RCT_CodeBook
Ishan Gupta
2024-07-03
Introduction
This document describes the data used while analyzing the impact of Shamiri’s intervention through an RCT on the mental health status of students captured by various pre-determined scales in the field of psychology.
These scales include: PCS - Perceived Control Scale (6 questions, Academic and Staying out of trouble) PHQ - Patient Health Questionaire (8 questions + Functioning) GAD - Generalised Anxiety Disorder (7 questions + Check + Functioning) MSPSS - Multidimensional Scale of Perceived Social Support (8 questions) GQ - Gratitude Questionaire (6 questions) SWEMBS - Short Warwick Edinburgh Mental Well Being Scale (7 questions) SES - School Engagement Scale (19 questions)
The document is divided into four parts.
- The first part checks for the missing values across the data set in both treatment and control groups.
- The second part lays out the descriptive analysis of each variable in the demographic details for both treatment and control group
- The third part checks for trends in responses to each of the questions in the scales mentioned above.
- Work in Progress: The fourth part does correlational analysis of each of the scales with demographic information of the participants such as age, gender, form, perceived, area type (urban/rural), household situation and school metrics (academic performance and co-curricular engageement). This will help us understand the implication of the correlations toward our analysis in the RCT.
First we load the libraries as well as the data sets to perform basic cleaning before we begin the analysis.
# Installing essential libraries
library(stargazer)##
## Please cite as:## Hlavac, Marek (2022). stargazer: Well-Formatted Regression and Summary Statistics Tables.## R package version 5.2.3. https://CRAN.R-project.org/package=stargazerlibrary(readxl)
library(fastDummies)## Thank you for using fastDummies!## To acknowledge our work, please cite the package:## Kaplan, J. & Schlegel, B. (2023). fastDummies: Fast Creation of Dummy (Binary) Columns and Rows from Categorical Variables. Version 1.7.1. URL: https://github.com/jacobkap/fastDummies, https://jacobkap.github.io/fastDummies/.library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(knitr)
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ forcats 1.0.0 ✔ readr 2.1.4
## ✔ ggplot2 3.4.4 ✔ stringr 1.5.0
## ✔ lubridate 1.9.2 ✔ tibble 3.2.1
## ✔ purrr 1.0.2 ✔ tidyr 1.3.0## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(lmtest)## Loading required package: zoo
##
## Attaching package: 'zoo'
##
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numericlibrary(AER)## Loading required package: car
## Loading required package: carData
##
## Attaching package: 'car'
##
## The following object is masked from 'package:purrr':
##
## some
##
## The following object is masked from 'package:dplyr':
##
## recode
##
## Loading required package: sandwich
## Loading required package: survivallibrary(sandwich)
library(fixest)
library(ggplot2)
library(ggthemes)
library(rddensity)
library(GGally)## Registered S3 method overwritten by 'GGally':
## method from
## +.gg ggplot2library(corrplot)## corrplot 0.92 loadedlibrary(reshape2)##
## Attaching package: 'reshape2'
##
## The following object is masked from 'package:tidyr':
##
## smithslibrary(kableExtra)##
## Attaching package: 'kableExtra'
##
## The following object is masked from 'package:dplyr':
##
## group_rowslibrary(MissMech)
library(gridExtra)##
