FB-Misinformation Descriptive Statistics

Goal

This document provides summary statistic for each variables, both continuous and categorical variables, and generates distribution for time duration variables. The data used in this script was processed by the main survey cleaning script and the follow up cleaning script. Specifically, we are using the following sets of data, which are located at ~fb_misinfo_interventions/data/chatfuel:

Main survey data

1. A wide-form dataset, misinfo_clean_wide.csv

• This dataset contains one row per respondent. It includes all the responents who consented to the study and passed data-quality checks (such as being from one of the eligible countries), regardless of whether they completed the intervention or the quiz. For the users who did not complete the intervention or the quiz, the relevant variables are set to missing.

2. A long-form dataset, misinfo_clean_long.csv

• This dataset contains 5 rows per respondent, one for each post (excluding attention check) that the user saw. It includes only the respondents who completed the intervention and the quiz, and passed data-quality checks (such as there being no missing values for the quiz questions).

Follow-up data

1. A wide-form dataset, misinfo_followup_clean_wide.csv

• This dataset contains one row per respondent. It includes all the responents who consented to the follow up study and passed data-quality checks, regardless of whether they the quiz. For the users who did not complete followup quiz, the relevant variables are set to missing.

2. long-form dataset, misinfo_followup_clean_long.csv

• This dataset contains 5 rows per respondent, one for each post (excluding attention check) that the user saw. It includes only the respondents who completed the followup quiz, and passed data-quality checks (such as there being no missing values for the quiz questions).

Summary

• This document begins with a data dictionary. It then provides a summary statistics for continuous variables, unordered categorical variables, and ordered categorical variables. The summary statistics are organized by several categories, including phase, treatment arm, chatbot completion status, payment amount, and whether the attention check was passed.

• Throughout this analysis we defined participants as follows:

  1. consented - participants who consented to start the main survey
  2. completed_quiz - participants who consented to start the main survey and completed the main survey misinformation quiz
  3. consented_followup - participants who completed the main survey misinformation quiz and consented to start the follow-up survey
  4. completed_followup_quiz - participants who consented to start the follow-up and completed the follow-up survey misinformation quiz

• We limited our analysis to individuals who consented to participate in the study when generating the summary statistics. All participants who consented must have a recorded value for phase_coded, country_coded, payment_amount_coded, and arm_coded.

• In the process of calculating the outcomes of the main survey data within the cleaning script, we conditioned on those completed_quiz and excluded participants missing any values in the sharing, reliability, and manipulative scores.

• In the follow up data, we only kept participants who consented to start the follow up survey. In the process of calculating the outcomes of the main survey data within the cleaning script, we conditioned on those completed_followup_quiz and excluded participants missing any values in the sharing, reliability, and manipulative scores.

• We then showed the distribution of time duration variables, including the time it took for participants to complete the intervention, the misinformation quiz, and the main survey chatbot. We compared the time variables between phases, country, payment amount, treatment arm, and passed attention check.

• We then showed the distribution of outcome variables and compare them between phases, country, payment amount, treatment arm, and passed attention check.

• We provided provided the distribution of missing values for each variable in the main survey and follow up data.

Loading R library

packages = c(
  "tidyverse", "data.table", "dtplyr", "rlang", "kableExtra", "haven", "ggcorrplot", "visdat", "VIM", "corrplot", "kableExtra", "fastDummies", "DescTools"
)

# Load the packages, install if necessary
new_packages = packages[!(packages %in% installed.packages()[,"Package"])]
if(length(new_packages)) install.packages(new_packages, dependencies = TRUE)
lapply(packages, require, character.only = TRUE) |> invisible()

Data Loading:

Main Survey

df_wide <- readRDS("./data/chatfuel/processed/misinfo_clean_wide.rds") |> filter(consent_coded_num == 1)

df_long <- readRDS("./data/chatfuel/processed/misinfo_clean_long.rds")

n_wide <- as.character(length(unique(df_wide$analytic_id)))
row_wide <- as.character(nrow(df_wide))
n_long <- as.character(length(unique(df_long$analytic_id)))
row_long <- as.character(nrow(df_long))

• The data is stored in the data folder in the Github directory. The data is loaded into R using the fread function from the data.table package. The data is loaded into two different formats: wide and long.

• We are currently having a total 70355 respondents who consented to the study and 42807 respondents who completed_quiz.

• The wide dataset contains 70355 rows and the long dataset contains 214035 rows.

Follow up

df_wide_followup = df_wide |> select(analytic_id, arm_coded, phase_coded) |> 
    inner_join(
        readRDS("data/chatfuel/processed/misinfo_followup_clean_wide.rds"),
        by = "analytic_id",
        relationship = "one-to-one",
        unmatched = c("drop", "error")
    )

df_long_followup <- readRDS("./data/chatfuel/processed/misinfo_followup_clean_long.rds")

n_followup_wide <- as.character(length(unique(df_wide_followup$analytic_id)))
row_followup_wide <- as.character(nrow(df_wide_followup))
n_followup_long <- as.character(length(unique(df_long_followup$analytic_id)))
row_followup_long <- as.character(nrow(df_long_followup))

• We are currently having a total 19950 respondents who consented to the follow up study and 18531 respondents who completed the follow up misinformation quiz.

• The wide dataset contains 19950 rows and the long dataset contains 92655 rows.

Data Dictionary

library(Hmisc)
a <- label(df_wide) %>% data.frame()
b <- label(df_long)  %>% data.frame() %>% slice(-1) # removing analytic_id to prevent duplication 
c <- label(df_wide_followup)  %>% data.frame() %>% slice(-1) # removing analytic_id to prevent duplication 
d <- label(df_long_followup)  %>% data.frame() %>% slice(-1) # removing analytic_id to prevent duplication 

# appending four dataframe and sorting by alphabetical order
combined_dictionary <- rbind(a, b, c, d)

# needing to put in alphabetical order

combined_dictionary %>%  kable(digits = 3, col.names = c("Variable Name", "Description")) |>
      kable_styling(bootstrap_options = c("striped", "hover")) |>
      kableExtra::scroll_box( height = "500px")

Variable Name	Description
analytic_id	Analytic ID
phase_coded	Phase
country_coded	Country
payment_amount_coded	Payment Amount
payment_condition_dollar
arm_coded	Treatment Arm
MisinfoChat_start_time	Misinformation Chatbot Start Time
total_duration	time duration between the start of the chatbot and the end of the misinformation quiz in minutes
intervention_duration	time duration between the start and the end of the intervention in minutes
misinfo_quiz_duration	time duration between the start and the end of the misinformation quiz in minutes
misinfoQuiz_end_time	Misinfo Quiz End Time
MisinfoQuiz_start_time	Misinfo Quiz Start Time
missing_treatment_arm
abnormal_posts
stage_reached_num	Stage Reached (Numeric, 1-7; 7 = Completed initial chatbot)
stage_reached_char	Stage Reached
consent_coded_num	Consent: 1 for I consent, start now, 0 for No let’s stop here
manipulative_reliable_order	Order of Manipulative and Reliable Questions
subphase	Sub-Phase (for Stratification)
strata	Subphase-Country-Payment Strata
mean_share_misinfo	Mean Share Score for Misinformation Posts
mean_reliable_misinfo	Mean Reliable Score for Misinformation Posts
mean_manipulative_misinfo	Mean Manipulative Score for Misinformation Posts
mean_share_nonmisinfo	Mean Share Score for Non-Misinformation Posts
mean_reliable_nonmisinfo	Mean Reliable Score for Non-Misinformation Posts
mean_manipulative_nonmisinfo	Mean Manipulative Score for Non-Misinformation Posts
sharing_discernment	Discernment in Sharing Misinformation and Non-Misinformation Posts
reliability_discernment	Discernment in Reliability of Misinformation and Non-Misinformation Posts
manipulation_discernment	Discernment in Manipulation of Misinformation and Non-Misinformation Posts
attention_check_passed	Attention Check Passed; 1 = PASS; 0 = FAIL
quiz_completed
quiz_completed_coded	Quiz Completion: 1 = Completed, 0 = Did not complete
post_id	Post ID
post_number	Post Order Number
post_type	Post Type
share_score	Share Score
reliable_score	Reliable Score
manipulative_score	Manipulative Score
phase_coded1	Phase
quiz_completed_coded_num	Quiz Completion
arm_coded1	Treatment arm
arm_coded2	Treatment Arm
phase_coded2	Phase
MisinfoChat_start_time_followup	Follow up Start Time
MisinfoQuiz_start_time_followup	Follow up Quiz Start Time
misinfoQuiz_end_time_followup	Follow up Quiz End Time
duration_complete_quiz_followup	Follow up Quiz Duration
time_since_completed_intervention	Follow up Time Since Completed Intervention in Main Survey
time_since_completed_quiz	Follow up Time Since Completed Misinformation Quiz in Main Survey
attention_check_passed_followup_coded_num
user_payment_amount_followup	Follow up Payment Amount
followup_entry_point	Follow up Entry Point
followup_reflection_1	Follow up Reflection 1
followup_reflection_2	Follow up Reflection 2
followup_reflection_3	Follow up Reflection 3
abnormal_followup_reflection_1	Follow up Abnormal Reflection 1
abnormal_followup_reflection_2	Follow up Abnormal Reflection 2
abnormal_followup_reflection_3	Follow up Abnormal Reflection 3
completed_quiz_followup_coded_num	Completed Followup Misinformation Quiz as Binary Variable
mean_share_misinfo_followup	Follow up Mean Misinfo Share Score
mean_reliable_misinfo_followup	Follow up Mean Misinfo Reliable Score
mean_manipulative_misinfo_followup	Follow up Mean Misinfo Manipulative Score
mean_share_nonmisinfo_followup	Follow up Mean Nonmisinfo Sharing Score
mean_reliable_nonmisinfo_followup	Follow up Mean Nonmisinfo Reliability Score
mean_manipulative_nonmisinfo_followup	Follow up Mean Nonmisinfo Manipulation Score
sharing_discernment_followup	Follow up Sharing Discernment
reliability_discernment_followup	Follow up Reliability Discernment
manipulation_discernment_followup	Follow up Manipulation Discernment
quiz_completed_followup	Follow up Quiz Completion
attention_check_passed_followup_coded_num1	Follow up Attention Check
completed_quiz_followup_coded_num1	Follow up Quiz Completion
post_number_followup	Follow up Post Order Number
post_id_followup	Follow up Post ID
manipulative_score_followup	Follow up Manipulative Score
reliable_score_followup	Follow up Reliable Score
share_score_followup	Follow up Share Score
post_type_followup	Follow up Post Type
is_followup	Is Follow up

Functions

Funtions for continuous variables

• We created a function to take each variable and group by a categorical variable to calculate the summary statistics for each group.

• The summary statistics include the number of missing values, the mean, the standard deviation, the standard error, the number of total observation, min, max, and decile values.

• The function combines the summary statistics for each selected variable and creates an html table

• The function also generates a csv file, which is saved in the Github directory ~./data/chatfuel_summary_statistics/

summary_stats_continous <- function(dt, variable, group = NULL){
  # Get descriptive statistics for a single continuous variable
  variable_name = variable 
  variable = sym(variable)

  if(!is.null(group)){
    group_name = group
    group = sym(group)
    dt <- dt |> group_by({{group}})
  } # grouping if group variable is selected

  tab = dt |> lazy_dt() |>
    mutate(
      `Number_Missing` = sum(is.na({{variable}})),  # calculate missing values
    ) |>
    filter(!is.na({{variable}})) |>  # filtering out the missing values for the variable
    summarise(
        Total_Observations = n(),
        Number_Missing = first(`Number_Missing`),
        Mean = as.numeric(mean({{variable}})),
        SD = as.numeric(sd({{variable}})),
        SE = as.numeric(sd({{variable}}) / sqrt(n())),
        Min = as.numeric(min({{variable}})),
        Max =  as.numeric(max({{variable}})),
        Decile_10 = as.numeric(quantile({{variable}}, probs = 0.1)),
        Decile_20 = as.numeric(quantile({{variable}}, probs = 0.2)),
        Decile_30 = as.numeric(quantile({{variable}}, probs = 0.3)),
        Decile_40 = as.numeric(quantile({{variable}}, probs = 0.4)),
        Decile_50 = as.numeric(quantile({{variable}}, probs = 0.5)),
        Decile_60 = as.numeric(quantile({{variable}}, probs = 0.6)),
        Decile_70 = as.numeric(quantile({{variable}}, probs = 0.7)),
        Decile_80 = as.numeric(quantile({{variable}}, probs = 0.8)),
        Decile_90 = as.numeric(quantile({{variable}}, probs = 0.9)),
        Decile_91 = as.numeric(quantile({{variable}}, probs = 0.91)),
        Decile_92 = as.numeric(quantile({{variable}}, probs = 0.92)),
        Decile_93 = as.numeric(quantile({{variable}}, probs = 0.93)),
        Decile_94 = as.numeric(quantile({{variable}}, probs = 0.94)),
        Decile_95 = as.numeric(quantile({{variable}}, probs = 0.95)),
        Decile_96 = as.numeric(quantile({{variable}}, probs = 0.96)),
        Decile_97 = as.numeric(quantile({{variable}}, probs = 0.97)),
        Decile_98 = as.numeric(quantile({{variable}}, probs = 0.98)),
        Decile_99 = as.numeric(quantile({{variable}}, probs = 0.99)),
        Decile_100 = as.numeric(quantile({{variable}}, probs = 1))
        
      ) |> # compute summary statistics 
      mutate(
        Proportion_Missing = Number_Missing / Total_Observations,
        Variable = variable_name
      ) |> # adds a column indicating the proportion of missing values relative to the total observations.
      collect()

  if (is.null(group)){
    tab <- tab |> select(Variable, everything())
  } else {
    tab <- tab |> mutate(Group = paste(group_name, ": ", {{group}})) |> select(Variable, Group, everything()) |> select(-{{group}})
  }

  return(tab)
}

summary_stats_table_continous <- function(dt, variables, group = NULL,
  filename = NULL, print_html = TRUE, caption = NULL
  ){
  # Get descriptive statistics for a list of continuous variables and return a combined table
  results <- lapply(variables, function(x) summary_stats_continous(dt, x, group)) # apply the summary_stats_continous function to each variable in the list
  results <- bind_rows(results) # bind the list of dataframes into a single dataframe

  if(!is.null(filename)){
    write.csv(results, paste0("./data/chatfuel_summary_statistics/", filename)) # save as csv file
  }

  if (print_html){
    results = results |> 
      kable(digits = 3,caption = caption) |>
      kable_styling(bootstrap_options = c("striped", "hover")) |>
      scroll_box( height = "500px") # generate html table 
  }
  
  return(results)
}

summary_stats_continous_outcomes <- function(dt, variable, group = NULL){
  # Get descriptive statistics for a single continuous variable
  variable_name = variable 
  variable = sym(variable)

  if(!is.null(group)){
    group_name = group
    group = sym(group)
    dt <- dt |> group_by({{group}})
  } # grouping if group variable is selected

  tab = dt |> lazy_dt() |>
    mutate(
      `Number_Missing` = sum(is.na({{variable}})),  # calculate missing values
    ) |>
    filter(!is.na({{variable}})) |>  # filtering out the missing values for the variable
    summarise(
        Total_Observations = n(),
        Number_Missing = first(`Number_Missing`),
        Mean = as.numeric(mean({{variable}})),
        SD = as.numeric(sd({{variable}})),
        SE = as.numeric(sd({{variable}}) / sqrt(n()))
      ) |> # compute summary statistics 
      mutate(
        Proportion_Missing = Number_Missing / Total_Observations,
        Variable = variable_name
      ) |> # adds a column indicating the proportion of missing values relative to the total observations.
      collect()

  if (is.null(group)){
    tab <- tab |> select(Variable, everything())
  } else {
    tab <- tab |> mutate(Group = paste(group_name, ": ", {{group}})) |> select(Variable, Group, everything()) |> select(-{{group}})
  }

  return(tab)
}

summary_stats_table_continous_outcomes <- function(dt, variables, group = NULL,
  filename = NULL, print_html = TRUE, caption = NULL
  ){
  # Get descriptive statistics for a list of continuous variables and return a combined table
  results <- lapply(variables, function(x) summary_stats_continous_outcomes(dt, x, group)) # apply the summary_stats_continous function to each variable in the list
  results <- bind_rows(results) # bind the list of dataframes into a single dataframe

  if(!is.null(filename)){
    write.csv(results, paste0("./data/chatfuel_summary_statistics/", filename)) # save as csv file
  }

  if (print_html){
    results = results |> 
      kable(digits = 3,caption = caption) |>
      kable_styling(bootstrap_options = c("striped", "hover")) |>
      scroll_box( height = "500px") # generate html table 
  }
  
  return(results)
}

Function for unordered categorical variables

• We created a function to take each variable and group by a categorical variable to calculate the summary statistics for each group.

• The summary statistics include the number for each count for each value and the proportion for each value

• The function combines the summary statistics for each selected variable and creates an html table

• The function also generates a csv file, which is saved in the Github directory ~./data/chatfuel_summary_statistics/

# Define the function to create summary statistics for unordered categorical variable
summary_stats_categorical <- function(data, variables, group = NULL) {
  combined_results <- data.frame(
    Variable = character(),
    Group = character(),
    Category = character(),
    Frequency = integer(),
    Proportion = numeric(),
    stringsAsFactors = FALSE
  ) # preset a dataframe
  
  for (variable in variables) {
    if (!is.null(group)) {
      # Grouped calculations
      summary_df <- data %>%
        group_by_at(vars(c(group, variable))) %>%
        summarise(Frequency = n(), .groups = 'drop') %>%
        mutate(Proportion = Frequency / sum(Frequency)) %>%
        ungroup() %>%
        mutate(Variable = variable,
               Group = ifelse(is.na(.data[[group]]), "Missing", as.character(.data[[group]])),
               Category = ifelse(is.na(.data[[variable]]), "Missing", as.character(.data[[variable]]))) %>%
        select(Variable, Group, Category, Frequency, Proportion)
    } else {
      # Ungrouped calculations
     summary_df <- data %>%
      select(!!sym(variable)) %>%
      mutate(Category = factor(!!sym(variable), exclude = NULL)) %>% # Include missing values as a factor level
      group_by(Category) %>%
      summarise(
        Frequency = n(),
        Proportion = n() / nrow(data),
        .groups = 'drop' # Avoid regrouping warning
      ) %>%
      ungroup() %>%
      mutate(Variable = variable) %>% # Add variable name
      select(Variable, everything())
    }
    
    # Combine the result with the overall results
    combined_results <- bind_rows(combined_results, summary_df)
  }
  
  return(combined_results)
}

summary_stats_tab_categorical <- function(dt, variables, group = NULL,
  filename = NULL, print_html = TRUE, caption = NULL
  ){
  
  results <- lapply(variables, function(x) summary_stats_categorical(dt, x, group))
  results <- bind_rows(results) 

  if(!is.null(filename)){
    write.csv(results, paste0("./data/chatfuel_summary_statistics/", filename)) # save as csv file
  }

  if (print_html){
    results = results |> 
      kable(digits = 3,caption = caption) |>
      kable_styling(bootstrap_options = c("striped", "hover")) |>
      scroll_box( height = "500px") # generate html table 
  }
  
  return(results)
}

Function for generating probability distribution

Density plot

create_density_plot <- function(data, x_var, group_var, x_min, x_max, title, x_label, y_label,  file_basename)  {
  plt <-
  ggplot(data, aes_string(x = x_var, group = group_var, fill = group_var)) +
    geom_density(alpha = 0.7, color = "black") +
    labs(title = title,
         x = x_label,
         y = y_label) +
    scale_x_continuous(limits = c(x_min, x_max)) +
    theme_minimal()+
    guides(fill = guide_legend(title = NULL))

  print(plt)

  # Export the data used for the graph to CSV
  write.csv(data, paste0("./data/chatfuel_summary_statistics/", file_basename, ".csv"))

  # Save the plot to JPG
  ggsave(paste0("./data/chatfuel_summary_statistics/", file_basename, ".jpg"), plt, device = "jpg", width = 10, height = 8)

  # Save the plot to PDF
  ggsave(paste0("./data/chatfuel_summary_statistics/", file_basename, ".pdf"), plt, device = "pdf", width = 10, height = 8)
}  

Histogram

create_histogram_grouped <- function(data, x_var, group_var, bin_width, title, x_label, y_label, line_color = "black", file_basename) {
  # Generate the histogram
  plt <- ggplot(data, aes_string(x = x_var, fill = group_var)) +
    geom_histogram(binwidth = bin_width, color = line_color, position = "identity", alpha = 0.5) +
    labs(title = title, x = x_label, y = y_label) +
    theme_minimal() +
    guides(fill = guide_legend(title = NULL))

  # Print the plot
  print(plt)
  
  # Export the data used for the graph to CSV
  write.csv(data, paste0("./data/chatfuel_summary_statistics/", file_basename, ".csv"))

  # Save the plot to JPG
  ggsave(paste0("./data/chatfuel_summary_statistics/", file_basename, ".jpg"), plt, device = "jpg", width = 10, height = 8)

  # Save the plot to PDF
  ggsave(paste0("./data/chatfuel_summary_statistics/", file_basename, ".pdf"), plt, device = "pdf", width = 10, height = 8)
}

Summary Statistics for Continous Variables

User level variables include responses from individuals who consented to participate in the study.

Post level variables only include responses from individuals who completed the misinformation quiz.

