()

#overall population age distribution
all_data |>
    count(age_rc) |>
    mutate(prop = n / sum(n)) |>
    ggplot(aes(x = age_rc, y=prop)) +
    geom_col(position = 'dodge') +
    labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

#never married singles age distribution
all_data |>
    filter(marital_rc == 4) |>
    count(age_rc) |>
    mutate(prop = n / sum(n)) |>
    ggplot(aes(x = age_rc, y=prop)) +
    geom_col(position = 'dodge') +
    labs(title = "Age Distribution of Never Married Singles", x = "Age", y = "Count")

#never married singles age distribution by gender
all_data |>
    filter(marital_rc %in% c(4)) |>
    group_by(female, marital_rc) |>
    count(age_rc) |>
    mutate(prop = n / sum(n)) |>
    ggplot(aes(x = age_rc, y=prop, fill=female, group=female)) +
    geom_col(position = 'dodge') +
    labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

plot_3 <- all_data |>
    filter(marital_rc == 4) |>
    filter(mormon == 1) |>
    mutate(age_rc = ifelse(age_rc == "66+"|age_rc == "46-65"|age_rc=="36-45", "2", age_rc)) |>
    mutate(age_rc = ifelse(age_rc == "1", '18-35','36+')) |>
    mutate(sex = ifelse(female == 1, "Female", "Male")) |>
    mutate(attend_church = ifelse(attend == 1, "Weekly+", "Less Than Weekly")) |>
    group_by(age_rc,sex) |>
    count(attend_church) |>
    mutate(prop = n / sum(n)) |>
    ggplot() +
    scale_y_continuous(labels = scales::percent_format(accuracy = 1)) +  # Format y-axis as percentages
    labs(x = "Age Group",
        y = "Proportion",
        caption = "data: Pew Religious Landscape Study 2007,14,24\n@mormon_metrics")+
    geom_col(aes(x = age_rc, y = prop, fill = attend_church), position = 'stack') +
    facet_wrap(~sex) +
    theme_minimal(base_family = "Cairo", base_size = 15) +
    theme(legend.position = "top",plot.background = element_rect(fill = "grey95")) +
    ggtitle("Mormon Single Church Attendance by Age X Gender (n=189)")

ggsave("./images/0_1_older_singles_4.png", plot = plot_3, width = 8, height = 8, units = "in", dpi = 300)


all_data |>
    filter(marital_rc == 4) |>
    filter(mormon == 1) |>
    mutate(age_rc = ifelse(age_rc == "66+"|age_rc == "46-65"|age_rc=="36-45", "2", age_rc)) |>
    mutate(age_rc = ifelse(age_rc == "1", '18-35','36+')) |>
    mutate(sex = ifelse(female == 1, "Female", "Male")) |>
    mutate(attend_church = ifelse(attend == 1, "Weekly+", "Less Than Weekly")) |>
    group_by(age_rc,sex) |>
    count(attend_church) |>
    mutate(prop = n / sum(n)) |>
    ungroup() |>
    summarise(sum(n))


#gist: https://gist.github.com/acbass49/d7975d28a960904330b8d209a982190e

# Adding divorce analysis
data |>
    filter(marital_rc == 3) |>
    count(still_id)
# there are 96 disaffiliated divorced individuals
# 78 divorced active individuals
# 174 total divorced individuals

plot_2 <- data |>
    filter(marital_rc == 3) |>
    mutate(
        sex = ifelse(female == 1, "female", "male"),
        type = ifelse(still_id == 1, "mormon", "disaffiliated")
    ) |>
    group_by(type) |>
    count(sex) |>
    mutate(prop = n / sum(n)) |>
    ggplot(aes(x = type, y = prop, fill = sex)) +
    geom_col(position = 'stack') +
    labs(
        title = "Sex Breakdown of Divorced By Active + Disaffiliated",
        x = "Sex",
        y = "Proportion",
        caption = "data: Pew Religious Landscape Study 2007,14,24\n@mormon_metrics"
    ) +
    scale_fill_manual(values = c("pink", "lightblue")) +
    scale_y_continuous(labels = scales::percent_format(accuracy = 1)) +  # Format y-axis as percentages
    theme_minimal(base_family = "Cairo", base_size = 15) +
    theme(legend.position = "top",plot.background = element_rect(fill = "grey95"))