## Attaching package: 'gridExtra'
##
## The following object is masked from 'package:dplyr':
##
## combinelibrary(grid)
# Setting the working directory
setwd('/Users/ishangupta/Desktop/Books for Masters/R Practice')
# Loading the data in R
treatment <- read_xlsx("Treatment.xlsx", sheet = "2023Studentlongtrialdata")
control <- read_xlsx("Control.xlsx", sheet = "2023NaiKiambuControl")
# Renaming the columns in the treatment group
treatment <- treatment %>%
rename(
Age = `student age`,
Form = form,
Gender = `student gender`,
School_ID = `school id`
) %>%
mutate(trt = 1)
# List of school IDs to be dropped
school_ids_to_drop <- c("ANS23_School_31", "ANS23_School_32", "ANS23_School_33",
"ANS23_School_34", "ANS23_School_35", "ANS23_School_36",
"ANS23_School_37")
# Assuming 'treatment' is your dataframe with the treatment group data
# Filtering out rows with specified school IDs
treatment <- treatment %>%
filter(!School_ID %in% school_ids_to_drop)
# Renaming the columns in the control group
control <- control %>%
rename(
School_ID = `school id`
) %>%
mutate(supervisor_id = "NA") %>%
mutate(hub = "NA") %>%
mutate(fellow_gender = "NA") %>%
mutate(fellow_age = "NA") %>%
mutate(trt = 0)
# Creating the RCT data set
rct <- rbind(treatment,control)
# Converting the columns into numeric
cols_to_numeric <- c(1, 11, 12, 13, 103, 104, 17:101)
rct[cols_to_numeric] <- lapply(rct[cols_to_numeric], as.numeric)## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercion
## Warning in lapply(rct[cols_to_numeric], as.numeric): NAs introduced by coercionPART 1: Checking for missingness
# Define a function to count NAs in a row
count_NAs <- function(row) {
sum(is.na(row)) + sum(row == "NA", na.rm = TRUE)
}
# Apply the function to the relevant columns (19 to 79) and create the new column NA_entries
rct$NA_entries <- apply(rct[, 27:90], 1, count_NAs)
# Checking for trends in NA entries for each time
rct_small <- rct %>%
group_by(trt,participant_id,time, NA_entries) %>%
summarise(count=n())## `summarise()` has grouped output by 'trt', 'participant_id', 'time'. You can
## override using the `.groups` argument.# Pivot the data to wide format
rct_wide <- rct_small %>%
pivot_wider(names_from = time, values_from = NA_entries, names_prefix = "NA_entries_in_time_")
# Rename the columns to match the desired output
colnames(rct_wide) <- gsub("NA_entries_in_time_", "NA_entries_in_time_", colnames(rct_wide))
# Checking for absence of students in each data collection period.
rct_wide <- rct_wide %>%
select(-count) %>%
mutate(cases = case_when(
NA_entries_in_time_0 == 64 & NA_entries_in_time_2 == 64 & NA_entries_in_time_4 == 64 & NA_entries_in_time_8 == 64 ~ "Absent in ALL",
NA_entries_in_time_2 == 64 & NA_entries_in_time_4 == 64 & NA_entries_in_time_8 == 64 ~ "Absent in 2, 4, 8",
NA_entries_in_time_0 == 64 & NA_entries_in_time_4 == 64 & NA_entries_in_time_8 == 64 ~ "Absent in 0, 4, 8",
NA_entries_in_time_0 == 64 & NA_entries_in_time_2 == 64 & NA_entries_in_time_8 == 64 ~ "Absent in 0, 2, 8",
NA_entries_in_time_0 == 64 & NA_entries_in_time_2 == 64 & NA_entries_in_time_4 == 64 ~ "Absent in 0, 2, 4",
NA_entries_in_time_4 == 64 & NA_entries_in_time_8 == 64 ~ "Absent in 4, 8",
NA_entries_in_time_2 == 64 & NA_entries_in_time_8 == 64 ~ "Absent in 2, 8",
NA_entries_in_time_2 == 64 & NA_entries_in_time_4 == 64 ~ "Absent in 2, 4",
NA_entries_in_time_0 == 64 & NA_entries_in_time_8 == 64 ~ "Absent in 0, 8",
NA_entries_in_time_0 == 64 & NA_entries_in_time_4 == 64 ~ "Absent in 0, 4",
NA_entries_in_time_0 == 64 & NA_entries_in_time_2 == 64 ~ "Absent in 0, 2",