User level variables

continous_var <- c("intervention_duration", "total_duration", "misinfo_quiz_duration",
                    "mean_share_misinfo", "mean_reliable_misinfo", "mean_manipulative_misinfo", "mean_share_nonmisinfo",
                    "mean_reliable_nonmisinfo", "mean_manipulative_nonmisinfo", "sharing_discernment", "reliability_discernment", "manipulation_discernment")


summary_stats_table_continous(df_wide,
  variables = continous_var,
  filename = "continous_stat_results_consent.csv",
  caption = "Summary Statistics for Continous Variables",
  print_html = TRUE
  )

Summary Statistics for Continous Variables

Variable	Total_Observations	Number_Missing	Mean	SD	SE	Min	Max	Decile_10	Decile_20	Decile_30	Decile_40	Decile_50	Decile_60	Decile_70	Decile_80	Decile_90	Decile_91	Decile_92	Decile_93	Decile_94	Decile_95	Decile_96	Decile_97	Decile_98	Decile_99	Decile_100	Proportion_Missing
intervention_duration	52068	18287	278.194	3178.435	13.929	3.967	128975.6	9.250	11.700	13.033	14.300	15.733	17.683	20.767	26.900	60.638	76.232	102.821	144.437	222.816	349.768	550.801	870.765	1458.586	3847.463	128975.6	0.351
total_duration	43978	26377	636.825	5498.200	26.218	6.850	135561.0	16.883	19.367	21.467	23.700	26.383	30.133	36.183	51.450	239.367	317.518	426.377	558.294	732.024	990.251	1361.682	2032.999	3890.957	12080.045	135561.0	0.600
misinfo_quiz_duration	43973	26382	299.701	4124.102	19.667	1.200	133550.8	3.817	4.517	5.117	5.717	6.400	7.250	8.400	10.400	17.367	19.709	23.387	29.049	40.917	73.880	169.487	401.799	844.704	2667.263	133550.8	0.600
mean_share_misinfo	42807	27548	0.529	0.383	0.002	0.000	1.0	0.000	0.000	0.333	0.333	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.0	0.644
mean_reliable_misinfo	42807	27548	2.595	1.245	0.006	1.000	5.0	1.000	1.333	1.667	2.000	2.333	2.667	3.333	3.667	4.667	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.0	0.644
mean_manipulative_misinfo	42807	27548	3.038	1.291	0.006	1.000	5.0	1.333	1.667	2.333	2.667	3.000	3.667	3.667	4.333	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.0	0.644
mean_share_nonmisinfo	42807	27548	0.720	0.354	0.002	0.000	1.0	0.000	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.0	0.644
mean_reliable_nonmisinfo	42807	27548	3.107	1.334	0.006	1.000	5.0	1.000	1.500	2.000	3.000	3.000	3.500	4.000	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.0	0.644
mean_manipulative_nonmisinfo	42807	27548	2.547	1.333	0.006	1.000	5.0	1.000	1.000	1.500	2.000	2.500	3.000	3.000	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.0	0.644
sharing_discernment	42807	27548	0.191	0.405	0.002	-1.000	1.0	-0.333	0.000	0.000	0.000	0.167	0.333	0.333	0.500	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.0	0.644
reliability_discernment	42807	27548	0.513	1.331	0.006	-4.000	4.0	-1.000	-0.500	0.000	0.000	0.333	0.667	1.000	1.667	2.333	2.500	2.667	2.667	2.667	2.833	3.000	3.333	3.667	4.000	4.0	0.644
manipulation_discernment	42807	27548	0.491	1.388	0.007	-4.000	4.0	-1.167	-0.500	0.000	0.000	0.333	0.667	1.000	1.667	2.333	2.500	2.667	2.667	2.667	3.000	3.000	3.500	3.667	4.000	4.0	0.644

User level variables, by arm

summary_stats_table_continous(df_wide,
  variables = continous_var,
  group = "arm_coded",
  filename = "continous_stat_results_consent_by_arm.csv",
  caption = "Summary Statistics for Continous Variables, by arm",
  print_html = TRUE
  )

Summary Statistics for Continous Variables, by arm

Variable	Group	Total_Observations	Number_Missing	Mean	SD	SE	Min	Max	Decile_10	Decile_20	Decile_30	Decile_40	Decile_50	Decile_60	Decile_70	Decile_80	Decile_90	Decile_91	Decile_92	Decile_93	Decile_94	Decile_95	Decile_96	Decile_97	Decile_98	Decile_99	Decile_100	Proportion_Missing
intervention_duration	arm_coded : Game	4634	2782	308.098	3392.556	49.837	5.767	128975.58	7.067	8.183	10.250	13.170	17.383	22.610	29.637	43.373	126.495	164.525	219.512	292.338	455.129	609.729	862.239	1209.109	1771.018	4360.392	128975.58	0.600
intervention_duration	arm_coded : Long Baseline	10733	3405	234.160	2572.233	24.828	3.967	83247.23	12.867	13.933	15.000	16.150	17.533	19.433	22.323	27.693	52.707	64.659	82.045	119.439	181.778	302.593	495.293	819.585	1373.283	3315.505	83247.23	0.317
intervention_duration	arm_coded : Original Baseline	11208	2399	179.598	2291.842	21.648	4.217	93628.07	7.083	7.967	8.750	9.650	10.667	11.900	13.617	16.467	26.488	29.012	33.191	39.635	52.534	86.364	166.042	350.876	712.181	2261.329	93628.07	0.214
intervention_duration	arm_coded : SMS	20577	7145	316.006	3583.288	24.980	5.683	125646.87	12.300	13.333	14.367	15.433	16.867	18.727	21.717	27.483	61.457	76.488	105.147	143.144	232.875	351.197	549.554	854.909	1464.843	4106.158	125646.87	0.347
intervention_duration	arm_coded : Video	4916	2556	412.666	3992.914	56.949	9.000	106594.78	10.967	11.933	12.850	13.983	15.533	18.117	23.033	39.917	248.008	311.084	425.330	521.353	657.562	853.650	1197.477	1576.660	2577.520	5481.641	106594.78	0.520
total_duration	arm_coded : Game	3481	3935	764.863	6309.786	106.946	8.150	132509.32	14.433	17.683	21.100	24.917	30.067	36.783	48.117	75.733	398.183	518.893	669.323	842.643	1126.410	1417.883	2095.240	3206.360	5220.943	13115.590	132509.32	1.130
total_duration	arm_coded : Long Baseline	9173	4965	557.098	4724.967	49.334	12.033	98018.13	18.983	21.000	22.993	25.113	27.633	31.303	36.950	49.950	198.173	279.461	378.037	516.126	681.453	927.490	1278.223	1798.385	3765.373	10519.382	98018.13	0.541
total_duration	arm_coded : Original Baseline	9152	4455	583.971	5461.230	57.086	6.850	133564.78	13.385	15.353	17.117	18.950	21.167	23.983	28.323	37.443	137.642	192.395	267.711	383.994	533.313	715.074	984.319	1468.388	2802.173	9961.019	133564.78	0.487
total_duration	arm_coded : SMS	18036	9686	642.503	5469.059	40.723	11.733	125650.88	18.767	20.850	22.800	24.983	27.667	31.250	37.300	52.167	229.658	302.993	405.570	523.650	687.817	962.025	1343.770	1973.753	4018.640	13533.750	125650.88	0.537
total_duration	arm_coded : Video	4136	3336	798.080	6477.553	100.721	11.450	135561.02	17.717	19.883	21.767	23.833	26.733	30.950	39.675	83.050	529.158	633.308	772.633	977.788	1259.270	1506.842	1937.587	2729.385	4744.033	12510.097	135561.02	0.807
misinfo_quiz_duration	arm_coded : Game	3481	3935	436.080	5499.362	93.210	1.517	132476.25	3.550	4.233	4.800	5.350	6.000	6.767	7.850	9.750	16.083	19.490	23.417	33.820	53.370	106.617	237.983	549.193	1425.103	5131.283	132476.25	1.130
misinfo_quiz_duration	arm_coded : Long Baseline	9172	4966	230.460	3179.488	33.199	1.400	93547.08	3.600	4.250	4.788	5.350	5.958	6.717	7.817	9.633	16.032	17.683	20.683	26.104	35.168	60.845	132.783	360.972	731.122	1950.640	93547.08	0.541
misinfo_quiz_duration	arm_coded : Original Baseline	9151	4456	337.267	4459.293	46.616	1.333	133550.80	3.867	4.583	5.200	5.833	6.533	7.400	8.600	10.783	18.367	21.000	25.217	32.083	52.200	101.400	211.367	467.992	824.100	2734.067	133550.80	0.487
misinfo_quiz_duration	arm_coded : SMS	18033	9689	269.978	3873.453	28.845	1.200	121920.62	3.967	4.650	5.250	5.867	6.567	7.450	8.567	10.617	17.197	19.352	22.365	27.433	35.951	57.333	126.447	332.730	756.945	2138.951	121920.62	0.537
misinfo_quiz_duration	arm_coded : Video	4136	3336	384.948	4849.992	75.414	1.617	120152.10	3.983	4.717	5.350	6.033	6.767	7.600	8.700	10.767	20.450	24.064	29.397	40.720	60.685	162.488	301.123	594.364	1164.892	3107.946	120152.10	0.807
mean_share_misinfo	arm_coded : Game	3242	4174	0.730	0.316	0.006	0.000	1.00	0.333	0.333	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	1.287
mean_share_misinfo	arm_coded : Long Baseline	9163	4975	0.627	0.371	0.004	0.000	1.00	0.000	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.543
mean_share_misinfo	arm_coded : Original Baseline	8834	4773	0.621	0.367	0.004	0.000	1.00	0.000	0.333	0.333	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.540
mean_share_misinfo	arm_coded : SMS	17714	10008	0.369	0.351	0.003	0.000	1.00	0.000	0.000	0.000	0.333	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.565
mean_share_misinfo	arm_coded : Video	3854	3618	0.651	0.366	0.006	0.000	1.00	0.000	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.939
mean_reliable_misinfo	arm_coded : Game	3242	4174	2.908	1.257	0.022	1.000	5.00	1.333	1.667	2.000	2.333	2.667	3.333	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	1.287
mean_reliable_misinfo	arm_coded : Long Baseline	9163	4975	2.865	1.265	0.013	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.333	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.543
mean_reliable_misinfo	arm_coded : Original Baseline	8834	4773	2.819	1.273	0.014	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.000	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.540
mean_reliable_misinfo	arm_coded : SMS	17714	10008	2.242	1.117	0.008	1.000	5.00	1.000	1.000	1.333	1.667	2.000	2.333	2.667	3.333	3.667	4.000	4.000	4.000	4.333	4.667	5.000	5.000	5.000	5.000	5.00	0.565
mean_reliable_misinfo	arm_coded : Video	3854	3618	2.794	1.279	0.021	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.000	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.939
mean_manipulative_misinfo	arm_coded : Game	3242	4174	2.788	1.264	0.022	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.000	3.667	4.000	4.667	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	1.287
mean_manipulative_misinfo	arm_coded : Long Baseline	9163	4975	3.021	1.282	0.013	1.000	5.00	1.000	1.667	2.333	2.667	3.000	3.667	3.667	4.333	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.543
mean_manipulative_misinfo	arm_coded : Original Baseline	8834	4773	2.957	1.291	0.014	1.000	5.00	1.000	1.667	2.000	2.333	3.000	3.333	3.667	4.333	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.540
mean_manipulative_misinfo	arm_coded : SMS	17714	10008	3.171	1.285	0.010	1.000	5.00	1.333	2.000	2.333	2.667	3.333	3.667	4.000	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.565
mean_manipulative_misinfo	arm_coded : Video	3854	3618	2.861	1.296	0.021	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.333	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.939
mean_share_nonmisinfo	arm_coded : Game	3242	4174	0.797	0.309	0.005	0.000	1.00	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	1.287
mean_share_nonmisinfo	arm_coded : Long Baseline	9163	4975	0.769	0.333	0.003	0.000	1.00	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.543
mean_share_nonmisinfo	arm_coded : Original Baseline	8834	4773	0.759	0.340	0.004	0.000	1.00	0.000	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.540
mean_share_nonmisinfo	arm_coded : SMS	17714	10008	0.642	0.372	0.003	0.000	1.00	0.000	0.500	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.565
mean_share_nonmisinfo	arm_coded : Video	3854	3618	0.804	0.314	0.005	0.000	1.00	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.939
mean_reliable_nonmisinfo	arm_coded : Game	3242	4174	3.110	1.334	0.023	1.000	5.00	1.500	1.500	2.000	2.500	3.000	3.500	4.000	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	1.287
mean_reliable_nonmisinfo	arm_coded : Long Baseline	9163	4975	3.285	1.320	0.014	1.000	5.00	1.500	2.000	2.500	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.543
mean_reliable_nonmisinfo	arm_coded : Original Baseline	8834	4773	3.219	1.356	0.014	1.000	5.00	1.500	2.000	2.500	3.000	3.000	3.500	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.540
mean_reliable_nonmisinfo	arm_coded : SMS	17714	10008	2.932	1.303	0.010	1.000	5.00	1.000	1.500	2.000	2.500	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.565
mean_reliable_nonmisinfo	arm_coded : Video	3854	3618	3.230	1.357	0.022	1.000	5.00	1.500	2.000	2.000	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.939
mean_manipulative_nonmisinfo	arm_coded : Game	3242	4174	2.498	1.320	0.023	1.000	5.00	1.000	1.000	1.500	2.000	2.000	3.000	3.000	4.000	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	1.287
mean_manipulative_nonmisinfo	arm_coded : Long Baseline	9163	4975	2.620	1.358	0.014	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.543
mean_manipulative_nonmisinfo	arm_coded : Original Baseline	8834	4773	2.573	1.362	0.014	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.540
mean_manipulative_nonmisinfo	arm_coded : SMS	17714	10008	2.528	1.305	0.010	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.565
mean_manipulative_nonmisinfo	arm_coded : Video	3854	3618	2.445	1.340	0.022	1.000	5.00	1.000	1.000	1.500	1.500	2.000	3.000	3.000	3.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.939
sharing_discernment	arm_coded : Game	3242	4174	0.067	0.366	0.006	-1.000	1.00	-0.500	-0.167	0.000	0.000	0.000	0.000	0.333	0.333	0.667	0.667	0.667	0.667	0.667	0.667	0.667	0.667	1.000	1.000	1.00	1.287
sharing_discernment	arm_coded : Long Baseline	9163	4975	0.142	0.382	0.004	-1.000	1.00	-0.333	0.000	0.000	0.000	0.000	0.167	0.333	0.500	0.667	0.667	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.00	0.543
sharing_discernment	arm_coded : Original Baseline	8834	4773	0.138	0.386	0.004	-1.000	1.00	-0.333	0.000	0.000	0.000	0.000	0.167	0.333	0.500	0.667	0.667	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.00	0.540
sharing_discernment	arm_coded : SMS	17714	10008	0.273	0.423	0.003	-1.000	1.00	-0.167	0.000	0.000	0.167	0.333	0.333	0.500	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.565
sharing_discernment	arm_coded : Video	3854	3618	0.153	0.374	0.006	-1.000	1.00	-0.167	0.000	0.000	0.000	0.000	0.167	0.333	0.500	0.667	0.667	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.00	0.939
reliability_discernment	arm_coded : Game	3242	4174	0.202	1.178	0.021	-4.000	4.00	-1.167	-0.667	-0.333	0.000	0.000	0.333	0.667	1.000	1.667	1.833	2.000	2.000	2.167	2.333	2.500	2.667	3.000	3.500	4.00	1.287
reliability_discernment	arm_coded : Long Baseline	9163	4975	0.420	1.325	0.014	-4.000	4.00	-1.167	-0.500	-0.167	0.000	0.167	0.667	1.000	1.500	2.167	2.333	2.500	2.667	2.667	2.667	2.833	3.000	3.500	4.000	4.00	0.543
reliability_discernment	arm_coded : Original Baseline	8834	4773	0.400	1.303	0.014	-4.000	4.00	-1.167	-0.500	-0.167	0.000	0.167	0.500	1.000	1.333	2.167	2.333	2.333	2.500	2.667	2.667	2.667	3.000	3.333	4.000	4.00	0.540
reliability_discernment	arm_coded : SMS	17714	10008	0.690	1.369	0.010	-4.000	4.00	-0.833	-0.333	0.000	0.167	0.500	0.833	1.333	2.000	2.667	2.667	2.667	2.667	3.000	3.000	3.333	3.667	4.000	4.000	4.00	0.565
reliability_discernment	arm_coded : Video	3854	3618	0.435	1.249	0.020	-4.000	4.00	-1.000	-0.500	0.000	0.000	0.333	0.500	1.000	1.333	2.167	2.167	2.333	2.500	2.667	2.667	2.667	3.000	3.333	3.667	4.00	0.939
manipulation_discernment	arm_coded : Game	3242	4174	0.290	1.238	0.022	-4.000	4.00	-1.167	-0.500	-0.167	0.000	0.167	0.500	0.667	1.167	2.000	2.000	2.000	2.167	2.333	2.667	2.667	2.667	3.167	3.667	4.00	1.287
manipulation_discernment	arm_coded : Long Baseline	9163	4975	0.401	1.351	0.014	-4.000	4.00	-1.333	-0.667	-0.167	0.000	0.167	0.667	1.000	1.500	2.167	2.333	2.333	2.667	2.667	2.667	3.000	3.000	3.500	4.000	4.00	0.543
manipulation_discernment	arm_coded : Original Baseline	8834	4773	0.384	1.351	0.014	-4.000	4.00	-1.333	-0.500	-0.167	0.000	0.167	0.500	1.000	1.333	2.167	2.333	2.500	2.667	2.667	2.667	3.000	3.333	3.667	4.000	4.00	0.540
manipulation_discernment	arm_coded : SMS	17714	10008	0.643	1.451	0.011	-4.000	4.00	-1.000	-0.500	0.000	0.167	0.500	0.833	1.333	2.000	2.667	2.667	2.667	3.000	3.000	3.167	3.500	3.667	4.000	4.000	4.00	0.565
manipulation_discernment	arm_coded : Video	3854	3618	0.416	1.316	0.021	-4.000	4.00	-1.167	-0.500	0.000	0.000	0.167	0.500	1.000	1.500	2.167	2.333	2.500	2.667	2.667	2.667	3.000	3.000	3.500	4.000	4.00	0.939

User level variables, by phase

summary_stats_table_continous(df_wide,
  variables = continous_var,
  group = "phase_coded",
  filename = "continous_stat_results_consent_by_phase.csv",
  caption = "Summary Statistics for Continous Variables, by phase",
  print_html = TRUE
  )

Summary Statistics for Continous Variables, by phase

Variable	Group	Total_Observations	Number_Missing	Mean	SD	SE	Min	Max	Decile_10	Decile_20	Decile_30	Decile_40	Decile_50	Decile_60	Decile_70	Decile_80	Decile_90	Decile_91	Decile_92	Decile_93	Decile_94	Decile_95	Decile_96	Decile_97	Decile_98	Decile_99	Decile_100	Proportion_Missing
intervention_duration	phase_coded : Phase 1	20694	9033	316.339	3571.987	24.831	4.617	128975.58	8.033	10.067	11.650	13.133	14.817	17.180	21.133	29.983	90.682	116.928	162.489	230.900	335.411	497.982	679.997	1078.406	1645.818	4576.045	128975.58	0.437
intervention_duration	phase_coded : Phase 2	31374	9254	253.034	2889.448	16.313	3.967	94308.27	11.067	12.617	13.717	14.833	16.167	17.917	20.583	25.533	47.895	57.674	72.997	102.904	151.953	260.885	436.356	713.886	1323.139	3180.126	94308.27	0.295
total_duration	phase_coded : Phase 1	16818	12909	686.387	6144.889	47.383	7.133	135561.02	15.767	18.483	20.917	23.400	26.450	30.850	38.550	59.873	325.932	419.484	530.287	668.444	846.980	1142.517	1477.040	2285.286	4229.497	9395.048	135561.02	0.768
total_duration	phase_coded : Phase 2	27160	13468	606.136	5056.339	30.681	6.850	98018.13	17.583	19.800	21.733	23.850	26.350	29.773	35.050	47.917	188.365	258.681	352.014	486.017	635.745	896.550	1291.778	1850.392	3753.917	13068.810	98018.13	0.496
misinfo_quiz_duration	phase_coded : Phase 1	16817	12910	359.649	4841.823	37.337	1.517	133550.80	3.983	4.750	5.417	6.083	6.817	7.717	8.933	11.200	19.947	23.159	28.595	39.063	59.500	122.323	261.159	521.861	1057.455	3236.010	133550.80	0.768
misinfo_quiz_duration	phase_coded : Phase 2	27156	13472	262.577	3608.363	21.897	1.200	97110.57	3.733	4.400	4.950	5.500	6.150	6.950	8.067	9.917	16.050	17.883	20.767	25.522	32.750	54.104	121.337	332.147	726.955	2057.485	97110.57	0.496
mean_share_misinfo	phase_coded : Phase 1	15693	14034	0.581	0.378	0.003	0.000	1.00	0.000	0.333	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.894
mean_share_misinfo	phase_coded : Phase 2	27114	13514	0.499	0.382	0.002	0.000	1.00	0.000	0.000	0.333	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.498
mean_reliable_misinfo	phase_coded : Phase 1	15693	14034	2.675	1.270	0.010	1.000	5.00	1.000	1.333	1.667	2.000	2.333	3.000	3.667	4.000	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.894
mean_reliable_misinfo	phase_coded : Phase 2	27114	13514	2.548	1.227	0.007	1.000	5.00	1.000	1.333	1.667	2.000	2.333	2.667	3.333	3.667	4.333	4.667	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.498
mean_manipulative_misinfo	phase_coded : Phase 1	15693	14034	2.945	1.299	0.010	1.000	5.00	1.000	1.667	2.000	2.333	3.000	3.333	3.667	4.333	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.894
mean_manipulative_misinfo	phase_coded : Phase 2	27114	13514	3.092	1.284	0.008	1.000	5.00	1.333	1.667	2.333	2.667	3.000	3.667	4.000	4.333	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.498
mean_share_nonmisinfo	phase_coded : Phase 1	15693	14034	0.747	0.344	0.003	0.000	1.00	0.000	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.894
mean_share_nonmisinfo	phase_coded : Phase 2	27114	13514	0.704	0.359	0.002	0.000	1.00	0.000	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.498
mean_reliable_nonmisinfo	phase_coded : Phase 1	15693	14034	3.129	1.350	0.011	1.000	5.00	1.000	1.700	2.000	3.000	3.000	3.500	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.894
mean_reliable_nonmisinfo	phase_coded : Phase 2	27114	13514	3.095	1.325	0.008	1.000	5.00	1.000	1.500	2.000	3.000	3.000	3.500	4.000	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.498
mean_manipulative_nonmisinfo	phase_coded : Phase 1	15693	14034	2.511	1.345	0.011	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.894
mean_manipulative_nonmisinfo	phase_coded : Phase 2	27114	13514	2.568	1.326	0.008	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.498
sharing_discernment	phase_coded : Phase 1	15693	14034	0.166	0.397	0.003	-1.000	1.00	-0.333	0.000	0.000	0.000	0.000	0.167	0.333	0.500	0.667	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.894
sharing_discernment	phase_coded : Phase 2	27114	13514	0.205	0.408	0.002	-1.000	1.00	-0.333	0.000	0.000	0.000	0.167	0.333	0.333	0.500	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.498
reliability_discernment	phase_coded : Phase 1	15693	14034	0.454	1.302	0.010	-4.000	4.00	-1.000	-0.500	0.000	0.000	0.333	0.667	1.000	1.500	2.167	2.333	2.500	2.667	2.667	2.667	3.000	3.167	3.500	4.000	4.00	0.894
reliability_discernment	phase_coded : Phase 2	27114	13514	0.547	1.347	0.008	-4.000	4.00	-1.000	-0.500	0.000	0.000	0.333	0.667	1.167	1.667	2.333	2.500	2.667	2.667	2.667	2.833	3.000	3.333	3.667	4.000	4.00	0.498
manipulation_discernment	phase_coded : Phase 1	15693	14034	0.433	1.363	0.011	-4.000	4.00	-1.167	-0.500	-0.167	0.000	0.167	0.667	1.000	1.500	2.333	2.500	2.667	2.667	2.667	2.833	3.000	3.333	3.667	4.000	4.00	0.894
manipulation_discernment	phase_coded : Phase 2	27114	13514	0.524	1.402	0.009	-4.000	4.00	-1.167	-0.500	0.000	0.000	0.333	0.667	1.167	1.667	2.500	2.667	2.667	2.667	2.667	3.000	3.167	3.500	4.000	4.000	4.00	0.498

