Sociodemographic Factors

rm(list = ls())
setwd("~/Downloads")
library(foreign)
sex_data <- read.spss("sex_data.sav", use.value.label=TRUE, to.data.frame=TRUE)

Libraries/themes

library(ggplot2)
library(tidyverse)
library(gridExtra)
library(finalfit)
library(kableExtra)
library(xtable)
jrothsch_theme <-  theme_bw() + 
  theme(text = element_text(size = 10, face = "bold", color = "deepskyblue4"),panel.grid = element_blank(),axis.text = element_text(size = 10, color = "gray13"), axis.title = element_text(size = 10, color = "red"), legend.text = element_text(colour="Black", size=10), legend.title = element_text(colour="Black", size=7), plot.subtitle = element_text(size=14, face="italic", color="black"))

---

Removing obserations without our key variables

sex_data <- sex_data[!is.na(sex_data$status),]
sex_data <- sex_data[!is.na(sex_data$age),]
sex_data <- sex_data[!is.na(sex_data$Years_PrimaryPartner),]
sex_data <- sex_data[!is.na(sex_data$MALE),]

---

Making variables

sex_data <- sex_data %>%
  mutate(married = status == "Married")

sex_data <- sex_data %>%
  mutate(married_gender = ifelse(married, 
                                 ifelse(MALE == "Male", "Married Male", "Married Female"),
                                 ifelse(MALE == "Male", "Unmarried Male", "Unmarried Female ")))

  married_num =   length(sex_data$married[sex_data$married == T])
  unmarried_num =   length(sex_data$married[sex_data$married == T])
  
  sex_data <- sex_data %>% mutate(freq_num =  as.numeric(ifelse(sexfreq == "At least once per day", 6, 
                                ifelse(sexfreq == "3-4 times per week", 5,
                                  ifelse(sexfreq == 'At least once a week', 4,
                                       ifelse(sexfreq == 'At least once per month', 3, 
                                              ifelse(sexfreq =='At least once per year', 2,
                                                     ifelse( sexfreq =="Less than once a year", 1, 0))))))))
  
  
  female <- sex_data %>%
  filter(sex_data$MALE == "Female")

---

Binary graphs

bin_graphs_want <- function(df, mod, modname, modlabs,  interaction_labs){
  
  
  df2 = df[!is.na(df$want) & !is.na(mod),]
  mod = mod[!is.na(df$want) & !is.na(mod)]
  
    df2$mod_married = interaction(df2$married, mod)
     
    age <- ggplot(data = df2, aes(x = age, y = want, color = mod))  +
      geom_smooth() +
      labs(title = paste("Effect Of", modname, "On", "Want", " -- Age"), x = "Age", y = "Want", color = "") + 
      scale_x_continuous(limits = c(18, 80)) +
      scale_color_discrete(labels = modlabs) + 
      jrothsch_theme
  
  sum_inter <- df2 %>%
      group_by( mod_married) %>%
      summarize(DV = mean(want))
    
    mod_married <- ggplot(data = sum_inter,  aes(x = mod_married, y = DV)) +
    geom_bar(stat = 'identity', position = 'identity', fill = "Black") + 
      scale_x_discrete(labels = interaction_labs) + 
      labs(title = paste0("Interaction Between ", modname, "And Marriage"), y = "Want", x = "") +
      jrothsch_theme
    
    
    grid.arrange(age, mod_married)
    
}

bin_graphs_like <- function(df, mod, modname, modlabs, interaction_labs){
  

  df <- sex_data
  mod <- sex_data$children_in_house
  modname <-"Children In House"
  modlabs <- c("Children", "No Children")
  interaction_labs <- c("Unmarried, Children", "Married, Children", "Unmarried, No Children", "Married, No Children")
  df2 = df[!is.na(df$totlike) & !is.na(mod),]
  mod = mod[!is.na(df$totlike) & !is.na(mod)]
  table(mod)
  
    df2$mod_married = interaction(df2$married, mod)
     
    age <- ggplot(data = df2, aes(x = age, y = totlike, color = mod))  +
       geom_smooth() + 
      labs(title = paste("Effect Of", modname, "On", "Like", " -- Age"), x = "Age", y = "Like", color = "") + 
      scale_x_continuous(limits = c(18, 80)) +
      scale_color_discrete(labels = modlabs) + 
      jrothsch_theme
  
  sum_inter <- df2 %>%
      group_by( mod_married) %>%
      summarize(DV = mean(totlike))
    
    mod_married <- ggplot(data = sum_inter,  aes(x = mod_married, y = DV)) +
    geom_bar(stat = 'identity', position = 'identity', fill = "black") +
         scale_x_discrete(labels = interaction_labs) + 
      labs(title = paste0("Interaction Between ", modname, "And Marriage"), y = "Like", x = "") +
      jrothsch_theme
    
    
    grid.arrange(age, mod_married)
    
}
bin_graphs_freq <- function(df, mod, modname, modlabs, interaction_labs){
  

  
  df2 = df[!is.na(df$freq_num) & !is.na(mod),]
  mod = mod[!is.na(df$freq_num) & !is.na(mod)]
  
    df2$mod_married = interaction(df2$married, mod)
     
    age <- ggplot(data = df2, aes(x = age, y = freq_num, color = mod))  +
      geom_smooth() +
      labs(title = paste("Effect Of", modname, "On", "Frequency", " -- Age"), x = "Age", y = "Frequency", color = "") + 
      scale_x_continuous(limits = c(18, 80)) +
      scale_color_discrete(labels = modlabs) + 
      jrothsch_theme
  
  sum_inter <- df2 %>%
      group_by(mod_married) %>%
      summarize(DV = mean(freq_num))
    
    mod_married <- ggplot(data = sum_inter,  aes(x = mod_married, y = DV)) +
    geom_bar(stat = 'identity', position = 'identity', fill = "black") +    
      scale_x_discrete(labels = interaction_labs) + 
      labs(title = paste0("Interaction Between ", modname, "And Marriage"), y = "Frequency", x = "") +
      jrothsch_theme
    
    
    grid.arrange(age, mod_married)
    