NA_entries_in_time_0 == 64 ~ "Absent in 0",
NA_entries_in_time_2 == 64 ~ "Absent in 2",
NA_entries_in_time_4 == 64 ~ "Absent in 4",
NA_entries_in_time_8 == 64 ~ "Absent in 8",
TRUE ~ "Present" # Default case when none of the above conditions are met
))
# Summarize the count of cases by trt and cases
rct_present <- rct_wide %>%
group_by(trt, cases) %>%
summarise(count = n(), .groups = 'drop')
# Summarize the total count of entries for trt = 0 and trt = 1
total_count_trt <- rct_present %>%
group_by(trt) %>%
summarise(total_count = sum(count))
# Calculating the count of students answering the questions in each scales
rct <- rct %>%
mutate(
PCS_count = rowSums(!is.na(select(., PCS_1:PCS_6))),
PHQ_count = rowSums(!is.na(select(., PHQ_1:PHQ_8))),
GAD_count = rowSums(!is.na(select(., GAD_1:GAD_7))),
MSPSS_count = rowSums(!is.na(select(., MSPSS_1:MSPSS_8))),
GQ_count = rowSums(!is.na(select(., GQ_1:GQ_6))),
SWEMWBS_count = rowSums(!is.na(select(., SWEMWBS_1:SWEMWBS_7))),
SES_count = rowSums(!is.na(select(., SES_1:SES_19)))
)
# Shortening the dataset to perform further analysis
rct_tiny <- rct %>%
select(trt,participant_id,PCS_count,PHQ_count,GAD_count,MSPSS_count,GQ_count,SWEMWBS_count,SES_count)
# Calculating the proportion of students in treatment and control group answering the survey as a whole
proportions_df <- rct_tiny %>%
group_by(trt) %>%
summarise(
PCS_proportion = mean(PCS_count > 0, na.rm = TRUE),
PHQ_proportion = mean(PHQ_count > 0, na.rm = TRUE),
GAD_proportion = mean(GAD_count > 0, na.rm = TRUE),
MSPSS_proportion = mean(MSPSS_count > 0, na.rm = TRUE),
GQ_proportion = mean(GQ_count > 0, na.rm = TRUE),
SWEMWBS_proportion = mean(SWEMWBS_count > 0, na.rm = TRUE),
SES_proportion = mean(SES_count > 0, na.rm = TRUE)
)# Step 1: Create a duplicate dataset with the necessary columns
phq_gad_columns <- c("time", "participant_id", paste0("PHQ_", 1:8), "PHQ_Functioning", paste0("GAD_", 1:7), "GAD_Check", "GAD_Functioning")
duplicate_dataset <- treatment %>%
select(all_of(phq_gad_columns))
# Step 2: Trim spaces and replace "NA" strings with actual NA values
duplicate_dataset <- duplicate_dataset %>%
mutate(across(c(paste0("PHQ_", 1:8), "PHQ_Functioning", paste0("GAD_", 1:7), "GAD_Check", "GAD_Functioning"), ~ str_trim(.))) %>%
mutate(across(c(paste0("PHQ_", 1:8), "PHQ_Functioning", paste0("GAD_", 1:7), "GAD_Check", "GAD_Functioning"), ~ na_if(., "NA"))) %>%
mutate(across(c(paste0("PHQ_", 1:8), "PHQ_Functioning", paste0("GAD_", 1:7), "GAD_Check", "GAD_Functioning"), ~ na_if(., "")))
# Step 3: Count the number of missing entries in the PHQ and GAD columns for each row
duplicate_dataset <- duplicate_dataset %>%
rowwise() %>%
mutate(num_missing_entries = sum(
is.na(`PHQ_1`) | `PHQ_1` == "NA", is.na(`PHQ_2`) | `PHQ_2` == "NA", is.na(`PHQ_3`) | `PHQ_3` == "NA",
is.na(`PHQ_4`) | `PHQ_4` == "NA", is.na(`PHQ_5`) | `PHQ_5` == "NA", is.na(`PHQ_6`) | `PHQ_6` == "NA",
is.na(`PHQ_7`) | `PHQ_7` == "NA", is.na(`PHQ_8`) | `PHQ_8` == "NA", is.na(`PHQ_Functioning`) | `PHQ_Functioning` == "NA",
is.na(`GAD_1`) | `GAD_1` == "NA", is.na(`GAD_2`) | `GAD_2` == "NA", is.na(`GAD_3`) | `GAD_3` == "NA",
is.na(`GAD_4`) | `GAD_4` == "NA", is.na(`GAD_5`) | `GAD_5` == "NA", is.na(`GAD_6`) | `GAD_6` == "NA",
is.na(`GAD_7`) | `GAD_7` == "NA", is.na(`GAD_Check`) | `GAD_Check` == "NA", is.na(`GAD_Functioning`) | `GAD_Functioning` == "NA"
))
# Step 4: Calculate the percentage of missingness
duplicate_dataset <- duplicate_dataset %>%
mutate(missingness_percentage = (num_missing_entries / 18) * 100) %>%
ungroup()