ggsave("./images/0_1_older_singles_2.png", plot = plot_2, width = 8, height = 8, units = "in", dpi = 300)

data |>
    filter(marital_rc == 3 & still_id == 1) |>
    group_by(female) |>
    count(age_rc) |>
    mutate(prop = n / sum(n))

data |>
    filter(marital_rc == 3 & still_id == 0) |>
    group_by(female) |>
    count(age_rc) |>
    mutate(prop = n / sum(n))

plot_1 <- data |>
    filter(marital_rc == 3) |>
    mutate(type = ifelse(still_id == 1, "Mormon", "Disaffiliated")) |>
    group_by(age_rc) |>
    count(type) |>
    mutate(prop = n / sum(n)) |>
    ggplot(aes(x = age_rc, y = prop, fill = type)) +
    geom_col(position = 'stack') +
    labs(
        title = "Age Distribution of Divorced Individuals",
        x = "Age",
        y = "Proportion",
        caption = "data: Pew Religious Landscape Study 2007,14,24\n@mormon_metrics"
    ) +
    scale_fill_manual(values = c("orange", "#0000ffbe")) +
    scale_y_continuous(labels = scales::percent_format(accuracy = 1)) +  # Format y-axis as percentages
    theme_minimal(base_family = "Cairo", base_size = 15) +
    theme(legend.position = "top",plot.background = element_rect(fill = "grey95"))

ggsave("./images/0_1_older_singles_3.png", plot = plot_1, width = 8, height = 8, units = "in", dpi = 300)

# Adding CES analysis
library(patchwork)

# load data
data <- read_rds("./data/CES/until2023.rds")

data$mormon <- ifelse(data$religion == 3, "Mormon", "Non-Mormon")

data$attend <- ifelse(data$relig_church %in% 1:2, "Weekly+", "Less Than Weekly")

#men vs. women by age
data |>
    mutate(age = car::recode(age, "18:25='18-25';26:35='26-35';36:45='36-45';46:55='46-55';56:65='56-65';66:120='66+'")) |>
    mutate(age = factor(age, levels = c("18-25", "26-35", "36-45", "46-55", "56-65", "66+"))) |>
    filter(marstat == 5) |>
    filter(mormon == "Mormon") |>
    group_by(age) |>
    count(gender) |>
    drop_na() |>
    mutate(prop = n / sum(n))

#active men vs. women by age
data |>
    mutate(age = car::recode(age, "18:25='18-25';26:35='26-35';36:45='36-45';46:55='46-55';56:65='56-65';66:120='66+'")) |>
    mutate(age = factor(age, levels = c("18-25", "26-35", "36-45", "46-55", "56-65", "66+"))) |>
    filter(marstat == 5) |>
    filter(mormon == "Mormon") |>
    filter(attend == "Weekly+") |>
    group_by(age) |>
    count(gender) |>
    drop_na() |>
    mutate(prop = n / sum(n))

data |>
    filter(mormon == "Mormon") |>
    filter(marstat == 5) |>
    drop_na(age, attend, gender) |>
    mutate(age = as.numeric(age), attend = factor(attend),gender = sjlabelled::as_character(gender)) |>
    ggplot(aes(x = age, fill = attend)) +
    geom_density(alpha=0.6) +
    scale_fill_manual(values = c("orange", "blue")) +
    xlim(18,80) +
    ggtitle("Mormon Single Population By Church Attendance") +
    theme_minimal(base_family = "Cairo") +
    theme(legend.position = "top") +
    facet_grid(~gender)