}

---

Small continuous graphs

contsmall <- function(df, mod, modname, modlabs){

  
  df2 = df[!is.na(df$want) & !is.na(df$totlike) & !is.na(df$freq_num ) & !is.na(mod) ,]
  mod = mod[!is.na(df$want) & !is.na(df$totlike) & !is.na(df$freq_num ) & !is.na(mod)]




want <- ggplot(df2, aes(x=mod, y = want)) +
  stat_summary_bin(fun.y='mean', bins=20,
                    size=2, geom='point', mapping = aes(group = married, color = married)) +
  geom_smooth(method='lm', se = F,  size= 1, aes(color = married)) + jrothsch_theme +
    labs( x = modname, y = "Want")
  
like <- ggplot(df2, aes(x=mod, y = totlike)) +
  stat_summary_bin(fun.y='mean', bins=20,
                    size=2, geom='point', mapping = aes(group = married, color = married)) +
  geom_smooth(method='lm', se = F,  size= 1, aes(color = married)) + jrothsch_theme +
    labs( x = modname, y = "Like")




freq <- ggplot(df2, aes(x=mod, y=  freq_num)) +
  stat_summary_bin(fun.y='mean', bins=20,
                    size=2, geom='point', mapping = aes(group = married, color = married)) +
  geom_smooth(method='lm', se = F, aes(color = married)) + jrothsch_theme +
      labs( x = modname, y = "Frequency")

  

grid.arrange(want, like, freq)

}

---

Regressions – Standard

###################################################################################################################################
#REgressions
#########################################################################################################################
  reg_no_mod <- function(dv, df){
      dv <- as.numeric(dv)
  tx1 <- df %>% lm(dv ~ married, data = .)

  
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True")
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
    
  }


###########################################################################################################################
  reg_no_mod_i <- function(dv, df){

  dv <- as.numeric(dv)
  tx1 <- df %>% lm(dv ~ married + age + married*age, data = .)

  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: TRUE", "Age",  "Married X Age")
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
    
  }

##################################################################################################################

  reg_no_mod_control <- function(dv, df){

  dv <- as.numeric(dv)
  tx1 <- df %>% lm(dv ~ married + age +  Years_PrimaryPartner + MALE, data = .)

  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: TRUE", "Age",  "Duration", "MALE")
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
    
  }

##################################################################################################################

  reg_just_mod_i <- function(dv, df, mod, modname){
    
      dv <- as.numeric(dv)
  tx1 <- df %>% lm(dv ~ married + mod, data = .)

  
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True", modname)
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
    
  }
  
#############################################################################################################################################################################################################################################################
  reg_full_control <- function(dv, df, mod, modname){
     
     dv <- as.numeric(dv)
       tx1 <- df %>% lm(dv ~ married + mod + age + Years_PrimaryPartner + MALE, data = .)
  
      
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True", modname, "Age (years)", "Duration", "Male")
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
    
  
    
  }


 


#############################################################################################################################################################################################################################################################
  reg_full_control_i <- function(dv, df, mod, modname){
    
          dv <- as.numeric(dv)
  tx1 <- df %>% lm(dv ~ married + mod + age + Years_PrimaryPartner + MALE + married*age, data = .)

  
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True", modname, "Age (years)", "Duration", "Male", "Married X Age")
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
  }

---

Subsets where controls can’t be used – e.g. one gender subsets

     reg_full_control_gend <- function(dv, df, mod, modname){

    
      dv <- as.numeric(dv)
      
       tx1 <- df %>% lm(dv ~ married + mod + age + Years_PrimaryPartner, data = .)
  
      
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True", modname, "Age (years)", "Duration" )
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
    
  
    
     }

 reg_full_control_gend_i <- function(dv, df, mod, modname){
    
          dv <- as.numeric(dv)
  tx1 <- df %>% lm(dv ~ married + mod + age + Years_PrimaryPartner + age*married, data = .)

  
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True", modname, "Age (years)", "Duration", "Married X Age")
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
  }

---

Looking at modertators interaction with marriage

  ireg_just_mod_i <- function(dv, df, mod, modname){
    
      dv <- as.numeric(dv)
  tx1 <- df %>% lm(dv ~ married + mod + married*mod, data = .)

  
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True", modname, paste("Married X", modname))
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
    
  }
  
#############################################################################################################################################################################################################################################################
  ireg_full_control <- function(dv, df, mod, modname){
     
     dv <- as.numeric(dv)
       tx1 <- df %>% lm(dv ~ married + mod + age + Years_PrimaryPartner + MALE + married*mod, data = .)
  
      
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True", modname, "Age (years)", "Duration", "Male",  paste("Married X", modname))
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
    
  
    
  }


 


#############################################################################################################################################################################################################################################################
  ireg_full_control_i <- function(dv, df, mod, modname){
    
          dv <- as.numeric(dv)
  tx1 <- df %>% lm(dv ~ married + mod + age + Years_PrimaryPartner + MALE + age*married + married*mod, data = .)