# Step 5: Check the frequency of each missing value per time.
check_missing <- duplicate_dataset %>%
group_by(time, num_missing_entries) %>%
summarise(freq_miss = n())## `summarise()` has grouped output by 'time'. You can override using the
## `.groups` argument. stargazer(check_missing, type = "text", summary = FALSE, rownames = FALSE)##
## ==================================
## time num_missing_entries freq_miss
## ----------------------------------
## 0 0 496
## 0 1 184
## 0 2 91
## 0 3 38
## 0 4 12
## 0 5 8
## 0 6 5
## 0 7 2
## 0 8 2
## 0 9 9
## 0 10 1
## 0 11 2
## 0 12 2
## 0 13 1
## 0 14 1
## 0 15 4
## 0 18 8
## 2 0 362
## 2 1 126
## 2 2 107
## 2 3 17
## 2 4 10
## 2 5 4
## 2 8 3
## 2 9 4
## 2 10 1
## 2 11 1
## 2 14 4
## 2 15 1
## 2 17 2
## 2 18 224
## 4 0 297
## 4 1 114
## 4 2 115
## 4 3 21
## 4 4 7
## 4 5 7
## 4 6 1
## 4 7 2
## 4 8 3
## 4 11 5
## 4 14 2
## 4 15 1
## 4 16 1
## 4 17 2
## 4 18 288
## 8 0 511
## 8 1 89
## 8 2 71
## 8 3 14
## 8 4 6
## 8 5 1
## 8 6 1
## 8 7 1
## 8 8 1
## 8 9 1
## 8 10 3
## 8 12 1
## 8 13 1
## 8 14 1
## 8 18 164
## ----------------------------------# List of columns to convert
columns_to_convert <- c("PHQ_1", "PHQ_2", "PHQ_3", "PHQ_4", "PHQ_5", "PHQ_6", "PHQ_7", "PHQ_8",
"GAD_1", "GAD_2", "GAD_3", "GAD_4", "GAD_5", "GAD_6", "GAD_7",
"GAD_Check", "PHQ_Functioning", "GAD_Functioning")
# Convert the specified columns from character to numeric
duplicate_dataset[columns_to_convert] <- lapply(duplicate_dataset[columns_to_convert], function(x) as.numeric(as.character(x)))Part 2: Demographic Analysis**
- Participant ID, time, group number
This is a unique identifier of each participant. The variable on time is indicative of when the data was collected for which participant. The group number is the unique identifier for the group that was assigned to the students in treatment. We want to check for missing entries if any. We also want to check if there was any attrition with respect to time. We also want to check the number of students in each group.
# Checking for total number of entries
total_stu <- nrow(rct)
unique_stu <- length(unique(rct$participant_id))
missing_stu <- sum(is.na(rct$participant_id))
# Creating a data frame to store the results
summary_stu <- data.frame(
Metric = c("Total Entries", "Unique Entries", "Missing Entries"),
Value = c(total_stu, unique_stu, missing_stu)
)
# Printing the summary table
stargazer(summary_stu, type = "text", summary = FALSE, rownames = FALSE)##
## =====================
## Metric Value
## ---------------------
## Total Entries 6,864
## Unique Entries 1,716
## Missing Entries 0
## ---------------------time_stu <- rct %>%
group_by(trt,time) %>%
summarise(no_of_students = n_distinct(participant_id))## `summarise()` has grouped output by 'trt'. You can override using the `.groups`
## argument.stargazer(time_stu, type = "text", summary = FALSE, rownames = FALSE)##
## =======================
## trt time no_of_students
## -----------------------
## 0 0 850
## 0 2 850
## 0 4 850
## 0 8 850
## 1 0 866
## 1 2 866
## 1 4 866
## 1 8 866
## -----------------------We can note the following:
- Every student has a unique participant ID
- Every time period of data collection has equal number of students. After confirming with the team we found that the demographic information for each participant in the treatment and control group was carried forward regardless of them being actually present in the session and the subsequent survey - therefore this does not speak to the attrition or absenteeism
- Condition, Implementer and School ID
# Checking for entries for condition and implementer
condition_stu <- rct %>%
group_by(trt,time, condition) %>%
summarise(no_of_students = n())## `summarise()` has grouped output by 'trt', 'time'. You can override using the