User level variables by country

summary_stats_table_continous(df_wide,
  variables = continous_var,
  group = "country_coded",
  filename = "continous_stat_results_consent_by_country.csv",
  caption = "Summary Statistics for Continous Variables, by country",
  print_html = TRUE
  )

Summary Statistics for Continous Variables, by country

Variable	Group	Total_Observations	Number_Missing	Mean	SD	SE	Min	Max	Decile_10	Decile_20	Decile_30	Decile_40	Decile_50	Decile_60	Decile_70	Decile_80	Decile_90	Decile_91	Decile_92	Decile_93	Decile_94	Decile_95	Decile_96	Decile_97	Decile_98	Decile_99	Decile_100	Proportion_Missing
intervention_duration	country_coded : Ghana	6008	2129	160.389	1937.515	24.997	4.283	81425.83	9.783	12.117	13.650	15.183	16.817	18.883	22.232	27.950	50.488	57.902	71.956	96.605	153.968	255.894	447.220	660.265	1108.237	2186.374	81425.83	0.354
intervention_duration	country_coded : Kenya	6865	2437	301.148	3513.647	42.407	4.617	93585.18	8.807	11.467	12.720	13.877	15.133	16.867	19.397	24.203	46.823	53.400	67.165	95.491	134.726	219.973	433.041	768.429	1434.685	4227.299	93585.18	0.355
intervention_duration	country_coded : Nigeria	23368	8010	279.834	3189.211	20.863	3.967	125646.87	9.567	11.933	13.333	14.683	16.200	18.250	21.483	28.010	65.872	82.899	113.055	163.544	258.979	403.581	602.914	953.932	1533.669	3881.215	125646.87	0.343
intervention_duration	country_coded : South Africa	15827	5711	310.536	3381.305	26.877	4.517	128975.58	8.800	11.400	12.633	13.717	15.000	16.767	19.687	25.767	68.757	88.572	120.595	164.367	243.095	355.145	547.163	849.014	1476.629	4740.829	128975.58	0.361
total_duration	country_coded : Ghana	5086	3051	303.026	2845.602	39.901	9.150	84883.20	18.100	20.817	23.117	25.650	28.467	32.367	38.425	50.633	150.083	203.699	267.993	379.063	504.225	667.679	933.637	1344.075	1997.582	4612.342	84883.20	0.600
total_duration	country_coded : Kenya	5883	3419	817.022	6643.332	86.614	6.850	119665.73	16.570	19.073	21.160	23.250	25.550	28.870	34.057	46.093	190.547	278.467	382.649	556.186	764.495	1114.907	1471.279	2766.620	5249.461	18642.040	119665.73	0.581
total_duration	country_coded : Nigeria	19236	12142	672.953	5760.379	41.533	7.050	135561.02	17.383	20.000	22.267	24.683	27.500	31.467	37.983	55.700	293.333	391.603	504.663	624.505	820.407	1119.546	1444.673	2213.253	4023.983	12557.414	135561.02	0.631
total_duration	country_coded : South Africa	13773	7765	632.661	5318.933	45.322	7.467	133564.78	16.183	18.400	20.183	22.083	24.367	27.700	33.483	48.803	226.153	301.007	392.541	522.774	697.836	903.903	1341.753	2019.923	4252.535	13463.291	133564.78	0.564
misinfo_quiz_duration	country_coded : Ghana	5086	3051	100.775	1900.496	26.649	1.617	84875.15	4.317	5.067	5.717	6.350	7.133	8.017	9.133	11.150	16.908	18.523	20.967	24.034	28.255	37.263	64.977	127.277	392.175	1198.757	84875.15	0.600
misinfo_quiz_duration	country_coded : Kenya	5882	3420	500.689	5740.277	74.846	1.200	119539.68	3.817	4.600	5.200	5.773	6.417	7.217	8.317	10.100	15.745	16.900	19.259	23.961	32.981	56.361	146.015	379.651	995.730	6127.840	119539.68	0.581
misinfo_quiz_duration	country_coded : Nigeria	19234	12144	322.763	4257.412	30.698	1.400	132476.25	3.967	4.717	5.367	6.050	6.800	7.750	8.967	11.233	19.645	22.668	27.533	35.856	57.394	122.306	265.355	538.912	1101.331	2924.565	132476.25	0.631
misinfo_quiz_duration	country_coded : South Africa	13771	7767	255.112	3681.688	31.374	1.533	133550.80	3.567	4.150	4.633	5.133	5.667	6.333	7.267	8.850	15.133	17.512	20.633	27.000	36.000	58.983	131.363	316.183	710.117	2001.438	133550.80	0.564
mean_share_misinfo	country_coded : Ghana	4922	3215	0.482	0.388	0.006	0.000	1.00	0.000	0.000	0.333	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.653
mean_share_misinfo	country_coded : Kenya	5714	3588	0.507	0.385	0.005	0.000	1.00	0.000	0.000	0.333	0.333	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.628
mean_share_misinfo	country_coded : Nigeria	18626	12752	0.552	0.387	0.003	0.000	1.00	0.000	0.000	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.685
mean_share_misinfo	country_coded : South Africa	13545	7993	0.523	0.373	0.003	0.000	1.00	0.000	0.000	0.333	0.333	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.590
mean_reliable_misinfo	country_coded : Ghana	4922	3215	2.569	1.262	0.018	1.000	5.00	1.000	1.333	1.667	2.000	2.333	2.667	3.333	3.667	4.667	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.653
mean_reliable_misinfo	country_coded : Kenya	5714	3588	2.432	1.217	0.016	1.000	5.00	1.000	1.333	1.667	2.000	2.333	2.333	3.000	3.667	4.333	4.667	4.667	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.628
mean_reliable_misinfo	country_coded : Nigeria	18626	12752	2.622	1.275	0.009	1.000	5.00	1.000	1.333	1.667	2.000	2.333	2.667	3.333	3.667	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.685
mean_reliable_misinfo	country_coded : South Africa	13545	7993	2.635	1.201	0.010	1.000	5.00	1.000	1.333	1.667	2.333	2.333	3.000	3.333	3.667	4.333	4.667	4.667	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.590
mean_manipulative_misinfo	country_coded : Ghana	4922	3215	3.240	1.264	0.018	1.000	5.00	1.333	2.000	2.333	3.000	3.333	3.667	4.000	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.653
mean_manipulative_misinfo	country_coded : Kenya	5714	3588	3.069	1.319	0.017	1.000	5.00	1.333	1.667	2.000	2.400	3.000	3.667	4.000	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.628
mean_manipulative_misinfo	country_coded : Nigeria	18626	12752	2.899	1.302	0.010	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.333	3.667	4.333	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.685
mean_manipulative_misinfo	country_coded : South Africa	13545	7993	3.142	1.253	0.011	1.000	5.00	1.333	2.000	2.333	2.667	3.333	3.667	4.000	4.333	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.590
mean_share_nonmisinfo	country_coded : Ghana	4922	3215	0.713	0.357	0.005	0.000	1.00	0.000	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.653
mean_share_nonmisinfo	country_coded : Kenya	5714	3588	0.711	0.352	0.005	0.000	1.00	0.000	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.628
mean_share_nonmisinfo	country_coded : Nigeria	18626	12752	0.740	0.350	0.003	0.000	1.00	0.000	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.685
mean_share_nonmisinfo	country_coded : South Africa	13545	7993	0.697	0.358	0.003	0.000	1.00	0.000	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.590
mean_reliable_nonmisinfo	country_coded : Ghana	4922	3215	3.253	1.324	0.019	1.000	5.00	1.500	2.000	2.500	3.000	3.000	3.500	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.653
mean_reliable_nonmisinfo	country_coded : Kenya	5714	3588	2.967	1.351	0.018	1.000	5.00	1.000	1.500	2.000	2.500	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.628
mean_reliable_nonmisinfo	country_coded : Nigeria	18626	12752	3.140	1.361	0.010	1.000	5.00	1.000	1.500	2.000	3.000	3.000	3.500	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.685
mean_reliable_nonmisinfo	country_coded : South Africa	13545	7993	3.068	1.285	0.011	1.000	5.00	1.500	2.000	2.000	3.000	3.000	3.500	4.000	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.590
mean_manipulative_nonmisinfo	country_coded : Ghana	4922	3215	2.834	1.376	0.020	1.000	5.00	1.000	1.500	2.000	2.500	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.653
mean_manipulative_nonmisinfo	country_coded : Kenya	5714	3588	2.655	1.346	0.018	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.628
mean_manipulative_nonmisinfo	country_coded : Nigeria	18626	12752	2.456	1.336	0.010	1.000	5.00	1.000	1.000	1.500	1.500	2.000	3.000	3.000	3.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.685
mean_manipulative_nonmisinfo	country_coded : South Africa	13545	7993	2.522	1.290	0.011	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	3.500	4.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.590
sharing_discernment	country_coded : Ghana	4922	3215	0.231	0.409	0.006	-1.000	1.00	-0.167	0.000	0.000	0.000	0.167	0.333	0.500	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.653
sharing_discernment	country_coded : Kenya	5714	3588	0.203	0.404	0.005	-1.000	1.00	-0.167	0.000	0.000	0.000	0.167	0.333	0.333	0.500	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.628
sharing_discernment	country_coded : Nigeria	18626	12752	0.189	0.393	0.003	-1.000	1.00	-0.167	0.000	0.000	0.000	0.000	0.333	0.333	0.500	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.685
sharing_discernment	country_coded : South Africa	13545	7993	0.174	0.418	0.004	-1.000	1.00	-0.333	-0.167	0.000	0.000	0.167	0.333	0.333	0.500	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.590
reliability_discernment	country_coded : Ghana	4922	3215	0.683	1.353	0.019	-4.000	4.00	-0.833	-0.333	0.000	0.167	0.500	1.000	1.333	1.833	2.667	2.667	2.667	2.667	3.000	3.000	3.167	3.500	4.000	4.000	4.00	0.653
reliability_discernment	country_coded : Kenya	5714	3588	0.535	1.327	0.018	-4.000	4.00	-0.833	-0.333	0.000	0.000	0.333	0.667	1.000	1.667	2.333	2.500	2.667	2.667	2.667	3.000	3.000	3.500	4.000	4.000	4.00	0.628
reliability_discernment	country_coded : Nigeria	18626	12752	0.518	1.307	0.010	-4.000	4.00	-1.000	-0.333	0.000	0.000	0.333	0.667	1.000	1.500	2.333	2.500	2.500	2.667	2.667	2.667	3.000	3.333	3.667	4.000	4.00	0.685
reliability_discernment	country_coded : South Africa	13545	7993	0.433	1.352	0.012	-4.000	4.00	-1.167	-0.667	-0.167	0.000	0.333	0.667	1.000	1.500	2.167	2.333	2.500	2.667	2.667	2.667	3.000	3.167	3.500	4.000	4.00	0.590
manipulation_discernment	country_coded : Ghana	4922	3215	0.406	1.447	0.021	-4.000	4.00	-1.333	-0.667	-0.167	0.000	0.167	0.667	1.000	1.667	2.333	2.500	2.667	2.667	2.667	3.000	3.167	3.500	3.667	4.000	4.00	0.653
manipulation_discernment	country_coded : Kenya	5714	3588	0.414	1.360	0.018	-4.000	4.00	-1.167	-0.500	-0.167	0.000	0.167	0.500	1.000	1.500	2.167	2.333	2.500	2.667	2.667	2.725	3.000	3.333	3.667	4.000	4.00	0.628
manipulation_discernment	country_coded : Nigeria	18626	12752	0.443	1.369	0.010	-4.000	4.00	-1.167	-0.500	0.000	0.000	0.167	0.667	1.000	1.500	2.333	2.500	2.667	2.667	2.667	2.833	3.000	3.333	3.667	4.000	4.00	0.685
manipulation_discernment	country_coded : South Africa	13545	7993	0.619	1.397	0.012	-4.000	4.00	-1.000	-0.500	0.000	0.167	0.500	0.833	1.333	1.833	2.667	2.667	2.667	2.667	3.000	3.000	3.333	3.500	3.667	4.000	4.00	0.590

User level variables by arm and phase

df_wide = df_wide |> mutate(phase_arm = paste(phase_coded, arm_coded, sep = " - "))

summary_stats_table_continous(df_wide,
  variables = continous_var,
  group = "phase_arm",
  filename = "continous_stat_results_consent_by_phase_arm.csv",
  caption = "Summary Statistics for Continous Variables, by phase and arm",
  print_html = TRUE
  )