  
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True", modname, "Age (years)", "Duration", "Male", "Married X Age",  paste("Married X", modname))
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
  }
  
  
  
   
    
    #############################################################################################################################################################################################################################################################
   #For subsets wheere we can't use all controls
    
    #############################################################################################################################################################################################################################################################
    
     ireg_full_control_gend <- function(dv, df, mod, modname){

    
      dv <- as.numeric(dv)
      
       tx1 <- df %>% lm(dv ~ married + mod + age + Years_PrimaryPartner + married*mod, data = .)
  
      
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True", modname, "Age (years)", "Duration",  paste("Married X", modname))
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r",  "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
    
  
    
     }

 ireg_full_control_gend_i <- function(dv, df, mod, modname){
    
          dv <- as.numeric(dv)
  tx1 <- df %>% lm(dv ~ married + mod + age + Years_PrimaryPartner + age*married + married*mod, data = .)

  
  tx1 <- xtable(tx1)
  colnames(tx1)[colnames(tx1)=="Pr(>|t|)"] <- "p_value"
    colnames(tx1)[colnames(tx1)=="t value"] <- "t_value"

  
  tx1 <- tx1 %>%
    mutate(p_value = round(p_value, 4)) %>%
    mutate(t_value = round(t_value, 3)) %>%
    mutate(p_value= cell_spec(p_value, "html", background = ifelse(p_value < .01, "gold", "white"))) %>%
    mutate(t_value= cell_spec(t_value, "html", background = ifelse(t_value < -2.5, "#FF6347", 
                                                                   ifelse(t_value > 2.5, "lightgreen", "white"))))

  
  rownames(tx1) <- c("Intercept", "Married: True", modname, "Age (years)", "Duration", "Married X Age",  paste("Married X", modname))
  
kable(tx1, row.names= T, align=c("l", "l", "r", "r", "r")  ,
            booktabs=TRUE, escape = F) %>% 
    kable_styling(font_size=8) 
  }

Marriage on Want/Age/Like without mediators, with or without controls, for comparison

Want - Marriage

reg_no_mod(sex_data$want, sex_data )

	Estimate	Std. Error	t_value	p_value
Intercept	28.927462	0.2358428	122.656	0
Married: True	-5.062436	0.2804827	-18.049	0

Like - Marriage

reg_no_mod(sex_data$totlike, sex_data )

	Estimate	Std. Error	t_value	p_value
Intercept	38.524460	0.2654955	145.104	0
Married: True	-3.954296	0.3168241	-12.481	0

Frequency - Marriage

reg_no_mod(sex_data$freq_num, sex_data)

	Estimate	Std. Error	t_value	p_value
Intercept	3.6641623	0.0408283	89.746	0
Married: True	-0.6595098	0.0480844	-13.716	0

Want - Marriage, controls

reg_no_mod_control(sex_data$want, sex_data )

	Estimate	Std. Error	t_value	p_value
Intercept	33.6515234	0.3616805	93.042	0
Married: TRUE	-2.4498192	0.2836310	-8.637	0
Age	-0.1788330	0.0079330	-22.543	0
Duration	-0.0164498	0.0049803	-3.303	0.001
MALE	4.5428634	0.2430139	18.694	0

Like - Marriage, controls

reg_no_mod_control(sex_data$totlike, sex_data)

	Estimate	Std. Error	t_value	p_value
Intercept	40.9088281	0.4368889	93.637	0
Married: TRUE	-2.6911441	0.3412552	-7.886	0
Age	-0.0852356	0.0095901	-8.888	0
Duration	-0.0098875	0.0058960	-1.677	0.0936
MALE	1.9027594	0.2944406	6.462	0

Frequency - Marriage, controls

reg_no_mod_control(sex_data$freq_num, sex_data)

	Estimate	Std. Error	t_value	p_value
Intercept	4.8698702	0.0640338	76.052	0
Married: TRUE	-0.1485242	0.0488622	-3.04	0.0024
Age	-0.0334322	0.0014244	-23.471	0
Duration	-0.0032137	0.0010801	-2.975	0.0029
MALE	0.2339988	0.0411009	5.693	0

---

Economic Security Correlates With Higher W/L/F, But Doesn’t Affect Marriage Estimates

Graphs

sex_data$SecureNum = as.numeric(sex_data$econ_secure)

contsmall(sex_data, sex_data$SecureNum, "Economic Security", "")

Want - Only Economic Security

reg_just_mod_i(sex_data$want, sex_data, sex_data$SecureNum, "Economic Security")

	Estimate	Std. Error	t_value	p_value
Intercept	28.1024819	0.3539723	79.392	0
Married: True	-4.9898408	0.2910801	-17.143	0
Economic Security	0.2669037	0.0702588	3.799	1e-04

Want – Full Controls

reg_full_control(sex_data$want, sex_data, sex_data$SecureNum, "Economic Security")

	Estimate	Std. Error	t_value	p_value
Intercept	32.2402534	0.4269611	75.511	0
Married: True	-2.5032814	0.2928144	-8.549	0
Economic Security	0.4279101	0.0657507	6.508	0
Age (years)	-0.1780372	0.0082507	-21.579	0
Duration	-0.0164754	0.0049954	-3.298	0.001
Male	4.4305888	0.2528991	17.519	0

Like - Only Economic Security

reg_just_mod_i(sex_data$totlike, sex_data, sex_data$SecureNum,"Economic Security")

	Estimate	Std. Error	t_value	p_value
Intercept	35.579222	0.3894035	91.369	0
Married: True	-4.215278	0.3202160	-13.164	0
Economic Security	0.798112	0.0772914	10.326	0

Like – Full Controls

reg_full_control(sex_data$totlike,sex_data, sex_data$SecureNum, "Economic Security")