(plot_4 <- data |>
    filter(marstat == 5) |>
    filter(mormon == "Mormon") |>
    drop_na(age, attend, gender) |>
    mutate(age = as.numeric(age), attend = factor(attend),gender = sjlabelled::as_character(gender)) |>
    mutate(age_rc = factor(car::recode(age, "1:35='18-35';36:120='36+'"),levels = c("18-35", "36+"))) |>
    group_by(age_rc,gender) |>
    count(attend) |>
    mutate(prop = n / sum(n)) |>
    ggplot() +
    scale_y_continuous(labels = scales::percent_format(accuracy = 1)) +  # Format y-axis as percentages
    labs(x = "Age Group",
        y = "Proportion",
        caption = "data: Cooperative Election Study 2007-23\n@mormon_metrics")+
    geom_col(aes(x = age_rc, y = prop, fill = attend), position = 'stack') +
    facet_wrap(~gender) +
    theme_minimal(base_family = "Cairo", base_size = 15) +
    theme(legend.position = "top",plot.background = element_rect(fill = "grey95")) +
    ggtitle("Mormon Single Church Attendance by Age X Gender (n=1446)"))

ggsave("./images/0_1_older_singles_5.png", plot = plot_4, width = 8, height = 8, units = "in", dpi = 300)

p1 <- data |>
    mutate(
        age = as.numeric(data$age)
    ) |>
    drop_na(age, mormon) |>
    filter(marstat == 5) |>
    ggplot(aes(x = age,fill=mormon)) +
    geom_density(alpha=0.6, position = 'identity') +
    scale_fill_manual(values = c("orange", "blue")) +
    xlim(18,80) +
    ggtitle("Age Distribution of Singles") +
    theme_minimal(base_family = "Cairo") +
    theme(legend.position = "top")

p2 <- data |>
    mutate(
        gender = sjlabelled::as_character(data$gender),
        age = as.numeric(data$age)
    ) |>
    drop_na(gender, age) |>
    filter(marstat == 5) |>
    ggplot(aes(x = age,fill=gender)) +
    geom_density(alpha=0.6, position = 'identity') +
    scale_fill_manual(values = c("pink", "lightblue")) +
    xlim(18,80) +
    theme_minimal(base_family = "Cairo") +
    theme(legend.position = "top") +
    ggtitle("US Single Population By Gender")

p3 <- data |>
    mutate(
        gender = sjlabelled::as_character(data$gender),
        age = as.numeric(data$age)
    ) |>
    drop_na(gender, age) |>
    filter(marstat == 5 & religion == 3) |>
    ggplot(aes(x = age,fill=gender)) +
    geom_density(alpha=0.6, position = 'identity') +
    scale_fill_manual(values = c("pink", "lightblue")) +
    xlim(18,80) +
    theme_minimal(base_family = "Cairo") +
    ggtitle("Mormon Single Population By Gender") +
    theme(legend.position = "top")

combined_plot <- p1 + p2 + p3 +
    plot_annotation(
        caption = "Data: Cooperative Election Study 2007-23\n@mormon_metrics",
        theme = theme(
            text = element_text(family = "Cairo"),
            plot.background = element_rect(fill = "grey95")
        )
    )
ggsave("./images/0_1_older_singles_1.png", plot = combined_plot, width = 12, height = 8, units = "in", dpi = 300)
	# Title: Exploring Singles
	# Author; Alex Bass
	# Date: 2025-03-08

	# load packages
	library(tidyverse)
	library(haven)
	library(sjlabelled)

	# load data
	data_07 <- read_sav('./data/rls/rls2007.sav')
	data_14 <- read_sav('./data/rls/rls2014.sav')
	data_24 <- read_sav('./data/rls/rls2024.sav')

	# Step 1: Get variables to align as best as possible

	vars_to_keep <- c(
	'disaffiliated',
	'active',
	'inactive',
	'inactive_or_disaffiliated',
	'still_id',
	'age_rc',
	'marital_rc',
	'year',
	'female'
	)

	# 2007
	data_07$age_rc <- factor(car::recode(
	as.numeric(data_07$age),
	"1:35='18-35';36:45='36-45';46:65='46-65';66:97='66+';99=NA", as.numeric = FALSE),
	levels = c("18-35", "36-45", "46-65", "66+")
	)
	data_07$marital_rc <- as.numeric(car::recode(data_07$marital, "4:5=2;6=4;9=NA"))
	data_07$female <- as.numeric(car::recode(data_07$sex, "1=0;2=1;9=NA"))
	data_07$attend <- as.numeric(car::recode(data_07$q20, "1:2=1;else=0"))
	data_07$mormon <- ifelse(data_07$reltrad == 20000, 1, 0)

	data_07_lds <- data_07 \|>
	filter(q50 == 3) \|>
	mutate(
	disaffiliated = ifelse(q50 == 3 & reltrad != 20000, 1, 0),
	active = ifelse(q50 == 3 & reltrad == 20000 & q20 %in% c(1,2), 1, 0),
	inactive = ifelse(q50 == 3 & reltrad == 20000 & !q20 %in% c(1,2), 1, 0),
	inactive_or_disaffiliated = ifelse(disaffiliated == 1 \| inactive == 1, 1, 0),
	still_id = ifelse(q50 == 3 & reltrad == 20000, 1, 0),
	year = 2007
	) \|>
	select(all_of(vars_to_keep))