Summary Statistics for Continous Variables, by phase and arm

Variable	Group	Total_Observations	Number_Missing	Mean	SD	SE	Min	Max	Decile_10	Decile_20	Decile_30	Decile_40	Decile_50	Decile_60	Decile_70	Decile_80	Decile_90	Decile_91	Decile_92	Decile_93	Decile_94	Decile_95	Decile_96	Decile_97	Decile_98	Decile_99	Decile_100	Proportion_Missing
intervention_duration	phase_arm : Phase 1 - Game	4634	2782	308.098	3392.556	49.837	5.767	128975.58	7.067	8.183	10.250	13.170	17.383	22.610	29.637	43.373	126.495	164.525	219.512	292.338	455.129	609.729	862.239	1209.109	1771.018	4360.392	128975.58	0.600
intervention_duration	phase_arm : Phase 1 - Original Baseline	6066	1303	205.088	2607.973	33.485	4.617	93628.07	7.150	8.067	8.858	9.733	10.792	12.067	13.900	16.900	27.942	31.905	37.357	46.204	66.938	107.146	222.417	451.298	827.512	2461.138	93628.07	0.215
intervention_duration	phase_arm : Phase 1 - SMS	5078	2392	363.503	4232.636	59.397	8.150	125646.87	12.300	13.600	14.750	15.930	17.483	19.783	22.982	29.880	81.493	107.532	142.082	222.652	313.132	478.135	655.457	989.990	1555.195	4781.591	125646.87	0.471
intervention_duration	phase_arm : Phase 1 - Video	4916	2556	412.666	3992.914	56.949	9.000	106594.78	10.967	11.933	12.850	13.983	15.533	18.117	23.033	39.917	248.008	311.084	425.330	521.353	657.562	853.650	1197.477	1576.660	2577.520	5481.641	106594.78	0.520
intervention_duration	phase_arm : Phase 2 - Long Baseline	10733	3405	234.160	2572.233	24.828	3.967	83247.23	12.867	13.933	15.000	16.150	17.533	19.433	22.323	27.693	52.707	64.659	82.045	119.439	181.778	302.593	495.293	819.585	1373.283	3315.505	83247.23	0.317
intervention_duration	phase_arm : Phase 2 - Original Baseline	5142	1096	149.528	1850.542	25.807	4.217	73585.40	7.000	7.867	8.633	9.540	10.550	11.667	13.367	15.983	24.548	26.749	29.723	34.336	43.253	57.961	110.519	241.675	601.654	2015.385	73585.40	0.213
intervention_duration	phase_arm : Phase 2 - SMS	15499	4753	300.444	3343.231	26.854	5.683	94308.27	12.300	13.283	14.250	15.300	16.667	18.450	21.267	26.623	56.580	69.512	91.711	131.326	193.683	310.528	498.529	783.105	1446.189	3679.220	94308.27	0.307
total_duration	phase_arm : Phase 1 - Game	3481	3935	764.863	6309.786	106.946	8.150	132509.32	14.433	17.683	21.100	24.917	30.067	36.783	48.117	75.733	398.183	518.893	669.323	842.643	1126.410	1417.883	2095.240	3206.360	5220.943	13115.590	132509.32	1.130
total_duration	phase_arm : Phase 1 - Original Baseline	4881	2488	618.807	6041.494	86.475	7.133	133564.78	13.667	15.733	17.467	19.550	21.867	25.000	29.400	40.433	175.950	234.607	330.917	431.727	551.953	723.367	963.963	1445.330	2696.797	7907.067	133564.78	0.510
total_duration	phase_arm : Phase 1 - SMS	4320	3150	592.571	5788.243	88.065	11.733	125650.88	19.367	21.833	23.950	26.343	29.442	33.727	40.838	59.017	282.873	361.074	487.053	589.122	754.562	977.183	1301.407	1862.905	3698.296	8035.663	125650.88	0.729
total_duration	phase_arm : Phase 1 - Video	4136	3336	798.080	6477.553	100.721	11.450	135561.02	17.717	19.883	21.767	23.833	26.733	30.950	39.675	83.050	529.158	633.308	772.633	977.788	1259.270	1506.842	1937.587	2729.385	4744.033	12510.097	135561.02	0.807
total_duration	phase_arm : Phase 2 - Long Baseline	9173	4965	557.098	4724.967	49.334	12.033	98018.13	18.983	21.000	22.993	25.113	27.633	31.303	36.950	49.950	198.173	279.461	378.037	516.126	681.453	927.490	1278.223	1798.385	3765.373	10519.382	98018.13	0.541
total_duration	phase_arm : Phase 2 - Original Baseline	4271	1967	544.160	4711.784	72.098	6.850	92191.73	13.100	14.967	16.683	18.433	20.450	23.050	26.833	34.400	103.700	148.847	201.103	300.078	485.053	690.425	1003.283	1505.658	2863.767	12893.532	92191.73	0.461
total_duration	phase_arm : Phase 2 - SMS	13716	6536	658.229	5364.721	45.807	12.217	97132.32	18.650	20.600	22.450	24.583	27.150	30.600	36.183	50.250	214.333	285.341	381.310	509.023	668.212	955.758	1350.100	1988.740	4024.917	17180.468	97132.32	0.477
misinfo_quiz_duration	phase_arm : Phase 1 - Game	3481	3935	436.080	5499.362	93.210	1.517	132476.25	3.550	4.233	4.800	5.350	6.000	6.767	7.850	9.750	16.083	19.490	23.417	33.820	53.370	106.617	237.983	549.193	1425.103	5131.283	132476.25	1.130
misinfo_quiz_duration	phase_arm : Phase 1 - Original Baseline	4881	2488	356.181	4807.964	68.819	1.550	133550.80	4.133	4.917	5.583	6.250	7.017	7.933	9.267	11.633	20.950	25.073	31.637	44.660	81.580	161.400	323.090	535.070	884.727	2795.163	133550.80	0.510
misinfo_quiz_duration	phase_arm : Phase 1 - SMS	4319	3151	277.739	4273.269	65.023	1.867	121920.62	4.380	5.200	5.907	6.567	7.350	8.297	9.777	12.150	20.877	23.365	28.200	35.941	50.243	77.693	180.608	388.771	855.293	2786.456	121920.62	0.730
misinfo_quiz_duration	phase_arm : Phase 1 - Video	4136	3336	384.948	4849.992	75.414	1.617	120152.10	3.983	4.717	5.350	6.033	6.767	7.600	8.700	10.767	20.450	24.064	29.397	40.720	60.685	162.488	301.123	594.364	1164.892	3107.946	120152.10	0.807
misinfo_quiz_duration	phase_arm : Phase 2 - Long Baseline	9172	4966	230.460	3179.488	33.199	1.400	93547.08	3.600	4.250	4.788	5.350	5.958	6.717	7.817	9.633	16.032	17.683	20.683	26.104	35.168	60.845	132.783	360.972	731.122	1950.640	93547.08	0.541
misinfo_quiz_duration	phase_arm : Phase 2 - Original Baseline	4270	1968	315.646	4024.348	61.586	1.333	92066.67	3.617	4.300	4.867	5.383	6.033	6.733	7.817	9.717	15.502	18.016	20.332	24.620	31.467	57.041	121.501	346.172	737.365	2400.820	92066.67	0.461
misinfo_quiz_duration	phase_arm : Phase 2 - SMS	13714	6538	267.534	3738.859	31.927	1.200	97110.57	3.867	4.533	5.100	5.667	6.333	7.163	8.267	10.140	16.133	17.917	20.916	25.650	32.498	50.642	110.711	296.492	722.232	1944.756	97110.57	0.477
mean_share_misinfo	phase_arm : Phase 1 - Game	3242	4174	0.730	0.316	0.006	0.000	1.00	0.333	0.333	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	1.287
mean_share_misinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	0.618	0.367	0.005	0.000	1.00	0.000	0.333	0.333	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.612
mean_share_misinfo	phase_arm : Phase 1 - SMS	4026	3444	0.352	0.346	0.005	0.000	1.00	0.000	0.000	0.000	0.333	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.855
mean_share_misinfo	phase_arm : Phase 1 - Video	3854	3618	0.651	0.366	0.006	0.000	1.00	0.000	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.939
mean_share_misinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	0.627	0.371	0.004	0.000	1.00	0.000	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.543
mean_share_misinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	0.624	0.367	0.006	0.000	1.00	0.000	0.333	0.333	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.463
mean_share_misinfo	phase_arm : Phase 2 - SMS	13688	6564	0.374	0.352	0.003	0.000	1.00	0.000	0.000	0.000	0.333	0.333	0.333	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.480
mean_reliable_misinfo	phase_arm : Phase 1 - Game	3242	4174	2.908	1.257	0.022	1.000	5.00	1.333	1.667	2.000	2.333	2.667	3.333	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	1.287
mean_reliable_misinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	2.837	1.284	0.019	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.333	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.612
mean_reliable_misinfo	phase_arm : Phase 1 - SMS	4026	3444	2.188	1.118	0.018	1.000	5.00	1.000	1.000	1.333	1.667	2.000	2.333	2.667	3.333	3.667	4.000	4.000	4.000	4.333	4.667	4.667	5.000	5.000	5.000	5.00	0.855
mean_reliable_misinfo	phase_arm : Phase 1 - Video	3854	3618	2.794	1.279	0.021	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.000	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.939
mean_reliable_misinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	2.865	1.265	0.013	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.333	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.543
mean_reliable_misinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	2.799	1.261	0.019	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.000	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.463
mean_reliable_misinfo	phase_arm : Phase 2 - SMS	13688	6564	2.258	1.116	0.010	1.000	5.00	1.000	1.000	1.333	1.667	2.000	2.333	2.667	3.333	3.667	4.000	4.000	4.000	4.333	4.667	5.000	5.000	5.000	5.000	5.00	0.480
mean_manipulative_misinfo	phase_arm : Phase 1 - Game	3242	4174	2.788	1.264	0.022	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.000	3.667	4.000	4.667	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	1.287
mean_manipulative_misinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	2.942	1.290	0.019	1.000	5.00	1.000	1.667	2.000	2.333	3.000	3.333	3.667	4.333	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.612
mean_manipulative_misinfo	phase_arm : Phase 1 - SMS	4026	3444	3.154	1.313	0.021	1.000	5.00	1.333	1.667	2.333	2.667	3.333	3.667	4.000	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.855
mean_manipulative_misinfo	phase_arm : Phase 1 - Video	3854	3618	2.861	1.296	0.021	1.000	5.00	1.000	1.667	2.000	2.333	2.667	3.333	3.667	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.939
mean_manipulative_misinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	3.021	1.282	0.013	1.000	5.00	1.000	1.667	2.333	2.667	3.000	3.667	3.667	4.333	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.543
mean_manipulative_misinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	2.974	1.292	0.020	1.000	5.00	1.000	1.667	2.000	2.333	3.000	3.400	3.667	4.333	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.463
mean_manipulative_misinfo	phase_arm : Phase 2 - SMS	13688	6564	3.176	1.276	0.011	1.000	5.00	1.333	2.000	2.333	2.667	3.333	3.667	4.000	4.667	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.480
mean_share_nonmisinfo	phase_arm : Phase 1 - Game	3242	4174	0.797	0.309	0.005	0.000	1.00	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	1.287
mean_share_nonmisinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	0.762	0.338	0.005	0.000	1.00	0.000	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.612
mean_share_nonmisinfo	phase_arm : Phase 1 - SMS	4026	3444	0.634	0.375	0.006	0.000	1.00	0.000	0.500	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.855
mean_share_nonmisinfo	phase_arm : Phase 1 - Video	3854	3618	0.804	0.314	0.005	0.000	1.00	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.939
mean_share_nonmisinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	0.769	0.333	0.003	0.000	1.00	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.543
mean_share_nonmisinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	0.756	0.342	0.005	0.000	1.00	0.000	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.463
mean_share_nonmisinfo	phase_arm : Phase 2 - SMS	13688	6564	0.644	0.371	0.003	0.000	1.00	0.000	0.500	0.500	0.500	0.500	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.480
mean_reliable_nonmisinfo	phase_arm : Phase 1 - Game	3242	4174	3.110	1.334	0.023	1.000	5.00	1.500	1.500	2.000	2.500	3.000	3.500	4.000	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	1.287
mean_reliable_nonmisinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	3.246	1.354	0.020	1.000	5.00	1.500	2.000	2.500	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.612
mean_reliable_nonmisinfo	phase_arm : Phase 1 - SMS	4026	3444	2.914	1.324	0.021	1.000	5.00	1.000	1.500	2.000	2.500	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.855
mean_reliable_nonmisinfo	phase_arm : Phase 1 - Video	3854	3618	3.230	1.357	0.022	1.000	5.00	1.500	2.000	2.000	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.939
mean_reliable_nonmisinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	3.285	1.320	0.014	1.000	5.00	1.500	2.000	2.500	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.543
mean_reliable_nonmisinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	3.190	1.358	0.021	1.000	5.00	1.000	2.000	2.000	3.000	3.000	3.500	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.463
mean_reliable_nonmisinfo	phase_arm : Phase 2 - SMS	13688	6564	2.938	1.297	0.011	1.000	5.00	1.000	1.500	2.000	2.500	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.480
mean_manipulative_nonmisinfo	phase_arm : Phase 1 - Game	3242	4174	2.498	1.320	0.023	1.000	5.00	1.000	1.000	1.500	2.000	2.000	3.000	3.000	4.000	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	1.287
mean_manipulative_nonmisinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	2.589	1.380	0.020	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.612
mean_manipulative_nonmisinfo	phase_arm : Phase 1 - SMS	4026	3444	2.497	1.326	0.021	1.000	5.00	1.000	1.000	1.500	2.000	2.000	3.000	3.000	3.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.855
mean_manipulative_nonmisinfo	phase_arm : Phase 1 - Video	3854	3618	2.445	1.340	0.022	1.000	5.00	1.000	1.000	1.500	1.500	2.000	3.000	3.000	3.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.939
mean_manipulative_nonmisinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	2.620	1.358	0.014	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.543
mean_manipulative_nonmisinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	2.556	1.343	0.021	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	4.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.463
mean_manipulative_nonmisinfo	phase_arm : Phase 2 - SMS	13688	6564	2.537	1.298	0.011	1.000	5.00	1.000	1.000	1.500	2.000	2.500	3.000	3.000	3.500	4.500	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.000	5.00	0.480
sharing_discernment	phase_arm : Phase 1 - Game	3242	4174	0.067	0.366	0.006	-1.000	1.00	-0.500	-0.167	0.000	0.000	0.000	0.000	0.333	0.333	0.667	0.667	0.667	0.667	0.667	0.667	0.667	0.667	1.000	1.000	1.00	1.287
sharing_discernment	phase_arm : Phase 1 - Original Baseline	4571	2798	0.143	0.391	0.006	-1.000	1.00	-0.333	0.000	0.000	0.000	0.000	0.167	0.333	0.500	0.667	0.667	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.00	0.612
sharing_discernment	phase_arm : Phase 1 - SMS	4026	3444	0.283	0.422	0.007	-1.000	1.00	-0.167	0.000	0.000	0.167	0.333	0.500	0.500	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.855
sharing_discernment	phase_arm : Phase 1 - Video	3854	3618	0.153	0.374	0.006	-1.000	1.00	-0.167	0.000	0.000	0.000	0.000	0.167	0.333	0.500	0.667	0.667	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.00	0.939
sharing_discernment	phase_arm : Phase 2 - Long Baseline	9163	4975	0.142	0.382	0.004	-1.000	1.00	-0.333	0.000	0.000	0.000	0.000	0.167	0.333	0.500	0.667	0.667	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.000	1.00	0.543
sharing_discernment	phase_arm : Phase 2 - Original Baseline	4263	1975	0.133	0.380	0.006	-1.000	1.00	-0.333	0.000	0.000	0.000	0.000	0.167	0.333	0.500	0.667	0.667	0.667	0.667	0.667	0.667	1.000	1.000	1.000	1.000	1.00	0.463
sharing_discernment	phase_arm : Phase 2 - SMS	13688	6564	0.270	0.423	0.004	-1.000	1.00	-0.167	0.000	0.000	0.167	0.333	0.333	0.500	0.667	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.000	1.00	0.480
reliability_discernment	phase_arm : Phase 1 - Game	3242	4174	0.202	1.178	0.021	-4.000	4.00	-1.167	-0.667	-0.333	0.000	0.000	0.333	0.667	1.000	1.667	1.833	2.000	2.000	2.167	2.333	2.500	2.667	3.000	3.500	4.00	1.287
reliability_discernment	phase_arm : Phase 1 - Original Baseline	4571	2798	0.409	1.312	0.019	-4.000	4.00	-1.167	-0.500	-0.167	0.000	0.167	0.500	1.000	1.333	2.167	2.333	2.500	2.667	2.667	2.667	3.000	3.000	3.500	4.000	4.00	0.612
reliability_discernment	phase_arm : Phase 1 - SMS	4026	3444	0.726	1.386	0.022	-4.000	4.00	-0.833	-0.333	0.000	0.167	0.500	1.000	1.333	2.000	2.667	2.667	2.667	2.833	3.000	3.167	3.333	3.667	4.000	4.000	4.00	0.855
reliability_discernment	phase_arm : Phase 1 - Video	3854	3618	0.435	1.249	0.020	-4.000	4.00	-1.000	-0.500	0.000	0.000	0.333	0.500	1.000	1.333	2.167	2.167	2.333	2.500	2.667	2.667	2.667	3.000	3.333	3.667	4.00	0.939
reliability_discernment	phase_arm : Phase 2 - Long Baseline	9163	4975	0.420	1.325	0.014	-4.000	4.00	-1.167	-0.500	-0.167	0.000	0.167	0.667	1.000	1.500	2.167	2.333	2.500	2.667	2.667	2.667	2.833	3.000	3.500	4.000	4.00	0.543
reliability_discernment	phase_arm : Phase 2 - Original Baseline	4263	1975	0.391	1.292	0.020	-4.000	4.00	-1.167	-0.500	-0.167	0.000	0.167	0.500	1.000	1.333	2.167	2.167	2.333	2.500	2.667	2.667	2.667	3.000	3.333	4.000	4.00	0.463
reliability_discernment	phase_arm : Phase 2 - SMS	13688	6564	0.680	1.364	0.012	-4.000	4.00	-0.833	-0.333	0.000	0.167	0.500	0.833	1.333	2.000	2.500	2.667	2.667	2.667	3.000	3.000	3.333	3.667	4.000	4.000	4.00	0.480
manipulation_discernment	phase_arm : Phase 1 - Game	3242	4174	0.290	1.238	0.022	-4.000	4.00	-1.167	-0.500	-0.167	0.000	0.167	0.500	0.667	1.167	2.000	2.000	2.000	2.167	2.333	2.667	2.667	2.667	3.167	3.667	4.00	1.287
manipulation_discernment	phase_arm : Phase 1 - Original Baseline	4571	2798	0.353	1.362	0.020	-4.000	4.00	-1.333	-0.667	-0.167	0.000	0.167	0.500	0.833	1.333	2.167	2.333	2.500	2.667	2.667	2.667	3.000	3.333	3.667	4.000	4.00	0.612
manipulation_discernment	phase_arm : Phase 1 - SMS	4026	3444	0.657	1.471	0.023	-4.000	4.00	-1.000	-0.333	0.000	0.000	0.500	0.833	1.333	2.000	2.667	2.667	3.000	3.000	3.000	3.333	3.500	3.667	4.000	4.000	4.00	0.855
manipulation_discernment	phase_arm : Phase 1 - Video	3854	3618	0.416	1.316	0.021	-4.000	4.00	-1.167	-0.500	0.000	0.000	0.167	0.500	1.000	1.500	2.167	2.333	2.500	2.667	2.667	2.667	3.000	3.000	3.500	4.000	4.00	0.939
manipulation_discernment	phase_arm : Phase 2 - Long Baseline	9163	4975	0.401	1.351	0.014	-4.000	4.00	-1.333	-0.667	-0.167	0.000	0.167	0.667	1.000	1.500	2.167	2.333	2.333	2.667	2.667	2.667	3.000	3.000	3.500	4.000	4.00	0.543
manipulation_discernment	phase_arm : Phase 2 - Original Baseline	4263	1975	0.418	1.338	0.020	-4.000	4.00	-1.167	-0.500	-0.167	0.000	0.167	0.500	1.000	1.500	2.333	2.333	2.500	2.667	2.667	2.667	3.000	3.167	3.500	4.000	4.00	0.463
manipulation_discernment	phase_arm : Phase 2 - SMS	13688	6564	0.639	1.445	0.012	-4.000	4.00	-1.167	-0.500	0.000	0.167	0.500	0.833	1.333	2.000	2.667	2.667	2.667	2.833	3.000	3.167	3.500	3.667	4.000	4.000	4.00	0.480

Post-level variables

summary_stats_table_continous(
  df_long, 
  variables = c("share_score", "reliable_score", "manipulative_score"),
  filename = "continous_stat_results_consent_post.csv",
  caption = "Summary Statistics for Continous Variables",
  print_html = TRUE
  )

Summary Statistics for Continous Variables

Variable	Total_Observations	Number_Missing	Mean	SD	SE	Min	Max	Decile_10	Decile_20	Decile_30	Decile_40	Decile_50	Decile_60	Decile_70	Decile_80	Decile_90	Decile_91	Decile_92	Decile_93	Decile_94	Decile_95	Decile_96	Decile_97	Decile_98	Decile_99	Decile_100	Proportion_Missing
share_score	214035	0	0.605	0.489	0.001	0	1	0	0	0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	0
reliable_score	214035	0	2.800	1.645	0.004	1	5	1	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0
manipulative_score	214035	0	2.842	1.666	0.004	1	5	1	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0

Post-level variables, by arm

summary_stats_table_continous(
  df_long, 
  variables = c("share_score", "reliable_score", "manipulative_score"),
  group = "arm_coded",
  filename = "continous_stat_results_consent_post_by_arm.csv",
  caption = "Summary Statistics for Continous Variables, by arm",
  print_html = TRUE
  )

Summary Statistics for Continous Variables, by arm

Variable	Group	Total_Observations	Number_Missing	Mean	SD	SE	Min	Max	Decile_10	Decile_20	Decile_30	Decile_40	Decile_50	Decile_60	Decile_70	Decile_80	Decile_90	Decile_91	Decile_92	Decile_93	Decile_94	Decile_95	Decile_96	Decile_97	Decile_98	Decile_99	Decile_100	Proportion_Missing
share_score	arm_coded : Game	16210	0	0.757	0.429	0.003	0	1	0	0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	0
share_score	arm_coded : Long Baseline	45815	0	0.684	0.465	0.002	0	1	0	0	0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	0
share_score	arm_coded : Original Baseline	44170	0	0.676	0.468	0.002	0	1	0	0	0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	0
share_score	arm_coded : SMS	88570	0	0.478	0.500	0.002	0	1	0	0	0	0	0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	0
share_score	arm_coded : Video	19270	0	0.712	0.453	0.003	0	1	0	0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	0
reliable_score	arm_coded : Game	16210	0	2.989	1.612	0.013	1	5	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	5	0
reliable_score	arm_coded : Long Baseline	45815	0	3.033	1.650	0.008	1	5	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	5	0
reliable_score	arm_coded : Original Baseline	44170	0	2.979	1.658	0.008	1	5	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	5	0
reliable_score	arm_coded : SMS	88570	0	2.518	1.601	0.005	1	5	1	1	1	1	2	2	4	5	5	5	5	5	5	5	5	5	5	5	5	0
reliable_score	arm_coded : Video	19270	0	2.969	1.635	0.012	1	5	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	5	0
manipulative_score	arm_coded : Game	16210	0	2.672	1.609	0.013	1	5	1	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0
manipulative_score	arm_coded : Long Baseline	45815	0	2.861	1.670	0.008	1	5	1	1	1	2	3	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0
manipulative_score	arm_coded : Original Baseline	44170	0	2.804	1.669	0.008	1	5	1	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0
manipulative_score	arm_coded : SMS	88570	0	2.914	1.671	0.006	1	5	1	1	1	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	5	0
manipulative_score	arm_coded : Video	19270	0	2.694	1.649	0.012	1	5	1	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0

Summary statistics for unordered categorical variables

User-level variables

unordered_cat <- c("phase_coded", "country_coded", "payment_amount_coded", "manipulative_reliable_order")

summary_stats_tab_categorical(df_wide,  
                              variables = unordered_cat ,
  filename = "unordered_cat.csv",
  caption = "Summary Statistics for Unordered Categorical Variables",
  print_html = TRUE
  )

Summary Statistics for Unordered Categorical Variables

Variable	Group	Category	Frequency	Proportion
phase_coded	NA	Phase 1	29727	0.423
phase_coded	NA	Phase 2	40628	0.577
country_coded	NA	Kenya	9302	0.132
country_coded	NA	Nigeria	31378	0.446
country_coded	NA	South Africa	21538	0.306
country_coded	NA	Ghana	8137	0.116
payment_amount_coded	NA	150Ksh	5138	0.073
payment_amount_coded	NA	₦800	8917	0.127
payment_amount_coded	NA	R20	6330	0.090
payment_amount_coded	NA	₵12.5	5075	0.072
payment_amount_coded	NA	₦400	18155	0.258
payment_amount_coded	NA	R10	12394	0.176
payment_amount_coded	NA	R60	2814	0.040
payment_amount_coded	NA	₦2400	4308	0.061
payment_amount_coded	NA	450Ksh	1221	0.017
payment_amount_coded	NA	₵6	2076	0.030
payment_amount_coded	NA	75Ksh	2941	0.042
payment_amount_coded	NA	₵36	986	0.014
manipulative_reliable_order	NA	NA	18967	0.270
manipulative_reliable_order	NA	1	25738	0.366
manipulative_reliable_order	NA	2	25650	0.365

User-level variables, by arm

summary_stats_tab_categorical(
  df_wide,  
  variables = c("phase_coded", "payment_amount_coded", "manipulative_reliable_order"),
  group = "arm_coded",
  filename = "unordered_cat_arm.csv",
  caption = "Summary Statistics for Unordered Categorical Variables",
  print_html = TRUE
  )

Summary Statistics for Unordered Categorical Variables

Variable	Group	Category	Frequency	Proportion
phase_coded	Game	Phase 1	7416	0.105
phase_coded	Long Baseline	Phase 2	14138	0.201
phase_coded	Original Baseline	Phase 1	7369	0.105
phase_coded	Original Baseline	Phase 2	6238	0.089
phase_coded	SMS	Phase 1	7470	0.106
phase_coded	SMS	Phase 2	20252	0.288
phase_coded	Video	Phase 1	7472	0.106
payment_amount_coded	Game	150Ksh	995	0.014
payment_amount_coded	Game	R10	1398	0.020
payment_amount_coded	Game	R20	828	0.012
payment_amount_coded	Game	₦400	2092	0.030
payment_amount_coded	Game	₦800	1124	0.016
payment_amount_coded	Game	₵12.5	979	0.014
payment_amount_coded	Long Baseline	150Ksh	488	0.007
payment_amount_coded	Long Baseline	450Ksh	427	0.006
payment_amount_coded	Long Baseline	75Ksh	993	0.014
payment_amount_coded	Long Baseline	R10	2313	0.033
payment_amount_coded	Long Baseline	R20	1174	0.017
payment_amount_coded	Long Baseline	R60	933	0.013
payment_amount_coded	Long Baseline	₦2400	1363	0.019
payment_amount_coded	Long Baseline	₦400	3275	0.047
payment_amount_coded	Long Baseline	₦800	1723	0.024
payment_amount_coded	Long Baseline	₵12.5	403	0.006
payment_amount_coded	Long Baseline	₵36	318	0.005
payment_amount_coded	Long Baseline	₵6	728	0.010
payment_amount_coded	Original Baseline	150Ksh	1127	0.016
payment_amount_coded	Original Baseline	450Ksh	198	0.003
payment_amount_coded	Original Baseline	75Ksh	468	0.007
payment_amount_coded	Original Baseline	R10	2515	0.036
payment_amount_coded	Original Baseline	R20	1156	0.016
payment_amount_coded	Original Baseline	R60	514	0.007
payment_amount_coded	Original Baseline	₦2400	727	0.010
payment_amount_coded	Original Baseline	₦400	3756	0.053
payment_amount_coded	Original Baseline	₦800	1536	0.022
payment_amount_coded	Original Baseline	₵12.5	1115	0.016
payment_amount_coded	Original Baseline	₵36	173	0.002
payment_amount_coded	Original Baseline	₵6	322	0.005
payment_amount_coded	SMS	150Ksh	1565	0.022
payment_amount_coded	SMS	450Ksh	596	0.008
payment_amount_coded	SMS	75Ksh	1480	0.021
payment_amount_coded	SMS	R10	4787	0.068
payment_amount_coded	SMS	R20	2343	0.033
payment_amount_coded	SMS	R60	1367	0.019
payment_amount_coded	SMS	₦2400	2218	0.032
payment_amount_coded	SMS	₦400	6931	0.099
payment_amount_coded	SMS	₦800	3404	0.048
payment_amount_coded	SMS	₵12.5	1510	0.021
payment_amount_coded	SMS	₵36	495	0.007
payment_amount_coded	SMS	₵6	1026	0.015
payment_amount_coded	Video	150Ksh	963	0.014
payment_amount_coded	Video	R10	1381	0.020
payment_amount_coded	Video	R20	829	0.012
payment_amount_coded	Video	₦400	2101	0.030
payment_amount_coded	Video	₦800	1130	0.016
payment_amount_coded	Video	₵12.5	1068	0.015
manipulative_reliable_order	Game	1	2285	0.032
manipulative_reliable_order	Game	2	2272	0.032
manipulative_reliable_order	Game	Missing	2859	0.041
manipulative_reliable_order	Long Baseline	1	5322	0.076
manipulative_reliable_order	Long Baseline	2	5262	0.075
manipulative_reliable_order	Long Baseline	Missing	3554	0.051
manipulative_reliable_order	Original Baseline	1	5577	0.079
manipulative_reliable_order	Original Baseline	2	5475	0.078
manipulative_reliable_order	Original Baseline	Missing	2555	0.036
manipulative_reliable_order	SMS	1	10094	0.143
manipulative_reliable_order	SMS	2	10243	0.146
manipulative_reliable_order	SMS	Missing	7385	0.105
manipulative_reliable_order	Video	1	2460	0.035
manipulative_reliable_order	Video	2	2398	0.034
manipulative_reliable_order	Video	Missing	2614	0.037