	Estimate	Std. Error	t_value	p_value
Intercept	38.0229422	0.4994597	76.128	0
Married: True	-2.7885252	0.3425347	-8.141	0
Economic Security	0.8980070	0.0769152	11.675	0
Age (years)	-0.0950957	0.0096516	-9.853	0
Duration	-0.0129868	0.0058436	-2.222	0.0263
Male	1.8234063	0.2958418	6.163	0

Frequency - Only Economic Security

reg_just_mod_i(sex_data$freq_num, sex_data, sex_data$SecureNum, "Economic Security")

	Estimate	Std. Error	t_value	p_value
Intercept	3.5719567	0.0597229	59.809	0
Married: True	-0.6416354	0.0495148	-12.958	0
Economic Security	0.0338985	0.0118347	2.864	0.0042

Frequency – Full Controls

reg_full_control(sex_data$freq_num, sex_data, sex_data$SecureNum, "Economic Security")

	Estimate	Std. Error	t_value	p_value
Intercept	4.6149388	0.0737187	62.602	0
Married: True	-0.1658769	0.0499455	-3.321	9e-04
Economic Security	0.0751823	0.0110847	6.783	0
Age (years)	-0.0330308	0.0014658	-22.535	0
Duration	-0.0031195	0.0010744	-2.904	0.0037
Male	0.2217620	0.0424541	5.224	0

---

Feeling Personally Hurt By the Economic Downturn Lowers Like And Frequency, But Doesn’t Affect Marriage

Graphs

sex_data$HurtNum = as.numeric(sex_data$econ_hurt)

contsmall(sex_data, sex_data$HurtNum, "Economic Hurt", "")

Want - Only Economic Hurt

reg_just_mod_i(sex_data$want, sex_data, sex_data$HurtNum, "Economic Hurt")

	Estimate	Std. Error	t_value	p_value
Intercept	29.2782602	0.4073395	71.877	0
Married: True	-4.8942777	0.2904324	-16.852	0
Economic Hurt	-0.0444954	0.0736562	-0.604	0.5458

Want – Full Controls

reg_full_control(sex_data$want, sex_data, sex_data$HurtNum, "Economic Hurt")

	Estimate	Std. Error	t_value	p_value
Intercept	33.7736842	0.4829793	69.928	0
Married: True	-2.4556816	0.2941520	-8.348	0
Economic Hurt	-0.0437978	0.0683929	-0.64	0.522
Age (years)	-0.1729108	0.0082511	-20.956	0
Duration	-0.0145745	0.0050102	-2.909	0.0036
Male	4.4804931	0.2539606	17.642	0

Like - Only Economic Hurt

reg_just_mod_i(sex_data$totlike, sex_data, sex_data$HurtNum,"Economic Hurt")

	Estimate	Std. Error	t_value	p_value
Intercept	40.272139	0.4514913	89.198	0
Married: True	-3.955757	0.3219125	-12.288	0
Economic Hurt	-0.398369	0.0816399	-4.88	0

Like – Full Controls

reg_full_control(sex_data$totlike,sex_data, sex_data$HurtNum, "Economic Hurt")

	Estimate	Std. Error	t_value	p_value
Intercept	42.5976932	0.5692948	74.825	0
Married: True	-2.7202765	0.3467212	-7.846	0
Economic Hurt	-0.3978147	0.0806157	-4.935	0
Age (years)	-0.0843628	0.0097257	-8.674	0
Duration	-0.0089079	0.0059056	-1.508	0.1315
Male	1.9298237	0.2993471	6.447	0

Frequency - Only Economic Hurt

reg_just_mod_i(sex_data$freq_num, sex_data, sex_data$HurtNum, "Economic Hurt")

	Estimate	Std. Error	t_value	p_value
Intercept	3.8205804	0.0688016	55.53	0
Married: True	-0.6292731	0.0492935	-12.766	0
Economic Hurt	-0.0285361	0.0124015	-2.301	0.0214

Frequency – Full Controls

reg_full_control(sex_data$freq_num, sex_data, sex_data$HurtNum, "Economic Hurt")

	Estimate	Std. Error	t_value	p_value
Intercept	4.9816202	0.0832645	59.829	0
Married: True	-0.1561676	0.0501474	-3.114	0.0019
Economic Hurt	-0.0312331	0.0115256	-2.71	0.0068
Age (years)	-0.0320833	0.0014656	-21.89	0
Duration	-0.0027764	0.0010779	-2.576	0.01
Male	0.2309884	0.0426271	5.419	0

---

Feeling Anxious About The Economy Lowers Like And Frequency, But Doesn’t Affect Marriage

Graphs

sex_data$AnxNum = as.numeric(sex_data$econ_anxious)

contsmall(sex_data, sex_data$AnxNum, "Economic Anxiety", "")

Want - Only Economic Anxiety

reg_just_mod_i(sex_data$want, sex_data, sex_data$AnxNum, "Economic Anxiety")

	Estimate	Std. Error	t_value	p_value
Intercept	29.0342250	0.4203434	69.073	0
Married: True	-4.8890687	0.2904131	-16.835	0
Economic Anxiety	0.0105147	0.0772909	0.136	0.8918

Want – Full Controls

reg_full_control(sex_data$want, sex_data, sex_data$AnxNum, "Economic Anxiety")

	Estimate	Std. Error	t_value	p_value
Intercept	33.8383957	0.5085759	66.536	0
Married: True	-2.4504488	0.2940631	-8.333	0
Economic Anxiety	-0.0544082	0.0719874	-0.756	0.4498
Age (years)	-0.1733125	0.0082684	-20.961	0
Duration	-0.0145353	0.0050105	-2.901	0.0037
Male	4.4749506	0.2540526	17.614	0