	# 2014
	data_14$age_rc <- factor(car::recode(as.numeric(data_14$agerec),
	"1:3='18-35';4:5='36-45';6:9='46-65';10:15='66+';99=NA", as.numeric = FALSE),
	levels = c("18-35", "36-45", "46-65", "66+"))
	data_14$marital_rc <- as.numeric(car::recode(data_14$marital, "4:5=2;6=4;9=NA"))
	data_14$female <- as.numeric(car::recode(data_14$SEX, "1=0;2=1"))
	data_14$attend <- as.numeric(car::recode(data_14$attend, "1:2=1;else=0"))
	data_14$mormon <- ifelse(data_14$RELTRAD == 20000, 1, 0)

	data_14_lds <- data_14 \|>
	filter(qj1 == 3) \|>
	mutate(
	disaffiliated = ifelse(qj1 == 3 & RELTRAD != 20000, 1, 0),
	active = ifelse(qj1 == 3 & RELTRAD == 20000 & attend %in% c(1,2), 1, 0),
	inactive = ifelse(qj1 == 3 & RELTRAD == 20000 & !attend %in% c(1,2), 1, 0),
	inactive_or_disaffiliated = ifelse(disaffiliated == 1 \| inactive == 1, 1, 0),
	still_id = ifelse(qj1 == 3 & RELTRAD == 20000, 1, 0),
	year = 2014
	) \|>
	select(all_of(vars_to_keep))

	# 2024
	data_24$age_rc <- factor(car::recode(
	as.numeric(data_24$BIRTHDECADE),
	"6:7='18-35';5='36-45';3:4='46-65';1:2='66+';99=NA", as.numeric = FALSE),
	levels = c(
	"18-35", "36-45", "46-65", "66+")
	)
	data_24$marital_rc <- as.numeric(car::recode(data_24$MARITAL, "4:5=2;6=4;99=NA"))
	data_24$female <- as.numeric(car::recode(data_24$GENDER, "1=0;2=1;3=NA;99=NA"))
	data_24$attend <- as.numeric(car::recode(data_24$ATTNDPERRLS, "1:2=1;else=0"))
	data_24$mormon <- ifelse(data_24$RELTRAD == 20000, 1, 0)

	data_24_lds <- data_24 \|>
	filter(FRMREL == 20000) \|>
	mutate(
	disaffiliated = ifelse(FRMREL == 20000 & RELTRAD != 20000, 1, 0),
	active = ifelse(FRMREL == 20000 & RELTRAD == 20000 & ATTNDPERRLS %in% c(1,2), 1, 0),
	inactive = ifelse(FRMREL == 20000 & RELTRAD == 20000 & !ATTNDPERRLS %in% c(1,2), 1, 0),
	inactive_or_disaffiliated = ifelse(disaffiliated == 1 \| inactive == 1, 1, 0),
	still_id = ifelse(FRMREL == 20000 & RELTRAD == 20000, 1, 0),
	year = 2024
	) \|>
	select(all_of(vars_to_keep))

	# combine data
	data <- bind_rows(data_07_lds, data_14_lds, data_24_lds) \|>
	drop_na() \|>
	mutate(year = as.factor(year))

	# singles by year
	data \|>
	filter(marital_rc == 4) \|>
	group_by(year) \|>
	count(still_id)
	# This is really not very large Nsizes 😡

	data \|>
	filter(marital_rc == 4) \|>
	count(still_id)

	sum(data$marital_rc == 4) # total of 248 singles

	# Visualize age distribution by inactive_or_disaffiliated
	data \|>
	group_by(still_id) \|>
	count(age_rc) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot(aes(x = age_rc, y=prop, group=still_id, fill=still_id)) +
	geom_col(position = 'dodge') +
	labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