User-level variables, by phase

summary_stats_tab_categorical(
  df_wide,  
  variables = c("arm_coded", "payment_amount_coded", "manipulative_reliable_order"),
  group = "phase_coded",
  filename = "unordered_cat_phase.csv",
  caption = "Summary Statistics for Unordered Categorical Variables",
  print_html = TRUE
  )

Summary Statistics for Unordered Categorical Variables

Variable	Group	Category	Frequency	Proportion
arm_coded	Phase 1	Game	7416	0.105
arm_coded	Phase 1	Original Baseline	7369	0.105
arm_coded	Phase 1	SMS	7470	0.106
arm_coded	Phase 1	Video	7472	0.106
arm_coded	Phase 2	Long Baseline	14138	0.201
arm_coded	Phase 2	Original Baseline	6238	0.089
arm_coded	Phase 2	SMS	20252	0.288
payment_amount_coded	Phase 1	150Ksh	3872	0.055
payment_amount_coded	Phase 1	R10	5573	0.079
payment_amount_coded	Phase 1	R20	3328	0.047
payment_amount_coded	Phase 1	₦400	8462	0.120
payment_amount_coded	Phase 1	₦800	4465	0.063
payment_amount_coded	Phase 1	₵12.5	4027	0.057
payment_amount_coded	Phase 2	150Ksh	1266	0.018
payment_amount_coded	Phase 2	450Ksh	1221	0.017
payment_amount_coded	Phase 2	75Ksh	2941	0.042
payment_amount_coded	Phase 2	R10	6821	0.097
payment_amount_coded	Phase 2	R20	3002	0.043
payment_amount_coded	Phase 2	R60	2814	0.040
payment_amount_coded	Phase 2	₦2400	4308	0.061
payment_amount_coded	Phase 2	₦400	9693	0.138
payment_amount_coded	Phase 2	₦800	4452	0.063
payment_amount_coded	Phase 2	₵12.5	1048	0.015
payment_amount_coded	Phase 2	₵36	986	0.014
payment_amount_coded	Phase 2	₵6	2076	0.030
manipulative_reliable_order	Phase 1	1	10298	0.146
manipulative_reliable_order	Phase 1	2	10096	0.144
manipulative_reliable_order	Phase 1	Missing	9333	0.133
manipulative_reliable_order	Phase 2	1	15440	0.219
manipulative_reliable_order	Phase 2	2	15554	0.221
manipulative_reliable_order	Phase 2	Missing	9634	0.137

User-level variables, by payment amount

summary_stats_tab_categorical(
  df_wide,  
  variables = c("phase_coded", "arm_coded", "manipulative_reliable_order"),
  group = "payment_amount_coded",
  filename = "unordered_cat_phase.csv",
  caption = "Summary Statistics for Unordered Categorical Variables",
  print_html = TRUE
  )

Summary Statistics for Unordered Categorical Variables

Variable	Group	Category	Frequency	Proportion
phase_coded	150Ksh	Phase 1	3872	0.055
phase_coded	150Ksh	Phase 2	1266	0.018
phase_coded	450Ksh	Phase 2	1221	0.017
phase_coded	75Ksh	Phase 2	2941	0.042
phase_coded	R10	Phase 1	5573	0.079
phase_coded	R10	Phase 2	6821	0.097
phase_coded	R20	Phase 1	3328	0.047
phase_coded	R20	Phase 2	3002	0.043
phase_coded	R60	Phase 2	2814	0.040
phase_coded	₦2400	Phase 2	4308	0.061
phase_coded	₦400	Phase 1	8462	0.120
phase_coded	₦400	Phase 2	9693	0.138
phase_coded	₦800	Phase 1	4465	0.063
phase_coded	₦800	Phase 2	4452	0.063
phase_coded	₵12.5	Phase 1	4027	0.057
phase_coded	₵12.5	Phase 2	1048	0.015
phase_coded	₵36	Phase 2	986	0.014
phase_coded	₵6	Phase 2	2076	0.030
arm_coded	150Ksh	Game	995	0.014
arm_coded	150Ksh	Long Baseline	488	0.007
arm_coded	150Ksh	Original Baseline	1127	0.016
arm_coded	150Ksh	SMS	1565	0.022
arm_coded	150Ksh	Video	963	0.014
arm_coded	450Ksh	Long Baseline	427	0.006
arm_coded	450Ksh	Original Baseline	198	0.003
arm_coded	450Ksh	SMS	596	0.008
arm_coded	75Ksh	Long Baseline	993	0.014
arm_coded	75Ksh	Original Baseline	468	0.007
arm_coded	75Ksh	SMS	1480	0.021
arm_coded	R10	Game	1398	0.020
arm_coded	R10	Long Baseline	2313	0.033
arm_coded	R10	Original Baseline	2515	0.036
arm_coded	R10	SMS	4787	0.068
arm_coded	R10	Video	1381	0.020
arm_coded	R20	Game	828	0.012
arm_coded	R20	Long Baseline	1174	0.017
arm_coded	R20	Original Baseline	1156	0.016
arm_coded	R20	SMS	2343	0.033
arm_coded	R20	Video	829	0.012
arm_coded	R60	Long Baseline	933	0.013
arm_coded	R60	Original Baseline	514	0.007
arm_coded	R60	SMS	1367	0.019
arm_coded	₦2400	Long Baseline	1363	0.019
arm_coded	₦2400	Original Baseline	727	0.010
arm_coded	₦2400	SMS	2218	0.032
arm_coded	₦400	Game	2092	0.030
arm_coded	₦400	Long Baseline	3275	0.047
arm_coded	₦400	Original Baseline	3756	0.053
arm_coded	₦400	SMS	6931	0.099
arm_coded	₦400	Video	2101	0.030
arm_coded	₦800	Game	1124	0.016
arm_coded	₦800	Long Baseline	1723	0.024
arm_coded	₦800	Original Baseline	1536	0.022
arm_coded	₦800	SMS	3404	0.048
arm_coded	₦800	Video	1130	0.016
arm_coded	₵12.5	Game	979	0.014
arm_coded	₵12.5	Long Baseline	403	0.006
arm_coded	₵12.5	Original Baseline	1115	0.016
arm_coded	₵12.5	SMS	1510	0.021
arm_coded	₵12.5	Video	1068	0.015
arm_coded	₵36	Long Baseline	318	0.005
arm_coded	₵36	Original Baseline	173	0.002
arm_coded	₵36	SMS	495	0.007
arm_coded	₵6	Long Baseline	728	0.010
arm_coded	₵6	Original Baseline	322	0.005
arm_coded	₵6	SMS	1026	0.015
manipulative_reliable_order	150Ksh	1	1749	0.025
manipulative_reliable_order	150Ksh	2	1836	0.026
manipulative_reliable_order	150Ksh	Missing	1553	0.022
manipulative_reliable_order	450Ksh	1	517	0.007
manipulative_reliable_order	450Ksh	2	460	0.007
manipulative_reliable_order	450Ksh	Missing	244	0.003
manipulative_reliable_order	75Ksh	1	1110	0.016
manipulative_reliable_order	75Ksh	2	1090	0.015
manipulative_reliable_order	75Ksh	Missing	741	0.011
manipulative_reliable_order	R10	1	4279	0.061
manipulative_reliable_order	R10	2	4249	0.060
manipulative_reliable_order	R10	Missing	3866	0.055
manipulative_reliable_order	R20	1	2362	0.034
manipulative_reliable_order	R20	2	2441	0.035
manipulative_reliable_order	R20	Missing	1527	0.022
manipulative_reliable_order	R60	1	1208	0.017
manipulative_reliable_order	R60	2	1089	0.015
manipulative_reliable_order	R60	Missing	517	0.007
manipulative_reliable_order	₦2400	1	1809	0.026
manipulative_reliable_order	₦2400	2	1609	0.023
manipulative_reliable_order	₦2400	Missing	890	0.013
manipulative_reliable_order	₦400	1	6460	0.092
manipulative_reliable_order	₦400	2	6470	0.092
manipulative_reliable_order	₦400	Missing	5225	0.074
manipulative_reliable_order	₦800	1	3293	0.047
manipulative_reliable_order	₦800	2	3443	0.049
manipulative_reliable_order	₦800	Missing	2181	0.031
manipulative_reliable_order	₵12.5	1	1734	0.025
manipulative_reliable_order	₵12.5	2	1819	0.026
manipulative_reliable_order	₵12.5	Missing	1522	0.022
manipulative_reliable_order	₵36	1	405	0.006
manipulative_reliable_order	₵36	2	396	0.006
manipulative_reliable_order	₵36	Missing	185	0.003
manipulative_reliable_order	₵6	1	812	0.012
manipulative_reliable_order	₵6	2	748	0.011
manipulative_reliable_order	₵6	Missing	516	0.007

User-level variables, by attention check passed

summary_stats_tab_categorical(
  df_wide,  
  variables = c("phase_coded", "arm_coded", "manipulative_reliable_order"),
  group = "attention_check_passed",
  filename = "unordered_cat_phase.csv",
  caption = "Summary Statistics for Unordered Categorical Variables",
  print_html = TRUE
  )

Summary Statistics for Unordered Categorical Variables

Variable	Group	Category	Frequency	Proportion
phase_coded	0	Phase 1	15241	0.217
phase_coded	0	Phase 2	4609	0.066
phase_coded	1	Phase 1	443	0.006
phase_coded	1	Phase 2	22503	0.320
phase_coded	Missing	Phase 1	14043	0.200
phase_coded	Missing	Phase 2	13516	0.192
arm_coded	0	Game	3124	0.044
arm_coded	0	Long Baseline	1426	0.020
arm_coded	0	Original Baseline	5074	0.072
arm_coded	0	SMS	6479	0.092
arm_coded	0	Video	3747	0.053
arm_coded	1	Game	115	0.002
arm_coded	1	Long Baseline	7736	0.110
arm_coded	1	Original Baseline	3758	0.053
arm_coded	1	SMS	11232	0.160
arm_coded	1	Video	105	0.001
arm_coded	Missing	Game	4177	0.059
arm_coded	Missing	Long Baseline	4976	0.071
arm_coded	Missing	Original Baseline	4775	0.068
arm_coded	Missing	SMS	10011	0.142
arm_coded	Missing	Video	3620	0.051
manipulative_reliable_order	0	1	10000	0.142
manipulative_reliable_order	0	2	9850	0.140
manipulative_reliable_order	1	1	11444	0.163
manipulative_reliable_order	1	2	11502	0.163
manipulative_reliable_order	Missing	1	4294	0.061
manipulative_reliable_order	Missing	2	4298	0.061
manipulative_reliable_order	Missing	Missing	18967	0.270

Summary Statistics for ordered categorical variables

ordered_cat <- c("manipulative_score", "reliable_score", "share_score")
summary_stats_table_continous(df_long,
  variables = ordered_cat,
  group = ,
  filename = "ordered_cat.csv",
  caption = "Summary Statistics for Ordered Categorical Variables",
  print_html = TRUE
  )

Summary Statistics for Ordered Categorical Variables

Variable	Total_Observations	Number_Missing	Mean	SD	SE	Min	Max	Decile_10	Decile_20	Decile_30	Decile_40	Decile_50	Decile_60	Decile_70	Decile_80	Decile_90	Decile_91	Decile_92	Decile_93	Decile_94	Decile_95	Decile_96	Decile_97	Decile_98	Decile_99	Decile_100	Proportion_Missing
manipulative_score	214035	0	2.842	1.666	0.004	1	5	1	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0
reliable_score	214035	0	2.800	1.645	0.004	1	5	1	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0
share_score	214035	0	0.605	0.489	0.001	0	1	0	0	0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	0

summary_stats_tab_categorical(
  df_long,  
  variables = ordered_cat,
  group = ,
  filename = "ordered_cat.csv",
  caption = "Summary Statistics for Ordered Categorical Variables",
  print_html = TRUE
  )

Summary Statistics for Ordered Categorical Variables

Variable	Group	Category	Frequency	Proportion
manipulative_score	NA	1	72091	0.337
manipulative_score	NA	2	37038	0.173
manipulative_score	NA	3	20996	0.098
manipulative_score	NA	4	20518	0.096
manipulative_score	NA	5	63392	0.296
reliable_score	NA	1	71200	0.333
reliable_score	NA	2	41168	0.192
reliable_score	NA	3	21461	0.100
reliable_score	NA	4	19751	0.092
reliable_score	NA	5	60455	0.282
share_score	NA	0	84519	0.395
share_score	NA	1	129516	0.605

by phase

ordered_cat <- c("manipulative_score", "reliable_score", "share_score")
summary_stats_table_continous(df_long,
  variables = ordered_cat,
  group = "phase_coded" ,
  filename = "ordered_cat.csv",
  caption = "Summary Statistics for Ordered Categorical Variables",
  print_html = TRUE
  )

Summary Statistics for Ordered Categorical Variables

Variable	Group	Total_Observations	Number_Missing	Mean	SD	SE	Min	Max	Decile_10	Decile_20	Decile_30	Decile_40	Decile_50	Decile_60	Decile_70	Decile_80	Decile_90	Decile_91	Decile_92	Decile_93	Decile_94	Decile_95	Decile_96	Decile_97	Decile_98	Decile_99	Decile_100	Proportion_Missing
manipulative_score	phase_coded : Phase 1	78465	0	2.771	1.661	0.006	1	5	1	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0
manipulative_score	phase_coded : Phase 2	135570	0	2.882	1.668	0.005	1	5	1	1	1	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	5	0
reliable_score	phase_coded : Phase 1	78465	0	2.856	1.647	0.006	1	5	1	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0
reliable_score	phase_coded : Phase 2	135570	0	2.767	1.643	0.004	1	5	1	1	1	2	2	3	4	5	5	5	5	5	5	5	5	5	5	5	5	0
share_score	phase_coded : Phase 1	78465	0	0.647	0.478	0.002	0	1	0	0	0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	0
share_score	phase_coded : Phase 2	135570	0	0.581	0.493	0.001	0	1	0	0	0	0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	0

summary_stats_tab_categorical(
  df_long,  
  variables = ordered_cat,
  group = "phase_coded",
  filename = "ordered_cat.csv",
  caption = "Summary Statistics for Ordered Categorical Variables",
  print_html = TRUE
  )

Summary Statistics for Ordered Categorical Variables

Variable	Group	Category	Frequency	Proportion
manipulative_score	Phase 1	1	27679	0.129
manipulative_score	Phase 1	2	14084	0.066
manipulative_score	Phase 1	3	7369	0.034
manipulative_score	Phase 1	4	7179	0.034
manipulative_score	Phase 1	5	22154	0.104
manipulative_score	Phase 2	1	44412	0.207
manipulative_score	Phase 2	2	22954	0.107
manipulative_score	Phase 2	3	13627	0.064
manipulative_score	Phase 2	4	13339	0.062
manipulative_score	Phase 2	5	41238	0.193
reliable_score	Phase 1	1	24674	0.115
reliable_score	Phase 1	2	15607	0.073
reliable_score	Phase 1	3	7708	0.036
reliable_score	Phase 1	4	7274	0.034
reliable_score	Phase 1	5	23202	0.108
reliable_score	Phase 2	1	46526	0.217
reliable_score	Phase 2	2	25561	0.119
reliable_score	Phase 2	3	13753	0.064
reliable_score	Phase 2	4	12477	0.058
reliable_score	Phase 2	5	37253	0.174
share_score	Phase 1	0	27680	0.129
share_score	Phase 1	1	50785	0.237
share_score	Phase 2	0	56839	0.266
share_score	Phase 2	1	78731	0.368

Analysis of time variables

Main survey

• In the main survey, there are three time duration variables we recorded for each participants:

1. intervention_duration is the time duration between the start and the end of the intervention in minutes
2. misinfo_quiz_duration is the time duration between the start and the end of the misinformation quiz in minutes
3. total_duration is the between the start of the chatbot and the end of the misinformation quiz in minutes

• We will now look at the distribution of these time variables by phase, country, payment amount, treatment arm, quiz completion status and passing attention check.

• For the purpose of representation, we showed the distribution of time duration from the 1st percentile to the 85th percentile. The complete summary statistics for the time duration varaibles are reported above.

By user

Summary statistics

• We observed that the distribution of time duration is highly right skewed. The following table shows the quantiles for each of the time duration variables.

summary_stats_table_continous(df_wide,
  variables = c("intervention_duration", "misinfo_quiz_duration", "total_duration"),
  filename = "sum_time_duration_consent.csv",
  caption = "Summary Statistics for Time Duration Variables",
  print_html = TRUE
  )

Summary Statistics for Time Duration Variables

Variable	Total_Observations	Number_Missing	Mean	SD	SE	Min	Max	Decile_10	Decile_20	Decile_30	Decile_40	Decile_50	Decile_60	Decile_70	Decile_80	Decile_90	Decile_91	Decile_92	Decile_93	Decile_94	Decile_95	Decile_96	Decile_97	Decile_98	Decile_99	Decile_100	Proportion_Missing
intervention_duration	52068	18287	278.194	3178.435	13.929	3.967	128975.6	9.250	11.700	13.033	14.300	15.733	17.683	20.767	26.90	60.638	76.232	102.821	144.437	222.816	349.768	550.801	870.765	1458.586	3847.463	128975.6	0.351
misinfo_quiz_duration	43973	26382	299.701	4124.102	19.667	1.200	133550.8	3.817	4.517	5.117	5.717	6.400	7.250	8.400	10.40	17.367	19.709	23.387	29.049	40.917	73.880	169.487	401.799	844.704	2667.263	133550.8	0.600
total_duration	43978	26377	636.825	5498.200	26.218	6.850	135561.0	16.883	19.367	21.467	23.700	26.383	30.133	36.183	51.45	239.367	317.518	426.377	558.294	732.024	990.251	1361.682	2032.999	3890.957	12080.045	135561.0	0.600

Winsorizing

• For analysis purposes, we will winsorize the time duration variables at the 1st and 90th percentile.

df_winsorized <- df_wide %>% select(analytic_id, arm_coded, intervention_duration, misinfo_quiz_duration, total_duration)

df_winsorized$intervention_duration <- Winsorize(df_winsorized$intervention_duration, probs = c(0.01, 0.90), na.rm = TRUE)

df_winsorized$misinfo_quiz_duration <- Winsorize(df_winsorized$misinfo_quiz_duration, probs = c(0.01, 0.90), na.rm = TRUE)

df_winsorized$total_duration <- Winsorize(df_winsorized$total_duration, probs = c(0.01, 0.90), na.rm = TRUE)


# generating a csv

write.csv(df_winsorized, "./data/chatfuel_summary_statistics/winsorized_time_duration.csv")

Intervention duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "intervention_duration", 
                    group_var = NULL , 
                    x_min = quantile(df_wide$intervention_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$intervention_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Intervention (minutes)",
                    x_label = "Minutes", 
                    y_label = "Density",
                    file_basename = "intervention_duration")

Misinformation quiz duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "misinfo_quiz_duration", 
                    group_var = NULL , 
                    x_min = quantile(df_wide$misinfo_quiz_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$misinfo_quiz_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Misinformation Quiz (minutes)",
                    x_label = "Minutes)", 
                    y_label = "Density",
                    file_basename = "quiz_duration")

Total duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "total_duration", 
                    group_var = NULL , 
                    x_min = quantile(df_wide$total_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$total_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Main Survey (minutes)",
                    x_label = "Minutes", 
                    y_label = "Density",
                    file_basename = "total_duration")

By phase

Intervention duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "intervention_duration", 
                    group_var = "phase_coded" , 
                    x_min = quantile(df_wide$intervention_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$intervention_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Intervention (minutes) by Phase",
                    x_label = "Minutes", 
                    y_label = "Density", 
                    file_basename = "intervention_duration_phase")

Misinformation quiz duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "misinfo_quiz_duration", 
                    group_var = "phase_coded" , 
                    x_min = quantile(df_wide$misinfo_quiz_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$misinfo_quiz_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Misinformation Quiz (minutes) by Phase",
                    x_label = "Minutes)", 
                    y_label = "Density",
                    file_basename = "quiz_duration_phase")