Like - Only Economic Anxiety

reg_just_mod_i(sex_data$totlike, sex_data, sex_data$AnxNum,"Economic Anxiety")

	Estimate	Std. Error	t_value	p_value
Intercept	40.5510888	0.4656180	87.091	0
Married: True	-3.9518542	0.3216931	-12.285	0
Economic Anxiety	-0.4613249	0.0856158	-5.388	0

Like – Full Controls

reg_full_control(sex_data$totlike,sex_data, sex_data$AnxNum, "Economic Anxiety")

	Estimate	Std. Error	t_value	p_value
Intercept	43.2109527	0.5987891	72.164	0
Married: True	-2.6726777	0.3462251	-7.719	0
Economic Anxiety	-0.4995435	0.0847568	-5.894	0
Age (years)	-0.0880516	0.0097351	-9.045	0
Duration	-0.0085469	0.0058993	-1.449	0.1475
Male	1.8789597	0.2991174	6.282	0

Frequency - Only Economic Anxiety

reg_just_mod_i(sex_data$freq_num, sex_data, sex_data$AnxNum, "Economic Anxiety")

	Estimate	Std. Error	t_value	p_value
Intercept	3.7804814	0.0711089	53.165	0
Married: True	-0.6285247	0.0493099	-12.746	0
Economic Anxiety	-0.0194401	0.0130398	-1.491	0.1361

Frequency – Full Controls

reg_full_control(sex_data$freq_num, sex_data, sex_data$AnxNum, "Economic Anxiety")

	Estimate	Std. Error	t_value	p_value
Intercept	5.0477169	0.0876969	57.559	0
Married: True	-0.1523044	0.0501219	-3.039	0.0024
Economic Anxiety	-0.0428054	0.0121509	-3.523	4e-04
Age (years)	-0.0324144	0.0014679	-22.083	0
Duration	-0.0027405	0.0010771	-2.544	0.011
Male	0.2262950	0.0426178	5.31	0

---

Education Level Has Little Effect, Increases Marriage Like Estimate

Graphs

sex_data$EducNum = as.numeric(sex_data$educ)

contsmall(sex_data, sex_data$AnxNum, "Education Level", "")

Want - Only Education

reg_just_mod_i(sex_data$want, sex_data, sex_data$EducNum, "Education Level")

	Estimate	Std. Error	t_value	p_value
Intercept	29.5532773	0.3992271	74.026	0
Married: True	-4.8682097	0.2910248	-16.728	0
Education Level	-0.1404548	0.0917818	-1.53	0.126

Want – Full Controls

reg_full_control(sex_data$want, sex_data, sex_data$EducNum, "Education Level")

	Estimate	Std. Error	t_value	p_value
Intercept	33.5829662	0.4531463	74.111	0
Married: True	-2.4561218	0.2943536	-8.344	0
Education Level	-0.0016097	0.0862935	-0.019	0.9851
Age (years)	-0.1728834	0.0083183	-20.783	0
Duration	-0.0146329	0.0050097	-2.921	0.0035
Male	4.4753205	0.2548165	17.563	0

Like - Only Education

reg_just_mod_i(sex_data$totlike, sex_data, sex_data$EducNum,"Education Level")

	Estimate	Std. Error	t_value	p_value
Intercept	39.342560	0.4440245	88.604	0
Married: True	-3.885504	0.3236807	-12.004	0
Education Level	-0.240647	0.1020806	-2.357	0.0184

Like – Full Controls

reg_full_control(sex_data$totlike,sex_data, sex_data$EducNum, "Education Level")

	Estimate	Std. Error	t_value	p_value
Intercept	41.3409496	0.5360001	77.129	0
Married: True	-2.6848860	0.3481735	-7.711	0
Education Level	-0.1670266	0.1020715	-1.636	0.1018
Age (years)	-0.0827136	0.0098393	-8.406	0
Duration	-0.0091332	0.0059256	-1.541	0.1233
Male	1.9677495	0.3014075	6.529	0

Frequency - Only Education

reg_just_mod_i(sex_data$freq_num, sex_data, sex_data$EducNum, "Education Level")

	Estimate	Std. Error	t_value	p_value
Intercept	3.7804564	0.0680426	55.56	0
Married: True	-0.6235493	0.0494678	-12.605	0
Education Level	-0.0251669	0.0154655	-1.627	0.1037

Frequency – Full Controls

reg_full_control(sex_data$freq_num, sex_data, sex_data$EducNum, "Education Level")

	Estimate	Std. Error	t_value	p_value
Intercept	4.7968498	0.0791281	60.621	0
Married: True	-0.1550942	0.0502657	-3.085	0.002
Education Level	0.0157017	0.0145205	1.081	0.2796
Age (years)	-0.0322418	0.0014761	-21.843	0
Duration	-0.0027959	0.0010793	-2.591	0.0096
Male	0.2267135	0.0428281	5.294	0

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

No Income Effect, Possibly Some Specific Outlier Income Groups

Graphs

sex_data$IncomeNum = as.numeric(sex_data$HouseholdIncome)

contsmall(sex_data, sex_data$IncomeNum, "Income Level", "")

Want - Only Income

reg_just_mod_i(sex_data$want, sex_data, sex_data$IncomeNum, "Income Level")

	Estimate	Std. Error	t_value	p_value
Intercept	28.7239484	0.4557908	63.02	0
Married: True	-5.0549824	0.4141058	-12.207	0
Income Level	-0.0906243	0.1123780	-0.806	0.4201