## `.groups` argument.condition_stu_wide <- condition_stu %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(condition_stu_wide, type = "text", summary = FALSE, rownames = FALSE)##
## =============================
## trt condition 0 2 4 8
## -----------------------------
## 0 Control1 850 850 850 850
## 1 Shamiri 866 866 866 866
## -----------------------------imp_stu <- rct %>%
group_by(trt,time, implementer) %>%
summarise(no_of_students = n())## `summarise()` has grouped output by 'trt', 'time'. You can override using the
## `.groups` argument.imp_stu_wide <- imp_stu %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(imp_stu_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ===============================
## trt implementer 0 2 4 8
## -------------------------------
## 0 Imp_1 850 850 850 850
## 1 Imp_1 866 866 866 866
## -------------------------------school_ID <- rct %>%
group_by(trt,time, School_ID) %>%
summarise(no_of_students = n())## `summarise()` has grouped output by 'trt', 'time'. You can override using the
## `.groups` argument.school_ID_wide <- school_ID %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(school_ID_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ===================================
## trt School_ID 0 2 4 8
## -----------------------------------
## 0 ANS23_School_17 174 174 174 174
## 0 ANS23_School_18 133 133 133 133
## 0 ANS23_School_25 137 137 137 137
## 0 ANS23_School_47 94 94 94 94
## 0 ANS23_School_5 227 227 227 227
## 0 ANS23_School_7 85 85 85 85
## 1 ANS23_School_17 138 138 138 138
## 1 ANS23_School_18 134 134 134 134
## 1 ANS23_School_25 135 135 135 135
## 1 ANS23_School_47 105 105 105 105
## 1 ANS23_School_5 224 224 224 224
## 1 ANS23_School_7 130 130 130 130
## -----------------------------------school_group <- rct %>%
group_by(trt,time, School_ID,group_number) %>%
summarise(no_of_students = n())## `summarise()` has grouped output by 'trt', 'time', 'School_ID'. You can
## override using the `.groups` argument.school_group_wide <- school_group %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(school_group_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ================================================
## trt School_ID group_number 0 2 4 8
## ------------------------------------------------
## 0 ANS23_School_17 3-3-C004 174 174 174 174
## 0 ANS23_School_18 2-2-C003 133 133 133 133
## 0 ANS23_School_25 1-1-C002 137 137 137 137
## 0 ANS23_School_47 6-6-C009 94 94 94 94
## 0 ANS23_School_5 5-5-C007 227 227 227 227
## 0 ANS23_School_7 4-4-C005 85 85 85 85
## 1 ANS23_School_17 17_S001 15 15 15 15
## 1 ANS23_School_17 17_S002 18 18 18 18
## 1 ANS23_School_17 17_S003 15 15 15 15
## 1 ANS23_School_17 17_S004 15 15 15 15
## 1 ANS23_School_17 17_S005 14 14 14 14
## 1 ANS23_School_17 17_S006 15 15 15 15
## 1 ANS23_School_17 17_S007 15 15 15 15
## 1 ANS23_School_17 17_S008 15 15 15 15
## 1 ANS23_School_17 17_S009 15 15 15 15
## 1 ANS23_School_17 1 1 1 1
## 1 ANS23_School_18 18_S001 15 15 15 15
## 1 ANS23_School_18 18_S002 15 15 15 15
## 1 ANS23_School_18 18_S003 15 15 15 15
## 1 ANS23_School_18 18_S004 15 15 15 15
## 1 ANS23_School_18 18_S005 14 14 14 14
## 1 ANS23_School_18 18_S006 15 15 15 15
## 1 ANS23_School_18 18_S007 15 15 15 15
## 1 ANS23_School_18 18_S008 15 15 15 15
## 1 ANS23_School_18 18_S009 15 15 15 15
## 1 ANS23_School_25 25_S001 15 15 15 15
## 1 ANS23_School_25 25_S002 15 15 15 15
## 1 ANS23_School_25 25_S003 15 15 15 15
## 1 ANS23_School_25 25_S004 15 15 15 15
## 1 ANS23_School_25 25_S005 15 15 15 15
## 1 ANS23_School_25 25_S006 15 15 15 15
## 1 ANS23_School_25 25_S007 15 15 15 15
## 1 ANS23_School_25 25_S008 15 15 15 15
## 1 ANS23_School_25 25_S009 15 15 