Total duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "total_duration", 
                    group_var = "phase_coded" , 
                    x_min = quantile(df_wide$total_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$total_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Main Survey (minutes) by Phase",
                    x_label = "Minutes", 
                    y_label = "Density",
                    file_basename = "total_duration_phase")

By country

Intervention duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "intervention_duration", 
                    group_var = "country_coded" , 
                    x_min = quantile(df_wide$intervention_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$intervention_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Intervention (minutes) by Country",
                    x_label = "Minutes", 
                    y_label = "Density", 
                    file_basename = "intervention_duration_country")

Misinformation quiz duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "misinfo_quiz_duration", 
                    group_var = "country_coded" , 
                    x_min = quantile(df_wide$misinfo_quiz_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$misinfo_quiz_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Misinformation Quiz (minutes) by Country",
                    x_label = "Minutes)", 
                    y_label = "Density",
                    file_basename = "quiz_duration_country")

Total duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "total_duration", 
                    group_var = "country_coded" , 
                    x_min = quantile(df_wide$total_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$total_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Main Survey (minutes) by Country",
                    x_label = "Minutes", 
                    y_label = "Density", 
                    file_basename = "total_duration_country")

By payment amount

Intervention duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "intervention_duration", 
                    group_var = "payment_amount_coded" , 
                    x_min = quantile(df_wide$intervention_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$intervention_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Intervention (minutes) by Payment Amount",
                    x_label = "Minutes", 
                    y_label = "Density",
                    file_basename = "intervention_duration_payment")

Misinformation quiz duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "misinfo_quiz_duration", 
                    group_var = "payment_amount_coded" , 
                    x_min = quantile(df_wide$misinfo_quiz_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$misinfo_quiz_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Misinformation Quiz (minutes) by Payment Amount",
                    x_label = "Minutes)", 
                    y_label = "Density",
                    file_basename = "quiz_duration_payment")

Total duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "total_duration", 
                    group_var = "payment_amount_coded" , 
                    x_min = quantile(df_wide$total_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$total_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Main Survey (minutes) by Payment Amount",
                    x_label = "Minutes", 
                    y_label = "Density",
                    file_basename = "total_duration_payment")

By treatment arm

Intervention duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "intervention_duration", 
                    group_var = "arm_coded" , 
                    x_min = quantile(df_wide$intervention_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$intervention_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Intervention (minutes) by Arm",
                    x_label = "Minutes", 
                    y_label = "Density",
                    file_basename = "intervention_duration_arm")

Misinformation quiz duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "misinfo_quiz_duration", 
                    group_var = "arm_coded" , 
                    x_min = quantile(df_wide$misinfo_quiz_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$misinfo_quiz_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Misinformation Quiz (minutes) by Arm",
                    x_label = "Minutes)", 
                    y_label = "Density",
                    file_basename = "quiz_duration_arm")

Total duration in minutes

create_density_plot(data =df_wide, 
                    x_var = "total_duration", 
                    group_var = "arm_coded" , 
                    x_min = quantile(df_wide$total_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide$total_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Main Survey (minutes) by Arm",
                    x_label = "Minutes", 
                    y_label = "Density",
                    file_basename = "total_duration_arm"
                    )

By attention check passed

Intervention duration in minutes

df_wide_attention_check <- df_wide %>% filter(attention_check_passed == 1| attention_check_passed == 0)

create_density_plot(data =df_wide_attention_check, 
                    x_var = "intervention_duration", 
                    group_var = as.factor(df_wide_attention_check$attention_check_passed) , 
                    x_min = quantile(df_wide_attention_check$intervention_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide_attention_check$intervention_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Intervention (minutes) by Attention Check",
                    x_label = "Minutes", 
                    y_label = "Density", 
                    file_basename = "intervention_duration_attention_check")

Misinformation quiz duration in minutes

create_density_plot(data =df_wide_attention_check, 
                    x_var = "misinfo_quiz_duration", 
                    group_var = as.factor(df_wide_attention_check$attention_check_passed) , 
                    x_min = quantile(df_wide_attention_check$misinfo_quiz_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide_attention_check$misinfo_quiz_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Misinformation Quiz (minutes) by Attention Check",
                    x_label = "Minutes)", 
                    y_label = "Density",
                    file_basename = "quiz_duration_attention_check")

Total duration in minutes

create_density_plot(data =df_wide_attention_check, 
                    x_var = "total_duration", 
                    group_var = as.factor(df_wide_attention_check$attention_check_passed) , 
                    x_min = quantile(df_wide_attention_check$total_duration, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide_attention_check$total_duration, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Main Survey (minutes) by Attention Check",
                    x_label = "Minutes", 
                    y_label = "Density",
                    file_basename = "total_duration_attention_check")

Follow up

Time since main survey

• We have three main variables that measure the time duration between the follow-up survey and the main survey, which are defined as follows:

1. Variable time_since_completed_intervention indicates the time duration since the participant completed the intervention to starting the follow up in days.
2. Variable time_since_completed_quiz indicates the time duration since the participant completed the intervention to starting the follow up survey in days.

Distribution of time duration between started followup and intervention completion

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="time_since_completed_intervention", 
                 group_var = "phase_coded",
                 bin_width = 1,
                 title = "Histogram of Time Since Completed Intervention by Phase in days",
                 x_label = "Time Since Completed Quiz (days)", 
                 y_label = "Frequency",
                 file_basename = "time_since_completed_intervention_phase")

Distribution of time duration between started followup and quiz completion

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="time_since_completed_quiz", 
                 group_var = "phase_coded",
                 bin_width = 1,
                 title = "Histogram of Time Since Completed Quiz by Phase in days",
                 x_label = "Time Since Completed Quiz (days)", 
                 y_label = "Frequency",
                 file_basename = "time_since_completed_quiz_phase")

Time to complete follow-up misinformation quiz

create_density_plot(data =df_wide_followup, 
                    x_var = "duration_complete_quiz_followup", 
                    group_var = NULL , 
                    x_min = quantile(df_wide_followup$duration_complete_quiz_followup, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide_followup$duration_complete_quiz_followup, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Misinformation Quiz (minutes)",
                    x_label = "Minutes)", 
                    y_label = "Density",
                    file_basename = "quiz_duration_followup")

By Phase

create_density_plot(data =df_wide_followup, 
                    x_var = "duration_complete_quiz_followup", 
                    group_var = "phase_coded" , 
                    x_min = quantile(df_wide_followup$duration_complete_quiz_followup, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide_followup$duration_complete_quiz_followup, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Misinformation Quiz (minutes) by Phase",
                    x_label = "Minutes)", 
                    y_label = "Density",
                    file_basename = "quiz_duration_followup_phase")

By Arm

create_density_plot(data =df_wide_followup, 
                    x_var = "duration_complete_quiz_followup", 
                    group_var = "arm_coded" , 
                    x_min = quantile(df_wide_followup$duration_complete_quiz_followup, 0.01, na.rm = TRUE),
                    x_max = quantile(df_wide_followup$duration_complete_quiz_followup, 0.85, na.rm = TRUE),
                    title = "Density Plot of Time to Complete Misinformation Quiz (minutes) by Phase",
                    x_label = "Minutes)", 
                    y_label = "Density",
                    file_basename = "quiz_duration_followup_arm")

Summary of Survey Outcome

• mean_share_misinfo: mean (across posts) intention to share misinformation posts for a given user; range [0, 1]
• mean_reliable_misinfo: mean (across posts) reliable score for misinformation posts for a given user; range [1, 5]
• mean_manipulative_misinfo: mean (across posts) manipulative score for misinformation posts for a given user; range [1, 5]
• mean_share_nonmisinfo: mean (across posts) intention to share non-misinformation post for a given user; range [0, 1]
• mean_reliable_nonmisinfo: mean (across posts) reliable score for non-misinformation posts for a given user; range [1, 5]
• mean_manipulative_nonmisinfo: mean (across posts) manipulative score for non-misinformation posts for a given user; range [1, 5]
• sharing_discernment: difference in mean share score between non-misinformation and misinformation posts for a given user; range [-1, 1]
• reliability_discernment: difference in mean reliable score between non-misinformation and misinformation posts for a given user; range [-4, 4]
• manipulation_discernment: difference in mean manipulative score between misinformation and non-misinformation posts for a given user; range [-4, 4]

Note that the discernment scores are constructed in such a way that a positive value indicates better discernment. For example, positive value for reliability discernment indicates that user scored non-misinformation posts as more reliable than misinformation posts. On the other hand, positive monipulative discernment indicates that user scored non-misinformation posts as less manipulative than misinformation posts.

• This section provides summary statistics and visualizations for the survey outcomes of both the main survey and follow-up survey.

Outcomes by treatment-phase

Main Survey

outcomes <- c("mean_share_misinfo", "mean_reliable_misinfo", "mean_manipulative_misinfo", "mean_share_nonmisinfo", "mean_reliable_nonmisinfo", "mean_manipulative_nonmisinfo", "sharing_discernment", "reliability_discernment", "manipulation_discernment")


summary_stats_table_continous_outcomes(df_wide, 
  variables = outcomes,
  group = "phase_arm",
  filename = "summary_statistics_for_main_survey_outcomes_phase_arm.csv",
  caption = "Summary Statistics for Main Survey Outcomes",
  print_html = TRUE                              
                
)

Summary Statistics for Main Survey Outcomes

Variable	Group	Total_Observations	Number_Missing	Mean	SD	SE	Proportion_Missing
mean_share_misinfo	phase_arm : Phase 1 - Game	3242	4174	0.730	0.316	0.006	1.287
mean_share_misinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	0.618	0.367	0.005	0.612
mean_share_misinfo	phase_arm : Phase 1 - SMS	4026	3444	0.352	0.346	0.005	0.855
mean_share_misinfo	phase_arm : Phase 1 - Video	3854	3618	0.651	0.366	0.006	0.939
mean_share_misinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	0.627	0.371	0.004	0.543
mean_share_misinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	0.624	0.367	0.006	0.463
mean_share_misinfo	phase_arm : Phase 2 - SMS	13688	6564	0.374	0.352	0.003	0.480
mean_reliable_misinfo	phase_arm : Phase 1 - Game	3242	4174	2.908	1.257	0.022	1.287
mean_reliable_misinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	2.837	1.284	0.019	0.612
mean_reliable_misinfo	phase_arm : Phase 1 - SMS	4026	3444	2.188	1.118	0.018	0.855
mean_reliable_misinfo	phase_arm : Phase 1 - Video	3854	3618	2.794	1.279	0.021	0.939
mean_reliable_misinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	2.865	1.265	0.013	0.543
mean_reliable_misinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	2.799	1.261	0.019	0.463
mean_reliable_misinfo	phase_arm : Phase 2 - SMS	13688	6564	2.258	1.116	0.010	0.480
mean_manipulative_misinfo	phase_arm : Phase 1 - Game	3242	4174	2.788	1.264	0.022	1.287
mean_manipulative_misinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	2.942	1.290	0.019	0.612
mean_manipulative_misinfo	phase_arm : Phase 1 - SMS	4026	3444	3.154	1.313	0.021	0.855
mean_manipulative_misinfo	phase_arm : Phase 1 - Video	3854	3618	2.861	1.296	0.021	0.939
mean_manipulative_misinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	3.021	1.282	0.013	0.543
mean_manipulative_misinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	2.974	1.292	0.020	0.463
mean_manipulative_misinfo	phase_arm : Phase 2 - SMS	13688	6564	3.176	1.276	0.011	0.480
mean_share_nonmisinfo	phase_arm : Phase 1 - Game	3242	4174	0.797	0.309	0.005	1.287
mean_share_nonmisinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	0.762	0.338	0.005	0.612
mean_share_nonmisinfo	phase_arm : Phase 1 - SMS	4026	3444	0.634	0.375	0.006	0.855
mean_share_nonmisinfo	phase_arm : Phase 1 - Video	3854	3618	0.804	0.314	0.005	0.939
mean_share_nonmisinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	0.769	0.333	0.003	0.543
mean_share_nonmisinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	0.756	0.342	0.005	0.463
mean_share_nonmisinfo	phase_arm : Phase 2 - SMS	13688	6564	0.644	0.371	0.003	0.480
mean_reliable_nonmisinfo	phase_arm : Phase 1 - Game	3242	4174	3.110	1.334	0.023	1.287
mean_reliable_nonmisinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	3.246	1.354	0.020	0.612
mean_reliable_nonmisinfo	phase_arm : Phase 1 - SMS	4026	3444	2.914	1.324	0.021	0.855
mean_reliable_nonmisinfo	phase_arm : Phase 1 - Video	3854	3618	3.230	1.357	0.022	0.939
mean_reliable_nonmisinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	3.285	1.320	0.014	0.543
mean_reliable_nonmisinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	3.190	1.358	0.021	0.463
mean_reliable_nonmisinfo	phase_arm : Phase 2 - SMS	13688	6564	2.938	1.297	0.011	0.480
mean_manipulative_nonmisinfo	phase_arm : Phase 1 - Game	3242	4174	2.498	1.320	0.023	1.287
mean_manipulative_nonmisinfo	phase_arm : Phase 1 - Original Baseline	4571	2798	2.589	1.380	0.020	0.612
mean_manipulative_nonmisinfo	phase_arm : Phase 1 - SMS	4026	3444	2.497	1.326	0.021	0.855
mean_manipulative_nonmisinfo	phase_arm : Phase 1 - Video	3854	3618	2.445	1.340	0.022	0.939
mean_manipulative_nonmisinfo	phase_arm : Phase 2 - Long Baseline	9163	4975	2.620	1.358	0.014	0.543
mean_manipulative_nonmisinfo	phase_arm : Phase 2 - Original Baseline	4263	1975	2.556	1.343	0.021	0.463
mean_manipulative_nonmisinfo	phase_arm : Phase 2 - SMS	13688	6564	2.537	1.298	0.011	0.480
sharing_discernment	phase_arm : Phase 1 - Game	3242	4174	0.067	0.366	0.006	1.287
sharing_discernment	phase_arm : Phase 1 - Original Baseline	4571	2798	0.143	0.391	0.006	0.612
sharing_discernment	phase_arm : Phase 1 - SMS	4026	3444	0.283	0.422	0.007	0.855
sharing_discernment	phase_arm : Phase 1 - Video	3854	3618	0.153	0.374	0.006	0.939
sharing_discernment	phase_arm : Phase 2 - Long Baseline	9163	4975	0.142	0.382	0.004	0.543
sharing_discernment	phase_arm : Phase 2 - Original Baseline	4263	1975	0.133	0.380	0.006	0.463
sharing_discernment	phase_arm : Phase 2 - SMS	13688	6564	0.270	0.423	0.004	0.480
reliability_discernment	phase_arm : Phase 1 - Game	3242	4174	0.202	1.178	0.021	1.287
reliability_discernment	phase_arm : Phase 1 - Original Baseline	4571	2798	0.409	1.312	0.019	0.612
reliability_discernment	phase_arm : Phase 1 - SMS	4026	3444	0.726	1.386	0.022	0.855
reliability_discernment	phase_arm : Phase 1 - Video	3854	3618	0.435	1.249	0.020	0.939
reliability_discernment	phase_arm : Phase 2 - Long Baseline	9163	4975	0.420	1.325	0.014	0.543
reliability_discernment	phase_arm : Phase 2 - Original Baseline	4263	1975	0.391	1.292	0.020	0.463
reliability_discernment	phase_arm : Phase 2 - SMS	13688	6564	0.680	1.364	0.012	0.480
manipulation_discernment	phase_arm : Phase 1 - Game	3242	4174	0.290	1.238	0.022	1.287
manipulation_discernment	phase_arm : Phase 1 - Original Baseline	4571	2798	0.353	1.362	0.020	0.612
manipulation_discernment	phase_arm : Phase 1 - SMS	4026	3444	0.657	1.471	0.023	0.855
manipulation_discernment	phase_arm : Phase 1 - Video	3854	3618	0.416	1.316	0.021	0.939
manipulation_discernment	phase_arm : Phase 2 - Long Baseline	9163	4975	0.401	1.351	0.014	0.543
manipulation_discernment	phase_arm : Phase 2 - Original Baseline	4263	1975	0.418	1.338	0.020	0.463
manipulation_discernment	phase_arm : Phase 2 - SMS	13688	6564	0.639	1.445	0.012	0.480

Followup

df_wide_followup = df_wide_followup |> mutate(phase_arm = paste(phase_coded, arm_coded, sep = " - "))

outcomes_followup <- c("mean_share_misinfo_followup", "mean_reliable_misinfo_followup", "mean_manipulative_misinfo_followup", "mean_share_nonmisinfo_followup", "mean_reliable_nonmisinfo_followup", "mean_manipulative_nonmisinfo_followup", "sharing_discernment_followup", "reliability_discernment_followup", "manipulation_discernment_followup")


summary_stats_table_continous_outcomes(df_wide_followup, 
  variables = outcomes_followup,
  group = "phase_arm",
  filename = "summary_statistics_for_followup_outcomes_phase_arm.csv",
  caption = "Summary Statistics for Followup Outcomes",
  print_html = TRUE                              
                
)

Summary Statistics for Followup Outcomes

Variable	Group	Total_Observations	Number_Missing	Mean	SD	SE	Proportion_Missing
mean_share_misinfo_followup	phase_arm : Phase 1 - Game	2024	160	0.650	0.361	0.008	0.079
mean_share_misinfo_followup	phase_arm : Phase 1 - Original Baseline	2798	267	0.585	0.380	0.007	0.095
mean_share_misinfo_followup	phase_arm : Phase 1 - SMS	2447	229	0.423	0.369	0.007	0.094
mean_share_misinfo_followup	phase_arm : Phase 1 - Video	2334	184	0.610	0.383	0.008	0.079
mean_share_misinfo_followup	phase_arm : Phase 2 - Long Baseline	3061	192	0.570	0.389	0.007	0.063
mean_share_misinfo_followup	phase_arm : Phase 2 - Original Baseline	1408	86	0.572	0.393	0.010	0.061
mean_share_misinfo_followup	phase_arm : Phase 2 - SMS	4459	301	0.414	0.372	0.006	0.068
mean_reliable_misinfo_followup	phase_arm : Phase 1 - Game	2024	160	2.776	1.207	0.027	0.079
mean_reliable_misinfo_followup	phase_arm : Phase 1 - Original Baseline	2798	267	2.709	1.253	0.024	0.095
mean_reliable_misinfo_followup	phase_arm : Phase 1 - SMS	2447	229	2.299	1.126	0.023	0.094
mean_reliable_misinfo_followup	phase_arm : Phase 1 - Video	2334	184	2.707	1.253	0.026	0.079
mean_reliable_misinfo_followup	phase_arm : Phase 2 - Long Baseline	3061	192	2.741	1.228	0.022	0.063
mean_reliable_misinfo_followup	phase_arm : Phase 2 - Original Baseline	1408	86	2.722	1.253	0.033	0.061
mean_reliable_misinfo_followup	phase_arm : Phase 2 - SMS	4459	301	2.368	1.152	0.017	0.068
mean_manipulative_misinfo_followup	phase_arm : Phase 1 - Game	2024	160	2.798	1.241	0.028	0.079
mean_manipulative_misinfo_followup	phase_arm : Phase 1 - Original Baseline	2798	267	2.886	1.260	0.024	0.095
mean_manipulative_misinfo_followup	phase_arm : Phase 1 - SMS	2447	229	3.033	1.265	0.026	0.094
mean_manipulative_misinfo_followup	phase_arm : Phase 1 - Video	2334	184	2.858	1.291	0.027	0.079
mean_manipulative_misinfo_followup	phase_arm : Phase 2 - Long Baseline	3060	193	3.086	1.255	0.023	0.063
mean_manipulative_misinfo_followup	phase_arm : Phase 2 - Original Baseline	1408	86	3.028	1.268	0.034	0.061
mean_manipulative_misinfo_followup	phase_arm : Phase 2 - SMS	4459	301	3.173	1.252	0.019	0.068
mean_share_nonmisinfo_followup	phase_arm : Phase 1 - Game	2024	160	0.776	0.331	0.007	0.079
mean_share_nonmisinfo_followup	phase_arm : Phase 1 - Original Baseline	2798	267	0.754	0.349	0.007	0.095
mean_share_nonmisinfo_followup	phase_arm : Phase 1 - SMS	2447	229	0.664	0.377	0.008	0.094
mean_share_nonmisinfo_followup	phase_arm : Phase 1 - Video	2334	184	0.768	0.346	0.007	0.079
mean_share_nonmisinfo_followup	phase_arm : Phase 2 - Long Baseline	3061	192	0.747	0.353	0.006	0.063
mean_share_nonmisinfo_followup	phase_arm : Phase 2 - Original Baseline	1408	86	0.747	0.348	0.009	0.061
mean_share_nonmisinfo_followup	phase_arm : Phase 2 - SMS	4459	301	0.681	0.369	0.006	0.068
mean_reliable_nonmisinfo_followup	phase_arm : Phase 1 - Game	2024	160	3.076	1.266	0.028	0.079
mean_reliable_nonmisinfo_followup	phase_arm : Phase 1 - Original Baseline	2798	267	3.132	1.326	0.025	0.095
mean_reliable_nonmisinfo_followup	phase_arm : Phase 1 - SMS	2447	229	2.937	1.316	0.027	0.094
mean_reliable_nonmisinfo_followup	phase_arm : Phase 1 - Video	2334	184	3.092	1.332	0.028	0.079
mean_reliable_nonmisinfo_followup	phase_arm : Phase 2 - Long Baseline	3061	192	3.233	1.267	0.023	0.063
mean_reliable_nonmisinfo_followup	phase_arm : Phase 2 - Original Baseline	1408	86	3.225	1.260	0.034	0.061
mean_reliable_nonmisinfo_followup	phase_arm : Phase 2 - SMS	4459	301	3.040	1.282	0.019	0.068
mean_manipulative_nonmisinfo_followup	phase_arm : Phase 1 - Game	2023	161	2.456	1.261	0.028	0.080
mean_manipulative_nonmisinfo_followup	phase_arm : Phase 1 - Original Baseline	2798	267	2.482	1.316	0.025	0.095
mean_manipulative_nonmisinfo_followup	phase_arm : Phase 1 - SMS	2446	230	2.420	1.275	0.026	0.094
mean_manipulative_nonmisinfo_followup	phase_arm : Phase 1 - Video	2334	184	2.442	1.305	0.027	0.079
mean_manipulative_nonmisinfo_followup	phase_arm : Phase 2 - Long Baseline	3061	192	2.593	1.324	0.024	0.063
mean_manipulative_nonmisinfo_followup	phase_arm : Phase 2 - Original Baseline	1408	86	2.541	1.310	0.035	0.061
mean_manipulative_nonmisinfo_followup	phase_arm : Phase 2 - SMS	4458	302	2.537	1.305	0.020	0.068
sharing_discernment_followup	phase_arm : Phase 1 - Game	2024	160	0.126	0.381	0.008	0.079
sharing_discernment_followup	phase_arm : Phase 1 - Original Baseline	2798	267	0.169	0.378	0.007	0.095
sharing_discernment_followup	phase_arm : Phase 1 - SMS	2447	229	0.241	0.407	0.008	0.094
sharing_discernment_followup	phase_arm : Phase 1 - Video	2334	184	0.158	0.368	0.008	0.079
sharing_discernment_followup	phase_arm : Phase 2 - Long Baseline	3061	192	0.178	0.394	0.007	0.063
sharing_discernment_followup	phase_arm : Phase 2 - Original Baseline	1408	86	0.175	0.383	0.010	0.061
sharing_discernment_followup	phase_arm : Phase 2 - SMS	4459	301	0.268	0.423	0.006	0.068
reliability_discernment_followup	phase_arm : Phase 1 - Game	2024	160	0.300	1.159	0.026	0.079
reliability_discernment_followup	phase_arm : Phase 1 - Original Baseline	2798	267	0.423	1.225	0.023	0.095
reliability_discernment_followup	phase_arm : Phase 1 - SMS	2447	229	0.638	1.262	0.026	0.094
reliability_discernment_followup	phase_arm : Phase 1 - Video	2334	184	0.385	1.175	0.024	0.079
reliability_discernment_followup	phase_arm : Phase 2 - Long Baseline	3061	192	0.492	1.249	0.023	0.063
reliability_discernment_followup	phase_arm : Phase 2 - Original Baseline	1408	86	0.504	1.201	0.032	0.061
reliability_discernment_followup	phase_arm : Phase 2 - SMS	4459	301	0.672	1.306	0.020	0.068
manipulation_discernment_followup	phase_arm : Phase 1 - Game	2023	161	0.342	1.191	0.026	0.080
manipulation_discernment_followup	phase_arm : Phase 1 - Original Baseline	2798	267	0.405	1.252	0.024	0.095
manipulation_discernment_followup	phase_arm : Phase 1 - SMS	2446	230	0.613	1.321	0.027	0.094
manipulation_discernment_followup	phase_arm : Phase 1 - Video	2334	184	0.416	1.206	0.025	0.079
manipulation_discernment_followup	phase_arm : Phase 2 - Long Baseline	3060	193	0.493	1.315	0.024	0.063
manipulation_discernment_followup	phase_arm : Phase 2 - Original Baseline	1408	86	0.487	1.286	0.034	0.061
manipulation_discernment_followup	phase_arm : Phase 2 - SMS	4458	302	0.637	1.377	0.021	0.068