Want – Full Controls

reg_full_control(sex_data$want, sex_data, sex_data$IncomeNum, "Income Level")

	Estimate	Std. Error	t_value	p_value
Intercept	33.9389741	0.5825240	58.262	0
Married: True	-2.4014129	0.4276577	-5.615	0
Income Level	-0.0754337	0.1051256	-0.718	0.4731
Age (years)	-0.1838015	0.0115380	-15.93	0
Duration	-0.0072029	0.0053718	-1.341	0.1801
Male	4.2568551	0.3774423	11.278	0

Like - Only Income

reg_just_mod_i(sex_data$totlike, sex_data, sex_data$IncomeNum,"Income Level")

	Estimate	Std. Error	t_value	p_value
Intercept	38.1934517	0.5016493	76.136	0
Married: True	-3.8375362	0.4557702	-8.42	0
Income Level	-0.0461524	0.1236847	-0.373	0.7091

Like – Full Controls

reg_full_control(sex_data$totlike,sex_data, sex_data$IncomeNum, "Income Level")

	Estimate	Std. Error	t_value	p_value
Intercept	40.5011138	0.6815216	59.427	0
Married: True	-2.6851216	0.5003363	-5.367	0
Income Level	-0.0410305	0.1229913	-0.334	0.7387
Age (years)	-0.0819290	0.0134988	-6.069	0
Duration	-0.0018835	0.0062847	-0.3	0.7644
Male	1.9336434	0.4415870	4.379	0

Frequency - Only Income

reg_just_mod_i(sex_data$freq_num, sex_data, sex_data$IncomeNum, "Income Level")

	Estimate	Std. Error	t_value	p_value
Intercept	3.5291093	0.0772604	45.678	0
Married: True	-0.7205595	0.0697380	-10.332	0
Income Level	0.0309751	0.0188795	1.641	0.101

Frequency – Full Controls

reg_full_control(sex_data$freq_num, sex_data, sex_data$IncomeNum, "Income Level")

	Estimate	Std. Error	t_value	p_value
Intercept	4.5753121	0.1018641	44.916	0
Married: True	-0.2455662	0.0732116	-3.354	8e-04
Income Level	0.0475445	0.0178977	2.656	0.008
Age (years)	-0.0300907	0.0020378	-14.766	0
Duration	-0.0015402	0.0012051	-1.278	0.2014
Male	0.1844088	0.0640869	2.877	0.0041

---

Partner Age Shows A Similar Effect to Age; The Marriage Effect Comes From A Steeper Decline With Age

Graphs

sex_data <- sex_data %>% mutate(partner_agecat = ifelse(partner_age < 30, 1,
                                                        ifelse(partner_age < 40 & partner_age >= 30, 2,
                                                               ifelse(partner_age < 50 & partner_age >= 40, 3,
                                                                      ifelse(partner_age  < 60 & partner_age >= 50, 4 ,5))))
                                )

contsmall(sex_data, sex_data$partner_agecat, "Partner Age Bucket", "")

Want - Only Education

reg_just_mod_i(sex_data$want, sex_data, sex_data$partner_age, "Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	35.6876997	0.3667337	97.312	0
Married: True	-2.4991131	0.2881915	-8.672	0
Partner Age	-0.1801903	0.0077411	-23.277	0

Want – Full Controls

reg_full_control(sex_data$want, sex_data, sex_data$partner_age, "Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	34.0394728	0.3792510	89.754	0
Married: True	-2.3508670	0.2848575	-8.253	0
Partner Age	-0.0633540	0.0188267	-3.365	8e-04
Age (years)	-0.1232009	0.0183332	-6.72	0
Duration	-0.0147545	0.0050005	-2.951	0.0032
Male	4.2562491	0.2572687	16.544	0

Like - Only Education

reg_just_mod_i(sex_data$totlike, sex_data, sex_data$partner_age,"Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	42.1696109	0.4324816	97.506	0
Married: True	-2.6010701	0.3381217	-7.693	0
Partner Age	-0.0972413	0.0091714	-10.603	0

Like – Full Controls

reg_full_control(sex_data$totlike,sex_data, sex_data$partner_age, "Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	41.4568818	0.4576758	90.581	0
Married: True	-2.5494021	0.3425991	-7.441	0
Partner Age	-0.0900794	0.0227706	-3.956	1e-04
Age (years)	-0.0060467	0.0221898	-0.273	0.7852
Duration	-0.0076745	0.0059133	-1.298	0.1944
Male	1.4971609	0.3113452	4.809	0

Frequency - Only Education

reg_just_mod_i(sex_data$freq_num, sex_data, sex_data$partner_age, "Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	5.0380031	0.0620456	81.198	0
Married: True	-0.1575805	0.0480595	-3.279	0.001
Partner Age	-0.0359067	0.0012849	-27.944	0

Frequency – Full Controls

reg_full_control(sex_data$freq_num, sex_data, sex_data$partner_age, "Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	4.9779925	0.0672407	74.032	0
Married: True	-0.1265578	0.0489196	-2.587	0.0097
Partner Age	-0.0164814	0.0032074	-5.139	0
Age (years)	-0.0191987	0.0031130	-6.167	0
Duration	-0.0024071	0.0010886	-2.211	0.0271
Male	0.1605635	0.0434115	3.699	2e-04