	# Visualize single age distribution by inactive_or_disaffiliated
	data \|>
	filter(marital_rc %in% c(4)) \|>
	group_by(still_id, marital_rc) \|>
	count(age_rc) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot(aes(x = age_rc, y=prop, fill=still_id, group=still_id)) +
	geom_col(position = 'dodge') +
	labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

	# Visualize age distribution by inactive_or_disaffiliated - stacked
	data \|>
	filter(marital_rc %in% c(4) & still_id == 1) \|>
	group_by(female, marital_rc) \|>
	count(age_rc) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot(aes(x = age_rc, y=prop, fill=female, group=female)) +
	geom_col(position = 'dodge') +
	labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

	data \|>
	filter(marital_rc %in% c(4) & still_id == 0) \|>
	group_by(female, marital_rc) \|>
	count(age_rc) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot(aes(x = age_rc, y=prop, fill=female, group=female)) +
	geom_col(position = 'dodge') +
	labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

	# Checking to see the age distribution of never married singles
	vars_to_keep <- c(
	'age_rc',
	'marital_rc',
	'female',
	'attend',
	'mormon'
	)
	all_data <- data_07 \|>
	select(all_of(vars_to_keep)) \|>
	bind_rows(data_14 \|> select(all_of(vars_to_keep))) \|>
	bind_rows(data_24 \|> select(all_of(vars_to_keep))) \|>
	drop_na()

	#overall population age distribution
	all_data \|>
	count(age_rc) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot(aes(x = age_rc, y=prop)) +
	geom_col(position = 'dodge') +
	labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

	#never married singles age distribution
	all_data \|>
	filter(marital_rc == 4) \|>
	count(age_rc) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot(aes(x = age_rc, y=prop)) +
	geom_col(position = 'dodge') +
	labs(title = "Age Distribution of Never Married Singles", x = "Age", y = "Count")

	#never married singles age distribution by gender
	all_data \|>
	filter(marital_rc %in% c(4)) \|>
	group_by(female, marital_rc) \|>
	count(age_rc) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot(aes(x = age_rc, y=prop, fill=female, group=female)) +
	geom_col(position = 'dodge') +
	labs(title = "Age Distribution of Everyone", x = "Age", y = "Count")

	plot_3 <- all_data \|>
	filter(marital_rc == 4) \|>
	filter(mormon == 1) \|>
	mutate(age_rc = ifelse(age_rc == "66+"\|age_rc == "46-65"\|age_rc=="36-45", "2", age_rc)) \|>
	mutate(age_rc = ifelse(age_rc == "1", '18-35','36+')) \|>
	mutate(sex = ifelse(female == 1, "Female", "Male")) \|>
	mutate(attend_church = ifelse(attend == 1, "Weekly+", "Less Than Weekly")) \|>
	group_by(age_rc,sex) \|>
	count(attend_church) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot() +
	scale_y_continuous(labels = scales::percent_format(accuracy = 1)) + # Format y-axis as percentages
	labs(x = "Age Group",
	y = "Proportion",
	caption = "data: Pew Religious Landscape Study 2007,14,24\n@mormon_metrics")+
	geom_col(aes(x = age_rc, y = prop, fill = attend_church), position = 'stack') +
	facet_wrap(~sex) +
	theme_minimal(base_family = "Cairo", base_size = 15) +
	theme(legend.position = "top",plot.background = element_rect(fill = "grey95")) +
	ggtitle("Mormon Single Church Attendance by Age X Gender (n=189)")

	ggsave("./images/0_1_older_singles_4.png", plot = plot_3, width = 8, height = 8, units = "in", dpi = 300)



	all_data \|>
	filter(marital_rc == 4) \|>
	filter(mormon == 1) \|>
	mutate(age_rc = ifelse(age_rc == "66+"\|age_rc == "46-65"\|age_rc=="36-45", "2", age_rc)) \|>
	mutate(age_rc = ifelse(age_rc == "1", '18-35','36+')) \|>
	mutate(sex = ifelse(female == 1, "Female", "Male")) \|>
	mutate(attend_church = ifelse(attend == 1, "Weekly+", "Less Than Weekly")) \|>
	group_by(age_rc,sex) \|>
	count(attend_church) \|>
	mutate(prop = n / sum(n)) \|>
	ungroup() \|>
	summarise(sum(n))