By Post Order

Main Survey

outcomes_post <- c("share_score", "reliable_score", "manipulative_score")


summary_stats_table_continous_outcomes(df_long, 
  variables = outcomes_post,
  group = "post_number",
  filename = "summary_statistics_post_number.csv",
  caption = "Summary Statistics for Score by Post Order Main Survey",
  print_html = TRUE                              
                
)

Summary Statistics for Score by Post Order Main Survey

Variable	Group	Total_Observations	Number_Missing	Mean	SD	SE	Proportion_Missing
share_score	post_number : 1	35619	0	0.582	0.493	0.003	0
share_score	post_number : 2	35697	0	0.598	0.490	0.003	0
share_score	post_number : 3	35712	0	0.610	0.488	0.003	0
share_score	post_number : 4	35605	0	0.609	0.488	0.003	0
share_score	post_number : 5	35641	0	0.614	0.487	0.003	0
share_score	post_number : 6	35761	0	0.617	0.486	0.003	0
reliable_score	post_number : 1	35619	0	2.657	1.641	0.009	0
reliable_score	post_number : 2	35697	0	2.782	1.639	0.009	0
reliable_score	post_number : 3	35712	0	2.829	1.643	0.009	0
reliable_score	post_number : 4	35605	0	2.851	1.648	0.009	0
reliable_score	post_number : 5	35641	0	2.838	1.643	0.009	0
reliable_score	post_number : 6	35761	0	2.840	1.645	0.009	0
manipulative_score	post_number : 1	35619	0	3.016	1.671	0.009	0
manipulative_score	post_number : 2	35697	0	2.856	1.667	0.009	0
manipulative_score	post_number : 3	35712	0	2.797	1.654	0.009	0
manipulative_score	post_number : 4	35605	0	2.805	1.665	0.009	0
manipulative_score	post_number : 5	35641	0	2.803	1.665	0.009	0
manipulative_score	post_number : 6	35761	0	2.773	1.663	0.009	0

Followup

outcomes_post_followup <- c("share_score_followup", "reliable_score_followup", "manipulative_score_followup")


summary_stats_table_continous_outcomes(df_long_followup, 
  variables = outcomes_post_followup,
  group = "post_number_followup",
  filename = "summary_statistics_post_number_followup.csv",
  caption = "Summary Statistics for Score by Post Order Followup",
  print_html = TRUE                              
                
)

Summary Statistics for Score by Post Order Followup

Variable	Group	Total_Observations	Number_Missing	Mean	SD	SE	Proportion_Missing
share_score_followup	post_number_followup : 1	15429	0	0.607	0.488	0.004	0
share_score_followup	post_number_followup : 2	15466	0	0.599	0.490	0.004	0
share_score_followup	post_number_followup : 3	15407	0	0.609	0.488	0.004	0
share_score_followup	post_number_followup : 4	15463	0	0.614	0.487	0.004	0
share_score_followup	post_number_followup : 5	15427	0	0.618	0.486	0.004	0
share_score_followup	post_number_followup : 6	15463	0	0.603	0.489	0.004	0
reliable_score_followup	post_number_followup : 1	15429	0	2.701	1.566	0.013	0
reliable_score_followup	post_number_followup : 2	15466	0	2.750	1.574	0.013	0
reliable_score_followup	post_number_followup : 3	15407	0	2.811	1.578	0.013	0
reliable_score_followup	post_number_followup : 4	15463	0	2.832	1.584	0.013	0
reliable_score_followup	post_number_followup : 5	15427	0	2.832	1.581	0.013	0
reliable_score_followup	post_number_followup : 6	15463	0	2.817	1.580	0.013	0
manipulative_score_followup	post_number_followup : 1	15427	2	2.869	1.597	0.013	0
manipulative_score_followup	post_number_followup : 2	15466	0	2.814	1.598	0.013	0
manipulative_score_followup	post_number_followup : 3	15407	0	2.777	1.602	0.013	0
manipulative_score_followup	post_number_followup : 4	15462	1	2.780	1.604	0.013	0
manipulative_score_followup	post_number_followup : 5	15426	1	2.785	1.598	0.013	0
manipulative_score_followup	post_number_followup : 6	15463	0	2.798	1.608	0.013	0

Correlation Matrix

Main Survey

df_correlation <- df_wide %>% 
  select(mean_share_misinfo, mean_reliable_misinfo, mean_manipulative_misinfo, mean_share_nonmisinfo, mean_reliable_nonmisinfo, mean_manipulative_nonmisinfo, sharing_discernment, reliability_discernment, manipulation_discernment, country_coded)

df_correlation <- dummy_cols(df_correlation, select_columns = "country_coded") %>% select(-country_coded)

correlation_matrix <- cor(df_correlation, use = "pairwise.complete.obs")

col <- colorRampPalette(c("#BB4444", "#EE9988", "#FFFFFF", "#77AADD", "#4477AA"))
colors <- col(200)
annotation_colors <- c("white", "black")

corrplot(correlation_matrix,
         method = "color", 
         col = colors,
         type = "upper",
         addCoef.col = "black",
         tl.col = "black",
         tl.srt = 90,
         diag = FALSE,
         bg = "white",
         addrect = 2,
         rect.col = "gray",
         cl.pos = "n",
         number.cex  = 0.4,  # Adjust the font size for correlation values
         tl.cex = 0.5,      # Adjust the font size for variable names
         addCoef.asPercent = TRUE,
         p.mat = NULL,
         sig.level = 0.01,
         insig = "blank",
         pch.col = "black",
         pch.cex = 0.8,
         col.cor = "black",
         mar = c(0,0,1,0))

Followup

df_correlation_followup <- df_wide_followup %>% 
  select(mean_share_misinfo_followup, mean_reliable_misinfo_followup, mean_manipulative_misinfo_followup, mean_share_nonmisinfo_followup, mean_reliable_nonmisinfo_followup, mean_manipulative_nonmisinfo_followup, sharing_discernment_followup, reliability_discernment_followup, manipulation_discernment_followup)

correlation_matrix_followup <- cor(df_correlation_followup, use = "pairwise.complete.obs")

col <- colorRampPalette(c("#BB4444", "#EE9988", "#FFFFFF", "#77AADD", "#4477AA"))
colors <- col(200)
annotation_colors <- c("white", "black")

corrplot(correlation_matrix_followup,
         method = "color", 
         col = colors,
         type = "upper",
         addCoef.col = "black",
         tl.col = "black",
         tl.srt = 90,
         diag = FALSE,
         bg = "white",
         addrect = 2,
         rect.col = "gray",
         cl.pos = "n",
         number.cex  = 0.4,  # Adjust the font size for correlation values
         tl.cex = 0.5,      # Adjust the font size for variable names
         addCoef.asPercent = TRUE,
         p.mat = NULL,
         sig.level = 0.01,
         insig = "blank",
         pch.col = "black",
         pch.cex = 0.8,
         col.cor = "black",
         mar = c(0,0,1,0))

Overtime

Main Survey

We started the experiment on November 16th, 2023 and ended the experiment on December 31st, 2023. The graphs below show the average score outcomes in days since the starting of the experiment.

first_start_time_chat <- min(df_wide$MisinfoChat_start_time)
df_wide <- df_wide |> 
  mutate(time_since_first_start = floor(as.numeric(difftime(MisinfoChat_start_time, first_start_time_chat, units = "days")))) # round to nearest day 

df_time_graph <- df_wide %>% 
  select(time_since_first_start, mean_share_misinfo, mean_reliable_misinfo, mean_manipulative_misinfo, mean_share_nonmisinfo, mean_reliable_nonmisinfo, mean_manipulative_nonmisinfo, sharing_discernment, reliability_discernment, manipulation_discernment, attention_check_passed) %>% 
  group_by(time_since_first_start) %>% 
  summarise_all(mean, na.rm = TRUE)

main_out_since_start <- ggplot(data=df_time_graph, aes(x=time_since_first_start)) +
  geom_line(aes(y=mean_share_misinfo, color = "Mean Share Misinfo"), size = 1) +
  geom_line(aes(y=mean_reliable_misinfo, color = "Mean Reliable Misinfo"), size = 1) +
  geom_line(aes(y=mean_manipulative_misinfo, color = "Mean Manipulative Misinfo"), size = 1) +
  geom_line(aes(y=mean_share_nonmisinfo, color = "Mean Share Nonmisinfo"), size = 1) +
  geom_line(aes(y=mean_reliable_nonmisinfo, color = "Mean Reliable Nonmisinfo"), size = 1) +
  geom_line(aes(y=mean_manipulative_nonmisinfo, color = "Mean Manipulative Nonmisinfo"), size = 1) +
  geom_line(aes(y=sharing_discernment, color = "Sharing Discernment"), size = 1) +
  geom_line(aes(y=reliability_discernment, color = "Reliability Discernment"), size = 1) +
  geom_line(aes(y=manipulation_discernment, color = "Manipulation Discernment"), size = 1) +
  labs(
    title = "Main Survey Outcomes Since Beginning of the Experiment",
    x = "Time Since First Start (days)",
    y = "Mean Score",
    color = "Legend"
  ) +
  theme_minimal()+
  scale_x_continuous(expand = expand_scale(add = c(0.5, 0.5))) +
    guides(fill = guide_legend(title = NULL))

print(main_out_since_start)

ggsave(paste0("./data/chatfuel_summary_statistics/", "main_outcomes_since_start", ".jpg"), main_out_since_start, device = "jpg", width = 10, height = 8)

ggsave(paste0("./data/chatfuel_summary_statistics/", "main_outcomes_since_start", ".pdf"), main_out_since_start, device = "pdf", width = 10, height = 8)

Distribution of outcomes since start

reliability_score <- ggplot(data = df_time_graph, aes(x = time_since_first_start)) +
  geom_smooth(aes(y = reliability_discernment, color = "Reliability Discernment"), size = 1) +
  geom_smooth(aes(y = mean_reliable_misinfo, color = "Mean Reliability Misinfo"), size = 1) +
  geom_smooth(aes(y = mean_reliable_nonmisinfo, color = "Mean Reliability Nonmisinfo"), size = 1) +
  scale_color_brewer(palette = "Dark2") +
  labs(
    title = "Reliability Score Since Beginning of the Experiment",
    x = "Time Since First Start (days)",
    y = "Reliability Score",
    color = "Legend"
  ) +
  theme_minimal()+
  scale_x_continuous(expand = expand_scale(add = c(0.5, 0.5))) +
    guides(fill = guide_legend(title = NULL))

print(reliability_score) 

ggsave(paste0("./data/chatfuel_summary_statistics/", "reliability_score_since_start", ".jpg"), reliability_score, device = "jpg", width = 10, height = 8)

ggsave(paste0("./data/chatfuel_summary_statistics/", "reliability_score_since_start", ".pdf"), reliability_score, device = "pdf", width = 10, height = 8)



manipulative_score <- ggplot(data = df_time_graph, aes(x = time_since_first_start)) +
  geom_smooth(aes(y = manipulation_discernment, color = "Manipulation Discernment"), size = 1) +
  geom_smooth(aes(y = mean_manipulative_misinfo, color = "Mean Manipulative Misinfo"), size = 1) +
  geom_smooth(aes(y = mean_manipulative_nonmisinfo, color = "Mean Manipulative Nonmisinfo"), size = 1) +
  scale_color_brewer(palette = "Dark2") +
  labs(
    title = "Manipulative Score Since Beginning of the Experiment",
    x = "Time Since First Start (days)",
    y = "Manipulative Score",
    color = "Legend"
  ) +
  theme_minimal()+
  scale_x_continuous(expand = expand_scale(add = c(0.5, 0.5))) +
    guides(fill = guide_legend(title = NULL))

print(manipulative_score) 

ggsave(paste0("./data/chatfuel_summary_statistics/", "manipulative_score_since_start", ".jpg"), manipulative_score, device = "jpg", width = 10, height = 8)

ggsave(paste0("./data/chatfuel_summary_statistics/", "manipulative_score_since_start", ".pdf"), manipulative_score, device = "pdf", width = 10, height = 8)


share_score <- ggplot(data = df_time_graph, aes(x = time_since_first_start)) +
  geom_smooth(aes(y = sharing_discernment, color = "Sharing Discernment"), size = 1) +
  geom_smooth(aes(y = mean_share_misinfo, color = "Mean Share Misinfo"), size = 1) +
  geom_smooth(aes(y = mean_share_nonmisinfo, color = "Mean Share Nonmisinfo"), size = 1) +
  scale_color_brewer(palette = "Dark2") +
  labs(
    title = "Share Score Since Beginning of the Experiment",
    x = "Time Since First Start (days)",
    y = "Share Score",
    color = "Legend"
  ) +
  theme_minimal()+
  scale_x_continuous(expand = expand_scale(add = c(0.5, 0.5))) +
    guides(fill = guide_legend(title = NULL))

print(share_score) 

ggsave(paste0("./data/chatfuel_summary_statistics/", "share_score_since_start", ".jpg"), share_score, device = "jpg", width = 10, height = 8)

ggsave(paste0("./data/chatfuel_summary_statistics/", "share_score_since_start", ".pdf"), share_score, device = "pdf", width = 10, height = 8)

attention_check_since_start<-ggplot(data=df_time_graph, aes(x=time_since_first_start)) +
  geom_line(aes(y=attention_check_passed, color = "Attention Check Passed"), size = 1) +
  labs(
    title = "Attention Check Passing Rate Since Beginning of the Experiment",
    x = "Time Since First Start (days)",
    y = "Passing Rate",
    color = "Legend"
  ) +
  theme_minimal()

print(attention_check_since_start)

ggsave(paste0("./data/chatfuel_summary_statistics/", "attention_check_since_start", ".jpg"), attention_check_since_start, device = "jpg", width = 10, height = 8)

ggsave(paste0("./data/chatfuel_summary_statistics/", "attention_check_since_start", ".pdf"), attention_check_since_start, device = "pdf", width = 10, height = 8)

Followup

first_start_time_chat_followup <- min(df_wide_followup$MisinfoChat_start_time_followup)
df_wide_followup <- df_wide_followup |> 
  mutate(time_since_first_start_followup = floor(as.numeric(difftime(MisinfoChat_start_time_followup, first_start_time_chat_followup, units = "days")))) # round to nearest day

df_time_graph_followup <- df_wide_followup %>% 
   select(time_since_first_start_followup, mean_share_misinfo_followup, mean_reliable_misinfo_followup, mean_manipulative_misinfo_followup, mean_share_nonmisinfo_followup, mean_reliable_nonmisinfo_followup, mean_manipulative_nonmisinfo_followup, sharing_discernment_followup, reliability_discernment_followup, manipulation_discernment_followup) %>% 
  group_by(time_since_first_start_followup) %>% 
  summarise_all(mean, na.rm = TRUE)

followup_out_since_start <- ggplot(data=df_time_graph_followup, aes(x=time_since_first_start_followup)) +
  geom_line(aes(y=mean_share_misinfo_followup, color = "Mean Share Misinfo"), size = 1) +
  geom_line(aes(y=mean_reliable_misinfo_followup, color = "Mean Reliable Misinfo"), size = 1) +
  geom_line(aes(y=mean_manipulative_misinfo_followup, color = "Mean Manipulative Misinfo"), size = 1) +
  geom_line(aes(y=mean_share_nonmisinfo_followup, color = "Mean Share Nonmisinfo"), size = 1) +
  geom_line(aes(y=mean_reliable_nonmisinfo_followup, color = "Mean Reliable Nonmisinfo"), size = 1) +
  geom_line(aes(y=mean_manipulative_nonmisinfo_followup, color = "mean_manipulative_nonmisinfo"), size = 1) +
  geom_line(aes(y=sharing_discernment_followup, color = "Sharing Discernment"), size = 1) +
  geom_line(aes(y=reliability_discernment_followup, color = "Reliability Discernment"), size = 1) +
  geom_line(aes(y=manipulation_discernment_followup, color = "Manipulation Discernment"), size = 1) +
  labs(
    title = "Followup Outcomes Since Beginning of the Followup",
    x = "Time Since First Start of Followup (days)",
    y = "Mean Score",
    color = "Legend"
  ) +
  theme_minimal()+
  scale_x_continuous(expand = expand_scale(add = c(0.5, 0.5))) +
    guides(fill = guide_legend(title = NULL))
print(followup_out_since_start)

ggsave(paste0("./data/chatfuel_summary_statistics/", "followup_out_since_start", ".jpg"), followup_out_since_start, device = "jpg", width = 10, height = 8)

ggsave(paste0("./data/chatfuel_summary_statistics/", "followup_out_since_start", ".pdf"), followup_out_since_start, device = "pdf", width = 10, height = 8)

Main Survey

By user

Reliable Score

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_reliable_misinfo", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Reliable Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_misinfo_score")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_reliable_nonmisinfo", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Reliable Non-Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_nonmisinfo_score")

create_histogram_grouped(data = df_wide, 
                 x_var ="reliability_discernment", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Reliable Discernment",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_discernment")

Manipulativeness score

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_manipulative_misinfo", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Manipulation Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_misinfo_score")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_manipulative_nonmisinfo", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Manipulation Non-Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_nonmisinfo_score")

create_histogram_grouped(data = df_wide, 
                 x_var ="manipulation_discernment", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Manipulation Discernment",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulation_discernment")

Sharing

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_share_misinfo", 
                 group_var = NULL,
                 bin_width = 0.5,
                 title = "Histogram of Sharing Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename= "sharing_misinfo_score")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_share_nonmisinfo", 
                 group_var = NULL,
                 bin_width = 0.5,
                 title = "Histogram of Sharing Non-Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_nonmisinfo_score")

create_histogram_grouped(data = df_wide, 
                 x_var ="sharing_discernment", 
                 group_var = NULL,
                 bin_width = 0.5,
                 title = "Histogram of Sharing Discernment",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_discernment")

By Phase

Reliable Score

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_reliable_misinfo", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_misinfo_score_by_phase")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_reliable_nonmisinfo", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Non-Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_nonmisinfo_score_by_phase")

create_histogram_grouped(data = df_wide, 
                 x_var ="reliability_discernment", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Discernment By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_discernment_by_phase")

Manipulativeness score

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_manipulative_misinfo", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_misinfo_score_by_phase")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_manipulative_nonmisinfo", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Non-Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_nonmisinfo_score_by_phase")

create_histogram_grouped(data = df_wide, 
                 x_var ="manipulation_discernment", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Discernment By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulation_discernment_by_phase")