15 15
## 1 ANS23_School_47 47_S001 15 15 15 15
## 1 ANS23_School_47 47_S002 15 15 15 15
## 1 ANS23_School_47 47_S003 15 15 15 15
## 1 ANS23_School_47 47_S004 15 15 15 15
## 1 ANS23_School_47 47_S005 15 15 15 15
## 1 ANS23_School_47 47_S006 15 15 15 15
## 1 ANS23_School_47 47_S007 15 15 15 15
## 1 ANS23_School_5 5_S001 15 15 15 15
## 1 ANS23_School_5 5_S002 14 14 14 14
## 1 ANS23_School_5 5_S003 16 16 16 16
## 1 ANS23_School_5 5_S004 15 15 15 15
## 1 ANS23_School_5 5_S005 15 15 15 15
## 1 ANS23_School_5 5_S006 15 15 15 15
## 1 ANS23_School_5 5_S007 15 15 15 15
## 1 ANS23_School_5 5_S008 15 15 15 15
## 1 ANS23_School_5 5_S009 15 15 15 15
## 1 ANS23_School_5 5_S010 15 15 15 15
## 1 ANS23_School_5 5_S011 15 15 15 15
## 1 ANS23_School_5 5_S012 15 15 15 15
## 1 ANS23_School_5 5_S013 15 15 15 15
## 1 ANS23_School_5 5_S014 15 15 15 15
## 1 ANS23_School_5 5_S015 14 14 14 14
## 1 ANS23_School_7 7_S001 13 13 13 13
## 1 ANS23_School_7 7_S002 15 15 15 15
## 1 ANS23_School_7 7_S003 16 16 16 16
## 1 ANS23_School_7 7_S004 15 15 15 15
## 1 ANS23_School_7 7_S005 15 15 15 15
## 1 ANS23_School_7 7_S006 15 15 15 15
## 1 ANS23_School_7 7_S007 12 12 12 12
## 1 ANS23_School_7 7_S008 14 14 14 14
## 1 ANS23_School_7 7_S009 15 15 15 15
## ------------------------------------------------We can note the following important points: - The number of groups in the control group are less - there’s only one control group for each school which makes sense as the purpose of those groups is only to collect data. But there are multiple groups within the treatment schools to possibly balance the fellow-to-student ratio
- Fellow ID, Supervisor ID, Hub, Fellow Gender, and Fellow Age
We want to check for the following things: - How many students under each fellow and supervisor, hub. - Gender and age wise distribution of students under the fellow
# Checking for entries for condition and implementer
stu_hub <- rct %>%
group_by(hub, time) %>%
summarise(no_of_students = n_distinct(participant_id),
no_of_supervisors = n_distinct(supervisor_id),
no_of_fellows = n_distinct(fellow_id)
)## `summarise()` has grouped output by 'hub'. You can override using the `.groups`
## argument.# Display the table using stargazer
stargazer(stu_hub, type = "text", summary = FALSE, rownames = FALSE)##
## ==============================================================
## hub time no_of_students no_of_supervisors no_of_fellows
## --------------------------------------------------------------
## Kariobangi 0 377 6 23
## Kariobangi 2 377 6 23
## Kariobangi 4 377 6 23
## Kariobangi 8 377 6 23
## Kasarani 0 224 3 15
## Kasarani 2 224 3 15
## Kasarani 4 224 3 15
## Kasarani 8 224 3 15
## Kiambu 0 135 1 9
## Kiambu 2 135 1 9
## Kiambu 4 135 1 9
## Kiambu 8 135 1 9
## NA 0 850 1 1
## NA 2 850 1 1
## NA 4 850 1 1
## NA 8 850 1 1
## Ruiru 0 130 1 9
## Ruiru 2 130 1 9
## Ruiru 4 130 1 9
## Ruiru 8 130 1 9
## --------------------------------------------------------------stu_fel <- rct %>%
group_by(fellow_id, time) %>%
summarise(no_of_students = n())## `summarise()` has grouped output by 'fellow_id'. You can override using the
## `.groups` argument.stu_fel_wide <- stu_fel %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_fel_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ================================
## fellow_id 0 2 4 8
## --------------------------------
## SHAMIRI RESEARCH 850 850 850 850
## TFW23_S_031 16 16 16 16
## TFW23_S_032 15 15 15 15
## TFW23_S_034 14 14 14 14
## TFW23_S_036 15 15 15 15
## TFW23_S_040 15 15 15 15
## TFW23_S_044 15 15 15 15
## TFW23_S_046 15 15 15 15
## TFW23_S_048 15 15 15 15
## TFW23_S_049 15 15 15 15
## TFW23_S_050 14 14 14 14
## TFW23_S_061 15 15 15 15
## TFW23_S_064 15 15 15 15
## TFW23_S_065 15 