	#gist: https://gist.github.com/acbass49/d7975d28a960904330b8d209a982190e

	# Adding divorce analysis
	data \|>
	filter(marital_rc == 3) \|>
	count(still_id)
	# there are 96 disaffiliated divorced individuals
	# 78 divorced active individuals
	# 174 total divorced individuals

	plot_2 <- data \|>
	filter(marital_rc == 3) \|>
	mutate(
	sex = ifelse(female == 1, "female", "male"),
	type = ifelse(still_id == 1, "mormon", "disaffiliated")
	) \|>
	group_by(type) \|>
	count(sex) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot(aes(x = type, y = prop, fill = sex)) +
	geom_col(position = 'stack') +
	labs(
	title = "Sex Breakdown of Divorced By Active + Disaffiliated",
	x = "Sex",
	y = "Proportion",
	caption = "data: Pew Religious Landscape Study 2007,14,24\n@mormon_metrics"
	) +
	scale_fill_manual(values = c("pink", "lightblue")) +
	scale_y_continuous(labels = scales::percent_format(accuracy = 1)) + # Format y-axis as percentages
	theme_minimal(base_family = "Cairo", base_size = 15) +
	theme(legend.position = "top",plot.background = element_rect(fill = "grey95"))

	ggsave("./images/0_1_older_singles_2.png", plot = plot_2, width = 8, height = 8, units = "in", dpi = 300)

	data \|>
	filter(marital_rc == 3 & still_id == 1) \|>
	group_by(female) \|>
	count(age_rc) \|>
	mutate(prop = n / sum(n))

	data \|>
	filter(marital_rc == 3 & still_id == 0) \|>
	group_by(female) \|>
	count(age_rc) \|>
	mutate(prop = n / sum(n))

	plot_1 <- data \|>
	filter(marital_rc == 3) \|>
	mutate(type = ifelse(still_id == 1, "Mormon", "Disaffiliated")) \|>
	group_by(age_rc) \|>
	count(type) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot(aes(x = age_rc, y = prop, fill = type)) +
	geom_col(position = 'stack') +
	labs(
	title = "Age Distribution of Divorced Individuals",
	x = "Age",
	y = "Proportion",
	caption = "data: Pew Religious Landscape Study 2007,14,24\n@mormon_metrics"
	) +
	scale_fill_manual(values = c("orange", "#0000ffbe")) +
	scale_y_continuous(labels = scales::percent_format(accuracy = 1)) + # Format y-axis as percentages
	theme_minimal(base_family = "Cairo", base_size = 15) +
	theme(legend.position = "top",plot.background = element_rect(fill = "grey95"))

	ggsave("./images/0_1_older_singles_3.png", plot = plot_1, width = 8, height = 8, units = "in", dpi = 300)

	# Adding CES analysis
	library(patchwork)

	# load data
	data <- read_rds("./data/CES/until2023.rds")

	data$mormon <- ifelse(data$religion == 3, "Mormon", "Non-Mormon")

	data$attend <- ifelse(data$relig_church %in% 1:2, "Weekly+", "Less Than Weekly")

	#men vs. women by age
	data \|>
	mutate(age = car::recode(age, "18:25='18-25';26:35='26-35';36:45='36-45';46:55='46-55';56:65='56-65';66:120='66+'")) \|>
	mutate(age = factor(age, levels = c("18-25", "26-35", "36-45", "46-55", "56-65", "66+"))) \|>
	filter(marstat == 5) \|>
	filter(mormon == "Mormon") \|>
	group_by(age) \|>
	count(gender) \|>
	drop_na() \|>
	mutate(prop = n / sum(n))