Sharing

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_share_misinfo", 
                 group_var = "phase_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_misinfo_score_by_phase")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_share_nonmisinfo", 
                 group_var = "phase_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Non-Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_nonmisinfo_score_by_phase")

create_histogram_grouped(data = df_wide, 
                 x_var ="sharing_discernment", 
                 group_var = "phase_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Discernment By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_discernment_by_phase")

By Country

Reliable Score

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_reliable_misinfo", 
                 group_var = "country_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Misinformation Score By Country",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_misinfo_score_by_country")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_reliable_nonmisinfo", 
                 group_var = "country_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Non-Misinformation Score By Country",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_nonmisinfo_score_by_country")

create_histogram_grouped(data = df_wide, 
                 x_var ="reliability_discernment", 
                 group_var = "country_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Discernment By Country",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_discernment_by_country")

Manipulativeness score

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_manipulative_misinfo", 
                 group_var = "country_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Misinformation Score By Country",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_misinfo_score_by_country")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_manipulative_nonmisinfo", 
                 group_var = "country_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Non-Misinformation Score By Country",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_nonmisinfo_score_by_country")

create_histogram_grouped(data = df_wide, 
                 x_var ="manipulation_discernment", 
                 group_var = "country_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Discernment By Country",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulation_discernment_by_country")

Sharing

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_share_misinfo", 
                 group_var = "country_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Misinformation Score By Country",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_misinfo_score_by_country")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_share_nonmisinfo", 
                 group_var = "country_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Non-Misinformation Score By Country",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_nonmisinfo_score_by_country")

create_histogram_grouped(data = df_wide, 
                 x_var ="sharing_discernment", 
                 group_var = "country_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Discernment By Country",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_discernment_by_country")

By Payment Amount

Reliable Score

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_reliable_misinfo", 
                 group_var = "payment_amount_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Misinformation Score By Payment Amount",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_misinfo_score_by_payment_amount")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_reliable_nonmisinfo", 
                 group_var = "payment_amount_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Non-Misinformation Score By Payment Amount",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_nonmisinfo_score_by_payment_amount")

create_histogram_grouped(data = df_wide, 
                 x_var ="reliability_discernment", 
                 group_var = "payment_amount_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Discernment By Payment Amount",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_discernment_by_payment_amount")

Manipulativeness score

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_manipulative_misinfo", 
                 group_var = "payment_amount_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Misinformation Score By Payment Amount",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_misinfo_score_by_payment_amount")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_manipulative_nonmisinfo", 
                 group_var = "payment_amount_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Non-Misinformation Score By Payment Amount",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_nonmisinfo_score_by_payment_amount")

create_histogram_grouped(data = df_wide, 
                 x_var ="manipulation_discernment", 
                 group_var = "payment_amount_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Discernment By Payment Amount",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulation_discernment_by_payment_amount")

Sharing

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_share_misinfo", 
                 group_var = "payment_amount_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Misinformation Score By Payment Amount",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_misinfo_score_by_payment_amount")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_share_nonmisinfo", 
                 group_var = "payment_amount_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Non-Misinformation Score By Payment Amount",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_nonmisinfo_score_by_payment_amount")

create_histogram_grouped(data = df_wide, 
                 x_var ="sharing_discernment", 
                 group_var = "payment_amount_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Discernment By Payment Amount",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_discernment_by_payment_amount")

By Arm

Reliable Score

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_reliable_misinfo", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_misinfo_score_by_arm")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_reliable_nonmisinfo", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Non-Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_nonmisinfo_score_by_arm")

create_histogram_grouped(data = df_wide, 
                 x_var ="reliability_discernment", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Discernment By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_discernment_by_arm")

Manipulativeness score

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_manipulative_misinfo", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_misinfo_score_by_arm")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_manipulative_nonmisinfo", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Non-Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_nonmisinfo_score_by_arm")

create_histogram_grouped(data = df_wide, 
                 x_var ="manipulation_discernment", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Discernment By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulation_discernment_by_arm")

Sharing

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_share_misinfo", 
                 group_var = "arm_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_misinfo_score_by_arm")

create_histogram_grouped(data = df_wide, 
                 x_var ="mean_share_nonmisinfo", 
                 group_var = "arm_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Non-Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_nonmisinfo_score_by_arm")

create_histogram_grouped(data = df_wide, 
                 x_var ="sharing_discernment", 
                 group_var = "arm_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Discernment By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_discernment_by_arm")

By Attention Check Passed

Reliable Score

create_histogram_grouped(data = df_wide_attention_check, 
                 x_var ="mean_reliable_misinfo", 
                 group_var = as.factor(df_wide_attention_check$attention_check_passed), 
                 bin_width = 1,
                 title = "Histogram of Reliable Misinformation Score By Attention Check",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_misinfo_score_by_attention_check")

create_histogram_grouped(data = df_wide_attention_check, 
                 x_var ="mean_reliable_nonmisinfo", 
                 group_var = as.factor(df_wide_attention_check$attention_check_passed), 
                 bin_width = 1,
                 title = "Histogram of Reliable Non-Misinformation Score By Attention Check",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_nonmisinfo_score_by_attention_check")

create_histogram_grouped(data = df_wide_attention_check, 
                 x_var ="reliability_discernment", 
                 group_var = as.factor(df_wide_attention_check$attention_check_passed), 
                 bin_width = 1,
                 title = "Histogram of Reliable Discernment By Attention Check",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_discernment_by_attention_check")

Manipulativeness score

create_histogram_grouped(data = df_wide_attention_check, 
                 x_var ="mean_manipulative_misinfo", 
                 group_var = as.factor(df_wide_attention_check$attention_check_passed), 
                 bin_width = 1,
                 title = "Histogram of Manipulation Misinformation Score By Attention Check",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_misinfo_score_by_attention_check")

create_histogram_grouped(data = df_wide_attention_check, 
                 x_var ="mean_manipulative_nonmisinfo", 
                 group_var = as.factor(df_wide_attention_check$attention_check_passed), 
                 bin_width = 1,
                 title = "Histogram of Manipulation Non-Misinformation Score By Attention Check",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_nonmisinfo_score_by_attention_check")

create_histogram_grouped(data = df_wide_attention_check, 
                 x_var ="manipulation_discernment", 
                 group_var = as.factor(df_wide_attention_check$attention_check_passed), 
                 bin_width = 1,
                 title = "Histogram of Manipulation Discernment By Attention Check",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulation_discernment_by_attention_check")

Sharing

create_histogram_grouped(data = df_wide_attention_check, 
                 x_var ="mean_share_misinfo", 
                 group_var = as.factor(df_wide_attention_check$attention_check_passed),
                 bin_width = 0.5,
                 title = "Histogram of Sharing Misinformation Score By Attention Check",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_misinfo_score_by_attention_check")

create_histogram_grouped(data = df_wide_attention_check, 
                 x_var ="mean_share_nonmisinfo", 
                 group_var = as.factor(df_wide_attention_check$attention_check_passed),
                 bin_width = 0.5,
                 title = "Histogram of Sharing Non-Misinformation Score By Attention Check",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_nonmisinfo_score_by_attention_check")

create_histogram_grouped(data = df_wide_attention_check, 
                 x_var ="sharing_discernment", 
                 group_var = as.factor(df_wide_attention_check$attention_check_passed),
                 bin_width = 0.5,
                 title = "Histogram of Sharing Discernment By Attention Check",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_discernment_by_attention_check")

Followup

By user

Reliable Score

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_reliable_misinfo_followup", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Reliable Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_misinfo_score_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_reliable_nonmisinfo_followup", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Reliable Non-Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_nonmisinfo_score_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="reliability_discernment_followup", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Reliable Discernment",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_discernment_followup")

Manipulativeness score

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_manipulative_misinfo_followup", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Manipulation Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_misinfo_score_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_manipulative_nonmisinfo_followup", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Manipulation Non-Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_nonmisinfo_score_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="manipulation_discernment_followup", 
                 group_var = NULL, 
                 bin_width = 1,
                 title = "Histogram of Manipulation Discernment",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulation_discernment_followup")

Sharing

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_share_misinfo_followup", 
                 group_var = NULL,
                 bin_width = 0.5,
                 title = "Histogram of Sharing Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename= "sharing_misinfo_score_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_share_nonmisinfo_followup", 
                 group_var = NULL,
                 bin_width = 0.5,
                 title = "Histogram of Sharing Non-Misinformation Score",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_nonmisinfo_score_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="sharing_discernment_followup", 
                 group_var = NULL,
                 bin_width = 0.5,
                 title = "Histogram of Sharing Discernment",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_discernment_followup")

By Phase

Reliable Score

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_reliable_misinfo_followup", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_misinfo_score_by_phase_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_reliable_nonmisinfo_followup", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Non-Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_nonmisinfo_score_by_phase_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="reliability_discernment_followup", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Discernment By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_discernment_by_phase_followup")

Manipulativeness score

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_manipulative_misinfo_followup", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_misinfo_score_by_phase_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_manipulative_nonmisinfo_followup", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Non-Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_nonmisinfo_score_by_phase_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="manipulation_discernment_followup", 
                 group_var = "phase_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Discernment By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulation_discernment_by_phase_followup")

Sharing

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_share_misinfo_followup", 
                 group_var = "phase_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_misinfo_score_by_phase_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_share_nonmisinfo_followup", 
                 group_var = "phase_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Non-Misinformation Score By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_nonmisinfo_score_by_phase_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="sharing_discernment_followup", 
                 group_var = "phase_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Discernment By Phase",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_discernment_by_phase_followup")

By Arm

Reliable Score

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_reliable_misinfo_followup", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_misinfo_score_by_arm_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_reliable_nonmisinfo_followup", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Non-Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_nonmisinfo_score_by_arm_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="reliability_discernment_followup", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Reliable Discernment By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "reliable_discernment_by_arm_followup")

Manipulativeness score

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_manipulative_misinfo_followup", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_misinfo_score_by_arm_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_manipulative_nonmisinfo_followup", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Non-Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulative_nonmisinfo_score_by_arm_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="manipulation_discernment_followup", 
                 group_var = "arm_coded", 
                 bin_width = 1,
                 title = "Histogram of Manipulation Discernment By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "manipulation_discernment_by_arm_followup")

Sharing

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_share_misinfo_followup", 
                 group_var = "arm_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_misinfo_score_by_arm_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="mean_share_nonmisinfo_followup", 
                 group_var = "arm_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Non-Misinformation Score By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_nonmisinfo_score_by_arm_followup")

create_histogram_grouped(data = df_wide_followup, 
                 x_var ="sharing_discernment_followup", 
                 group_var = "arm_coded",
                 bin_width = 0.5,
                 title = "Histogram of Sharing Discernment By Arm",
                 x_label = "Score", 
                 y_label = "Frequency",
                 file_basename = "sharing_discernment_by_arm_followup")

Analysis of Missing Values

options(scipen = 999)

create_aggr_plot_distribution <- function(data, colors = c('lavender', 'purple'), sort_vars = TRUE, axis_cex = 0.3, gap = 0.5, number_dec = 3) {
  # Generate the aggregate plot for missing data showing only the distribution
  aggr_plot <- aggr(data, col = colors, numbers = TRUE, sortVars = sort_vars, labels = names(data), 
                    cex.axis = axis_cex, gap = gap, which = c(1), number_dec = number_dec)
  
  # Return the plot object
  return(aggr_plot)
}

• We provided a distribution of missing values for user-level and post-level variables in the tables below

• Note that the variable count indeed represents the proportion of missing values in the first table of each set of tables

• For post-level variables, we expect there to be no missing values for any of the variables. However, there are a few observations in follow up we need to look at again.

User level variables

Consented participants

• We are currently having a total 70355 respondents who consented to the study

• The table below shows the number of missing values for each variable and their distribution

create_aggr_plot_distribution(df_wide)

## 
##  Variables sorted by number of missings: 
##                      Variable     Count
##        attention_check_passed 0.3917135
##            mean_share_misinfo 0.3915571
##         mean_reliable_misinfo 0.3915571
##     mean_manipulative_misinfo 0.3915571
##         mean_share_nonmisinfo 0.3915571
##      mean_reliable_nonmisinfo 0.3915571
##  mean_manipulative_nonmisinfo 0.3915571
##           sharing_discernment 0.3915571
##       reliability_discernment 0.3915571
##      manipulation_discernment 0.3915571
##         misinfo_quiz_duration 0.3749840
##                total_duration 0.3749129
##          misinfoQuiz_end_time 0.3749129
##   manipulative_reliable_order 0.2695899
##         intervention_duration 0.2599247
##        MisinfoQuiz_start_time 0.2598252
##                   analytic_id 0.0000000
##                   phase_coded 0.0000000
##                 country_coded 0.0000000
##          payment_amount_coded 0.0000000
##      payment_condition_dollar 0.0000000
##                     arm_coded 0.0000000
##        MisinfoChat_start_time 0.0000000
##         missing_treatment_arm 0.0000000
##                abnormal_posts 0.0000000
##             stage_reached_num 0.0000000
##            stage_reached_char 0.0000000
##             consent_coded_num 0.0000000
##                      subphase 0.0000000
##                        strata 0.0000000
##                quiz_completed 0.0000000
##          quiz_completed_coded 0.0000000
##                     phase_arm 0.0000000
##        time_since_first_start 0.0000000

## 
##  Missings in variables:
##                      Variable Count
##                total_duration 26377
##         intervention_duration 18287
##         misinfo_quiz_duration 26382
##          misinfoQuiz_end_time 26377
##        MisinfoQuiz_start_time 18280
##   manipulative_reliable_order 18967
##            mean_share_misinfo 27548
##         mean_reliable_misinfo 27548
##     mean_manipulative_misinfo 27548
##         mean_share_nonmisinfo 27548
##      mean_reliable_nonmisinfo 27548
##  mean_manipulative_nonmisinfo 27548
##           sharing_discernment 27548
##       reliability_discernment 27548
##      manipulation_discernment 27548
##        attention_check_passed 27559

Completed quiz participants

df_quiz_completed <- df_wide |> filter(quiz_completed == 1) 

quiz_completed <- as.character(nrow(df_quiz_completed))

create_aggr_plot_distribution(df_quiz_completed)

## 
##  Variables sorted by number of missings: 
##                      Variable        Count
##        attention_check_passed 0.0002569673
##         intervention_duration 0.0002336067
##         misinfo_quiz_duration 0.0001168033
##        MisinfoQuiz_start_time 0.0001168033
##                   analytic_id 0.0000000000
##                   phase_coded 0.0000000000
##                 country_coded 0.0000000000
##          payment_amount_coded 0.0000000000
##      payment_condition_dollar 0.0000000000
##                     arm_coded 0.0000000000
##        MisinfoChat_start_time 0.0000000000
##                total_duration 0.0000000000
##          misinfoQuiz_end_time 0.0000000000
##         missing_treatment_arm 0.0000000000
##                abnormal_posts 0.0000000000
##             stage_reached_num 0.0000000000
##            stage_reached_char 0.0000000000
##             consent_coded_num 0.0000000000
##   manipulative_reliable_order 0.0000000000
##                      subphase 0.0000000000
##                        strata 0.0000000000
##            mean_share_misinfo 0.0000000000
##         mean_reliable_misinfo 0.0000000000
##     mean_manipulative_misinfo 0.0000000000
##         mean_share_nonmisinfo 0.0000000000
##      mean_reliable_nonmisinfo 0.0000000000
##  mean_manipulative_nonmisinfo 0.0000000000
##           sharing_discernment 0.0000000000
##       reliability_discernment 0.0000000000
##      manipulation_discernment 0.0000000000
##                quiz_completed 0.0000000000
##          quiz_completed_coded 0.0000000000
##                     phase_arm 0.0000000000
##        time_since_first_start 0.0000000000

## 
##  Missings in variables:
##                Variable Count
##   intervention_duration    10
##   misinfo_quiz_duration     5
##  MisinfoQuiz_start_time     5
##  attention_check_passed    11

• We are currently having a total 42807 respondents who completed_quiz

Consented followup participants

• We are currently having a total 19950 respondents who consented to the follow up study

create_aggr_plot_distribution(df_wide_followup)

## 
##  Variables sorted by number of missings: 
##                                   Variable      Count
##  attention_check_passed_followup_coded_num 0.07137845
##          manipulation_discernment_followup 0.07132832
##      mean_manipulative_nonmisinfo_followup 0.07127820
##         mean_manipulative_misinfo_followup 0.07117794
##                mean_share_misinfo_followup 0.07112782
##             mean_reliable_misinfo_followup 0.07112782
##             mean_share_nonmisinfo_followup 0.07112782
##          mean_reliable_nonmisinfo_followup 0.07112782
##               sharing_discernment_followup 0.07112782
##           reliability_discernment_followup 0.07112782
##              misinfoQuiz_end_time_followup 0.07062657
##            duration_complete_quiz_followup 0.07062657
##            MisinfoQuiz_start_time_followup 0.01177945
##                                analytic_id 0.00000000
##                                  arm_coded 0.00000000
##                                phase_coded 0.00000000
##            MisinfoChat_start_time_followup 0.00000000
##          time_since_completed_intervention 0.00000000
##                  time_since_completed_quiz 0.00000000
##               user_payment_amount_followup 0.00000000
##                       followup_entry_point 0.00000000
##                      followup_reflection_1 0.00000000
##                      followup_reflection_2 0.00000000
##                      followup_reflection_3 0.00000000
##             abnormal_followup_reflection_1 0.00000000
##             abnormal_followup_reflection_2 0.00000000
##             abnormal_followup_reflection_3 0.00000000
##          completed_quiz_followup_coded_num 0.00000000
##                    quiz_completed_followup 0.00000000
##                                  phase_arm 0.00000000
##            time_since_first_start_followup 0.00000000

## 
##  Missings in variables:
##                                   Variable Count
##            MisinfoQuiz_start_time_followup   235
##              misinfoQuiz_end_time_followup  1409
##            duration_complete_quiz_followup  1409
##  attention_check_passed_followup_coded_num  1424
##                mean_share_misinfo_followup  1419
##             mean_reliable_misinfo_followup  1419
##         mean_manipulative_misinfo_followup  1420
##             mean_share_nonmisinfo_followup  1419
##          mean_reliable_nonmisinfo_followup  1419
##      mean_manipulative_nonmisinfo_followup  1422
##               sharing_discernment_followup  1419
##           reliability_discernment_followup  1419
##          manipulation_discernment_followup  1423

Completed followup quiz participants

df_quiz_completed_followup <- df_wide_followup |> filter(quiz_completed_followup == "TRUE")
n_quiz_completed_followup <- as.character(nrow(df_quiz_completed_followup))


create_aggr_plot_distribution(df_quiz_completed_followup)

## 
##  Variables sorted by number of missings: 
##                                   Variable         Count
##  attention_check_passed_followup_coded_num 0.00005397528
##                                analytic_id 0.00000000000
##                                  arm_coded 0.00000000000
##                                phase_coded 0.00000000000
##            MisinfoChat_start_time_followup 0.00000000000
##            MisinfoQuiz_start_time_followup 0.00000000000
##              misinfoQuiz_end_time_followup 0.00000000000
##            duration_complete_quiz_followup 0.00000000000
##          time_since_completed_intervention 0.00000000000
##                  time_since_completed_quiz 0.00000000000
##               user_payment_amount_followup 0.00000000000
##                       followup_entry_point 0.00000000000
##                      followup_reflection_1 0.00000000000
##                      followup_reflection_2 0.00000000000
##                      followup_reflection_3 0.00000000000
##             abnormal_followup_reflection_1 0.00000000000
##             abnormal_followup_reflection_2 0.00000000000
##             abnormal_followup_reflection_3 0.00000000000
##          completed_quiz_followup_coded_num 0.00000000000
##                mean_share_misinfo_followup 0.00000000000
##             mean_reliable_misinfo_followup 0.00000000000
##         mean_manipulative_misinfo_followup 0.00000000000
##             mean_share_nonmisinfo_followup 0.00000000000
##          mean_reliable_nonmisinfo_followup 0.00000000000
##      mean_manipulative_nonmisinfo_followup 0.00000000000
##               sharing_discernment_followup 0.00000000000
##           reliability_discernment_followup 0.00000000000
##          manipulation_discernment_followup 0.00000000000
##                    quiz_completed_followup 0.00000000000
##                                  phase_arm 0.00000000000
##            time_since_first_start_followup 0.00000000000

## 
##  Missings in variables:
##                                   Variable Count
##  attention_check_passed_followup_coded_num     1

• We are currently having a total 18527 respondents who completed_followup_quiz for the follow up study

User-post level

• In the cleaning scripts for the main survey and followup data, we have removed the rows with missing values for the user-post level variables. Therefore, there is no missing values for the user-post level variables.

Main Survey

create_aggr_plot_distribution(df_long)

## 
##  Variables sorted by number of missings: 
##                  Variable Count
##               analytic_id     0
##                   post_id     0
##               post_number     0
##                 post_type     0
##               share_score     0
##            reliable_score     0
##        manipulative_score     0
##               phase_coded     0
##  quiz_completed_coded_num     0
##                 arm_coded     0

## 
##  Missings in variables:
## [1] Variable Count   
## <0 rows> (or 0-length row.names)

Follow up

create_aggr_plot_distribution(df_long_followup)

## 
##  Variables sorted by number of missings: 
##                                   Variable         Count
##  attention_check_passed_followup_coded_num 0.00005396363
##                manipulative_score_followup 0.00004317090
##                                analytic_id 0.00000000000
##          completed_quiz_followup_coded_num 0.00000000000
##                       post_number_followup 0.00000000000
##                           post_id_followup 0.00000000000
##                    reliable_score_followup 0.00000000000
##                       share_score_followup 0.00000000000
##                         post_type_followup 0.00000000000
##                                is_followup 0.00000000000

## 
##  Missings in variables:
##                                   Variable Count
##  attention_check_passed_followup_coded_num     5
##                manipulative_score_followup     4