---

Being Older Than Partner Improves W/L/F, Unrelated To Marriage

Graphs

sex_data <-sex_data  %>% mutate(dif = age - partner_age)
sex_data <- sex_data %>% mutate(dif_cat = ifelse(dif < -10, 1,
                                                        ifelse(dif < -5  & dif >= -10, 2,
                                                               ifelse(dif < -1 & dif >= -5, 3,
                                                                      ifelse(dif <= 1 & dif >= -1, 4,
                                                                          ifelse(dif  <= 5 & dif > 1, 5, 
                                                                                 ifelse(dif <= 10 & dif > 5, 6, 7
                                                                                 )))))))
                                 


contsmall(sex_data, sex_data$dif_cat, "Own Age - Partner Age", "")

Want - Age Difference

reg_just_mod_i(sex_data$want, sex_data, sex_data$dif, "Own Age - Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	28.9078666	0.2359541	122.515	0
Married: True	-5.0575833	0.2803994	-18.037	0
Own Age - Partner Age	0.0374872	0.0180943	2.072	0.0383

Want – Full Controls

reg_full_control(sex_data$want, sex_data, sex_data$dif, "Own Age - Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	34.0394728	0.3792510	89.754	0
Married: True	-2.3508670	0.2848575	-8.253	0
Own Age - Partner Age	0.0633540	0.0188267	3.365	8e-04
Age (years)	-0.1865549	0.0082503	-22.612	0
Duration	-0.0147545	0.0050005	-2.951	0.0032
Male	4.2562491	0.2572687	16.544	0

Like - Age Difference

reg_just_mod_i(sex_data$totlike, sex_data, sex_data$dif,"Own Age - Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	38.4936517	0.2655207	144.974	0
Married: True	-3.9461633	0.3166001	-12.464	0
Own Age - Partner Age	0.0580273	0.0205295	2.827	0.0047

Like – Full Controls

reg_full_control(sex_data$totlike,sex_data, sex_data$dif, "Own Age - Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	41.4568818	0.4576758	90.581	0
Married: True	-2.5494021	0.3425991	-7.441	0
Own Age - Partner Age	0.0900794	0.0227706	3.956	1e-04
Age (years)	-0.0961261	0.0099630	-9.648	0
Duration	-0.0076745	0.0059133	-1.298	0.1944
Male	1.4971609	0.3113452	4.809	0

Frequency - Age Difference

reg_just_mod_i(sex_data$freq_num, sex_data, sex_data$dif, "Own Age - Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	3.6675712	0.0408555	89.769	0
Married: True	-0.6602885	0.0480726	-13.735	0
Own Age - Partner Age	-0.0058926	0.0030720	-1.918	0.0552

Frequency – Full Controls

reg_full_control(sex_data$freq_num, sex_data, sex_data$dif, "Own Age - Partner Age")

	Estimate	Std. Error	t_value	p_value
Intercept	4.9779925	0.0672407	74.032	0
Married: True	-0.1265578	0.0489196	-2.587	0.0097
Own Age - Partner Age	0.0164814	0.0032074	5.139	0
Age (years)	-0.0356801	0.0014865	-24.003	0
Duration	-0.0024071	0.0010886	-2.211	0.0271
Male	0.1605635	0.0434115	3.699	2e-04

---

Religion - There is some interesting stuff here so I’m going to wait until after call to analyze. Does’t account for marriage effect nonetheless

Frequency – Full Controls

sex_data %>% 
  lm(want ~ married + religion, data = .) %>%
  summary()

## 
## Call:
## lm(formula = want ~ married + religion, data = .)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -23.4643  -5.9643   0.1145   6.3976  22.4408 
## 
## Coefficients:
##                                                              Estimate
## (Intercept)                                                  28.67394
## marriedTRUE                                                  -5.07878
## religionProtestant (United, Anglican, Presbyterian, Baptist) -0.01831
## religionJewish                                                0.67898
## religionMuslim                                                2.43371
## religionHindu                                                 0.92838
## religionNone or Atheist or Agnostic                          -1.03592
## religionOther                                                 2.31537
##                                                              Std. Error
## (Intercept)                                                     0.45838
## marriedTRUE                                                     0.32525
## religionProtestant (United, Anglican, Presbyterian, Baptist)    0.46261
## religionJewish                                                  0.90977
## religionMuslim                                                  1.75232
## religionHindu                                                   2.48919
## religionNone or Atheist or Agnostic                             0.61280
## religionOther                                                   0.50644
##                                                              t value
## (Intercept)                                                   62.555
## marriedTRUE                                                  -15.615
## religionProtestant (United, Anglican, Presbyterian, Baptist)  -0.040
## religionJewish                                                 0.746
## religionMuslim                                                 1.389
## religionHindu                                                  0.373
## religionNone or Atheist or Agnostic                           -1.690
## religionOther                                                  4.572
##                                                              Pr(>|t|)    
## (Intercept)                                                    <2e-16 ***
## marriedTRUE                                                    <2e-16 ***
## religionProtestant (United, Anglican, Presbyterian, Baptist)    0.968    
## religionJewish                                                  0.456    
## religionMuslim                                                  0.165    
## religionHindu                                                   0.709    
## religionNone or Atheist or Agnostic                             0.091 .  
## religionOther                                                   5e-06 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 8.853 on 3622 degrees of freedom
##   (1445 observations deleted due to missingness)
## Multiple R-squared:  0.07993,    Adjusted R-squared:  0.07815 
## F-statistic: 44.95 on 7 and 3622 DF,  p-value: < 2.2e-16

sex_data %>% 
  lm(totlike ~ married + religion, data = .) %>%
  summary()