	#active men vs. women by age
	data \|>
	mutate(age = car::recode(age, "18:25='18-25';26:35='26-35';36:45='36-45';46:55='46-55';56:65='56-65';66:120='66+'")) \|>
	mutate(age = factor(age, levels = c("18-25", "26-35", "36-45", "46-55", "56-65", "66+"))) \|>
	filter(marstat == 5) \|>
	filter(mormon == "Mormon") \|>
	filter(attend == "Weekly+") \|>
	group_by(age) \|>
	count(gender) \|>
	drop_na() \|>
	mutate(prop = n / sum(n))

	data \|>
	filter(mormon == "Mormon") \|>
	filter(marstat == 5) \|>
	drop_na(age, attend, gender) \|>
	mutate(age = as.numeric(age), attend = factor(attend),gender = sjlabelled::as_character(gender)) \|>
	ggplot(aes(x = age, fill = attend)) +
	geom_density(alpha=0.6) +
	scale_fill_manual(values = c("orange", "blue")) +
	xlim(18,80) +
	ggtitle("Mormon Single Population By Church Attendance") +
	theme_minimal(base_family = "Cairo") +
	theme(legend.position = "top") +
	facet_grid(~gender)

	(plot_4 <- data \|>
	filter(marstat == 5) \|>
	filter(mormon == "Mormon") \|>
	drop_na(age, attend, gender) \|>
	mutate(age = as.numeric(age), attend = factor(attend),gender = sjlabelled::as_character(gender)) \|>
	mutate(age_rc = factor(car::recode(age, "1:35='18-35';36:120='36+'"),levels = c("18-35", "36+"))) \|>
	group_by(age_rc,gender) \|>
	count(attend) \|>
	mutate(prop = n / sum(n)) \|>
	ggplot() +
	scale_y_continuous(labels = scales::percent_format(accuracy = 1)) + # Format y-axis as percentages
	labs(x = "Age Group",
	y = "Proportion",
	caption = "data: Cooperative Election Study 2007-23\n@mormon_metrics")+
	geom_col(aes(x = age_rc, y = prop, fill = attend), position = 'stack') +
	facet_wrap(~gender) +
	theme_minimal(base_family = "Cairo", base_size = 15) +
	theme(legend.position = "top",plot.background = element_rect(fill = "grey95")) +
	ggtitle("Mormon Single Church Attendance by Age X Gender (n=1446)"))

	ggsave("./images/0_1_older_singles_5.png", plot = plot_4, width = 8, height = 8, units = "in", dpi = 300)

	p1 <- data \|>
	mutate(
	age = as.numeric(data$age)
	) \|>
	drop_na(age, mormon) \|>
	filter(marstat == 5) \|>
	ggplot(aes(x = age,fill=mormon)) +
	geom_density(alpha=0.6, position = 'identity') +
	scale_fill_manual(values = c("orange", "blue")) +
	xlim(18,80) +
	ggtitle("Age Distribution of Singles") +
	theme_minimal(base_family = "Cairo") +
	theme(legend.position = "top")

	p2 <- data \|>
	mutate(
	gender = sjlabelled::as_character(data$gender),
	age = as.numeric(data$age)
	) \|>
	drop_na(gender, age) \|>
	filter(marstat == 5) \|>
	ggplot(aes(x = age,fill=gender)) +
	geom_density(alpha=0.6, position = 'identity') +
	scale_fill_manual(values = c("pink", "lightblue")) +
	xlim(18,80) +
	theme_minimal(base_family = "Cairo") +
	theme(legend.position = "top") +
	ggtitle("US Single Population By Gender")

	p3 <- data \|>
	mutate(
	gender = sjlabelled::as_character(data$gender),
	age = as.numeric(data$age)
	) \|>
	drop_na(gender, age) \|>
	filter(marstat == 5 & religion == 3) \|>
	ggplot(aes(x = age,fill=gender)) +
	geom_density(alpha=0.6, position = 'identity') +
	scale_fill_manual(values = c("pink", "lightblue")) +
	xlim(18,80) +
	theme_minimal(base_family = "Cairo") +
	ggtitle("Mormon Single Population By Gender") +
	theme(legend.position = "top")

	combined_plot <- p1 + p2 + p3 +
	plot_annotation(
	caption = "Data: Cooperative Election Study 2007-23\n@mormon_metrics",
	theme = theme(
	text = element_text(family = "Cairo"),
	plot.background = element_rect(fill = "grey95")
	)
	)
	ggsave("./images/0_1_older_singles_1.png", plot = combined_plot, width = 12, height = 8, units = "in", dpi = 300)
No results found