15 15 15
## TFW23_S_067 15 15 15 15
## TFW23_S_069 15 15 15 15
## TFW23_S_074 15 15 15 15
## TFW23_S_075 29 29 29 29
## TFW23_S_077 15 15 15 15
## TFW23_S_081 15 15 15 15
## TFW23_S_083 15 15 15 15
## TFW23_S_084 15 15 15 15
## TFW23_S_085 15 15 15 15
## TFW23_S_086 15 15 15 15
## TFW23_S_090 15 15 15 15
## TFW23_S_092 15 15 15 15
## TFW23_S_094 15 15 15 15
## TFW23_S_095 15 15 15 15
## TFW23_S_099 15 15 15 15
## TFW23_S_100 18 18 18 18
## TFW23_S_101 15 15 15 15
## TFW23_S_102 15 15 15 15
## TFW23_S_107 15 15 15 15
## TFW23_S_108 15 15 15 15
## TFW23_S_111 15 15 15 15
## TFW23_S_114 30 30 30 30
## TFW23_S_115 15 15 15 15
## TFW23_S_117 15 15 15 15
## TFW23_S_141 15 15 15 15
## TFW23_S_232 15 15 15 15
## TFW23_S_234 15 15 15 15
## TFW23_S_235 15 15 15 15
## TFW23_S_236 15 15 15 15
## TFW23_S_237 15 15 15 15
## TFW23_S_238 15 15 15 15
## TFW23_S_239 15 15 15 15
## TFW23_S_240 15 15 15 15
## TFW23_S_241 15 15 15 15
## TFW23_S_312 12 12 12 12
## TFW23_S_313 15 15 15 15
## TFW23_S_314 16 16 16 16
## TFW23_S_316 15 15 15 15
## TFW23_S_317 14 14 14 14
## TFW23_S_318 15 15 15 15
## TFW23_S_319 15 15 15 15
## TFW23_S_320 13 13 13 13
## TFW23_S_321 15 15 15 15
## --------------------------------stu_sup <- rct %>%
group_by(supervisor_id, time) %>%
summarise(no_of_students = n())## `summarise()` has grouped output by 'supervisor_id'. You can override using the
## `.groups` argument.stu_sup_wide <- stu_sup %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_sup_wide, type = "text", summary = FALSE, rownames = FALSE)##
## =============================
## supervisor_id 0 2 4 8
## -----------------------------
## NA 850 850 850 850
## SPV23_S_03 130 130 130 130
## SPV23_S_11 75 75 75 75
## SPV23_S_13 75 75 75 75
## SPV23_S_16 135 135 135 135
## SPV23_S_17 93 93 93 93
## SPV23_S_18 15 15 15 15
## SPV23_S_20 60 60 60 60
## SPV23_S_25 75 75 75 75
## SPV23_S_30 75 75 75 75
## SPV23_S_38 59 59 59 59
## SPV23_S_40 74 74 74 74
## -----------------------------fel_gen <- rct %>%
group_by(fellow_gender, time) %>%
summarise(no_of_students = n())## `summarise()` has grouped output by 'fellow_gender'. You can override using the
## `.groups` argument.fel_gen_wide <- fel_gen %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(fel_gen_wide, type = "text", summary = FALSE, rownames = FALSE)##
## =============================
## fellow_gender 0 2 4 8
## -----------------------------
## F 402 402 402 402
## M 464 464 464 464
## NA 850 850 850 850
## -----------------------------fel_age <- rct %>%
group_by(fellow_age, time) %>%
summarise(no_of_students = n())## `summarise()` has grouped output by 'fellow_age'. You can override using the
## `.groups` argument.fel_age_wide <- fel_age %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(fel_age_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ==========================
## fellow_age 0 2 4 8
## --------------------------
## 18 87 87 87 87
## 19 124 124 124 124
## 20 45 45 45 45
## 21 194 194 194 194
## 22 328 328 328 328
## 23 59 59 59 59
## 24 29 29 29 29
## NA 850 850 850 850
## --------------------------We can note the following: - The total number of fellows are 75 in the RCT treatment group in comparison to the 400 fellows who we have data on - Similarly the total number of supervisors are 15, in comparison with the 40 supervisors who we have data on - Most fellows have 60 students under them. However, the number of students can be as high as 172 for one of the fellows. - Male fellows have more students than female fellows - The maximum number of students in the treatment group are with fellows who are of age 21-22. - The NAs in all the tables above represent the entries from the control group as they didn’t have the fellows assigned.