## 
## Call:
## lm(formula = totlike ~ married + religion, data = .)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -29.598  -6.693   1.330   8.375  16.944 
## 
## Coefficients:
##                                                              Estimate
## (Intercept)                                                  38.46321
## marriedTRUE                                                  -3.92725
## religionProtestant (United, Anglican, Presbyterian, Baptist)  0.08888
## religionJewish                                               -0.19173
## religionMuslim                                                1.23711
## religionHindu                                                -1.12337
## religionNone or Atheist or Agnostic                          -1.47959
## religionOther                                                 1.13424
##                                                              Std. Error
## (Intercept)                                                     0.51282
## marriedTRUE                                                     0.36388
## religionProtestant (United, Anglican, Presbyterian, Baptist)    0.51756
## religionJewish                                                  1.01783
## religionMuslim                                                  1.96047
## religionHindu                                                   2.78487
## religionNone or Atheist or Agnostic                             0.68559
## religionOther                                                   0.56660
##                                                              t value
## (Intercept)                                                   75.003
## marriedTRUE                                                  -10.793
## religionProtestant (United, Anglican, Presbyterian, Baptist)   0.172
## religionJewish                                                -0.188
## religionMuslim                                                 0.631
## religionHindu                                                 -0.403
## religionNone or Atheist or Agnostic                           -2.158
## religionOther                                                  2.002
##                                                              Pr(>|t|)    
## (Intercept)                                                    <2e-16 ***
## marriedTRUE                                                    <2e-16 ***
## religionProtestant (United, Anglican, Presbyterian, Baptist)   0.8637    
## religionJewish                                                 0.8506    
## religionMuslim                                                 0.5281    
## religionHindu                                                  0.6867    
## religionNone or Atheist or Agnostic                            0.0310 *  
## religionOther                                                  0.0454 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 9.905 on 3622 degrees of freedom
##   (1445 observations deleted due to missingness)
## Multiple R-squared:  0.03599,    Adjusted R-squared:  0.03413 
## F-statistic: 19.32 on 7 and 3622 DF,  p-value: < 2.2e-16

Having Children In House Seems Promising, But Not Nearly Enough Data

Want

sex_data <-sex_data  %>% mutate(children_in_house = ifelse(children == "Yes, my partner and I have 1 or more children under the age of 18 that live with me/my partner all of the time", 1, 0))


bin_graphs_want(sex_data, as.factor(sex_data$children_in_house), "Children In House", c("Children", "No Children"),  
                     c("Unmarried, Children", "Married, Children", "Unmarried, No Children", "Married, No Children"))

Like Graphs

bin_graphs_like(sex_data, as.factor(sex_data$children_in_house), "Children In House", c("Children", "No Children"),  
                     c("Unmarried, Children", "Married, Children", "Unmarried, No Children", "Married, No Children"))

Frequency Graphs

bin_graphs_freq(sex_data, as.factor(sex_data$children_in_house), "Children In House", c("Children", "No Children"),  
                     c("Unmarried, Children", "Married, Children", "Unmarried, No Children", "Married, No Children"))

Want - Children

reg_just_mod_i(sex_data$want, sex_data, sex_data$children_in_house, "Has Children In House")

	Estimate	Std. Error	t_value	p_value
Intercept	27.2613340	1.134890	24.021	0
Married: True	-1.3197076	1.798587	-0.734	0.4663
Has Children In House	0.4213146	2.205528	0.191	0.8492

Want – Full Controls

reg_full_control(sex_data$want, sex_data, sex_data$children_in_house, "Has Children In House")

	Estimate	Std. Error	t_value	p_value
Intercept	28.2289203	3.4994587	8.067	0
Married: True	-2.1847248	2.1088725	-1.036	0.3051
Has Children In House	0.6597606	2.0966675	0.315	0.7543
Age (years)	-0.1090315	0.1243599	-0.877	0.3847
Duration	0.0366755	0.1776905	0.206	0.8373
Male	5.3815719	1.5267409	3.525	9e-04

Like - Children In Houise

reg_just_mod_i(sex_data$totlike, sex_data, sex_data$children_in_house,"Has Children In House")

	Estimate	Std. Error	t_value	p_value
Intercept	36.291123	1.379154	26.314	0
Married: True	-1.652972	2.185700	-0.756	0.4528
Has Children In House	1.557435	2.680227	0.581	0.5636

Like – Full Controls

reg_full_control(sex_data$totlike,sex_data, sex_data$children_in_house, "Has Children In House")

	Estimate	Std. Error	t_value	p_value
Intercept	38.0384981	4.5673762	8.328	0
Married: True	-0.6739723	2.7524297	-0.245	0.8075
Has Children In House	1.9184509	2.7365002	0.701	0.4865
Age (years)	-0.0962328	0.1623103	-0.593	0.5559
Duration	-0.1560996	0.2319157	-0.673	0.5039
Male	3.4260734	1.9926510	1.719	0.0916

Frequency - Children In House

reg_just_mod_i(sex_data$freq_num, sex_data, sex_data$children_in_house, "Has Children In House")

	Estimate	Std. Error	t_value	p_value
Intercept	3.8571923	0.1571327	24.547	0
Married: True	-0.0505453	0.2434290	-0.208	0.8363
Has Children In House	-0.1441925	0.2946513	-0.489	0.6266

Frequency – Full Controls

reg_full_control(sex_data$freq_num, sex_data, sex_data$children_in_house, "Has Children In House")

	Estimate	Std. Error	t_value	p_value
Intercept	4.3420304	0.5181018	8.381	0
Married: True	-0.0072073	0.3138191	-0.023	0.9818
Has Children In House	-0.0652540	0.3097394	-0.211	0.834
Age (years)	-0.0181816	0.0184045	-0.988	0.3281
Duration	0.0012913	0.0262512	0.049	0.961
Male	0.0629352	0.2296261	0.274	0.7852

---