- Student age, Student form, student gender, tribe and county
# Checking for entries for condition and implementer
stu_age <- rct %>%
group_by(trt,Age, Form) %>%
summarise(no_of_students = n_distinct(participant_id))## `summarise()` has grouped output by 'trt', 'Age'. You can override using the
## `.groups` argument.stargazer(stu_age, type = "text", summary = FALSE, rownames = FALSE)##
## ===========================
## trt Age Form no_of_students
## ---------------------------
## 0 12 1 1
## 0 13 1 12
## 0 13 NA 1
## 0 14 1 70
## 0 14 2 19
## 0 14 3 1
## 0 14 NA 1
## 0 15 1 105
## 0 15 2 87
## 0 15 3 5
## 0 15 4 2
## 0 15 NA 1
## 0 16 1 34
## 0 16 2 123
## 0 16 3 59
## 0 16 4 2
## 0 16 NA 4
## 0 17 1 6
## 0 17 2 38
## 0 17 3 79
## 0 17 4 25
## 0 17 NA 5
## 0 18 1 3
## 0 18 2 10
## 0 18 3 31
## 0 18 4 40
## 0 18 NA 3
## 0 19 1 1
## 0 19 2 2
## 0 19 3 12
## 0 19 4 12
## 0 20 3 1
## 0 20 4 5
## 0 22 2 1
## 0 22 4 3
## 0 NA 1 12
## 0 NA 2 9
## 0 NA 3 7
## 0 NA 4 3
## 0 NA NA 15
## 1 13 1 15
## 1 13 2 1
## 1 13 3 1
## 1 14 1 89
## 1 14 2 20
## 1 14 4 1
## 1 14 NA 2
## 1 15 1 121
## 1 15 2 81
## 1 15 3 5
## 1 15 NA 2
## 1 16 1 65
## 1 16 2 84
## 1 16 3 55
## 1 17 1 13
## 1 17 2 41
## 1 17 3 109
## 1 17 4 17
## 1 17 NA 4
## 1 18 1 3
## 1 18 2 4
## 1 18 3 22
## 1 18 4 37
## 1 18 NA 1
## 1 19 2 2
## 1 19 3 8
## 1 19 4 9
## 1 20 3 1
## 1 20 4 3
## 1 NA 1 17
## 1 NA 2 6
## 1 NA 3 10
## 1 NA NA 17
## ---------------------------stu_age_gen <- rct %>%
group_by(trt,Age,Gender) %>%
summarise(no_of_students = n_distinct(participant_id))## `summarise()` has grouped output by 'trt', 'Age'. You can override using the
## `.groups` argument.stargazer(stu_age_gen, type = "text", summary = FALSE, rownames = FALSE)##
## =============================
## trt Age Gender no_of_students
## -----------------------------
## 0 12 Male 1
## 0 13 Female 5
## 0 13 Male 8
## 0 14 Female 43
## 0 14 Male 47
## 0 14 NA 1
## 0 15 Female 97
## 0 15 Male 102
## 0 15 NA 1
## 0 16 Female 123
## 0 16 Male 98
## 0 16 NA 1
## 0 17 Female 75
## 0 17 Male 73
## 0 17 NA 5
## 0 18 Female 46
## 0 18 Male 38
## 0 18 NA 3
## 0 19 Female 11
## 0 19 Male 16
## 0 20 Female 2
## 0 20 Male 4
## 0 22 Female 2
## 0 22 Male 1
## 0 22 NA 1
## 0 NA Female 13
## 0 NA Male 17
## 0 NA NA 16
## 1 13 Female 5
## 1 13 Male 12
## 1 14 Female 55
## 1 14 Male 57
## 1 15 Female 108
## 1 15 Male 97
## 1 15 NA 4
## 1 16 Female 99
## 1 16 Male 102
## 1 16 NA 3
## 1 17 Female 92
## 1 17 Male 89
## 1 17 NA 3
## 1 18 Female 32
## 1 18 Male 34
## 1 18 NA 1
## 1 19 Female 8
## 1 19 Male 11
## 1 20 Female 2
## 1 20 Male 2
## 1 NA Female 18
## 1 NA Male 13
## 1 NA NA 19
## -----------------------------stu_gen <- rct %>%
group_by(Gender, Form) %>%
summarise(no_of_students = n_distinct(participant_id))## `summarise()` has grouped output by 'Gender'. You can override using the
## `.groups` argument.stargazer(stu_gen, type = "text", summary = FALSE, rownames = FALSE)##
## ==========================
## Gender Form no_of_students
## --------------------------
## Female 1 233
## Female 2 278
## Female 3 207
## Female 4 113
## Female NA 5
## Male 1 327
## Male 2 245
## Male 3 194
## Male 4 45
## Male NA 11
## NA 1 7
## NA 2 5
## NA 3 5
## NA 4 1
## NA NA 40
## --------------------------stu_tribe <- rct %>%
group_by(trt,Tribe) %>%
summarise(no_of_students = n_distinct(participant_id))## `summarise()` has grouped output by 'trt'. You can override using the `.groups`
## argument.# Print the summary table with stargazer
stargazer(stu_tribe, type = "text", summary = FALSE, rownames = FALSE)##
## ===========================================
## trt Tribe no_of_students
## -------------------------------------------
## 0 Abagusii 2
## 0 Arab 1
## 0 Bantu 2
## 0 Borana 8
## 0 Burji 5
## 0 Chirakinandi Dr Congo 1
## 0 Congo 1
## 0 Congolese 2
## 0 Digo Kisii 1
## 0 Embu 2
## 0 Gaatta 1
## 0 Gabra 1
## 0 Giriama 4
## 0 Giriama Mijikenda 1
## 0 Half Cast 1
## 0 Half Of Kikuyu And Kisii 2
## 0 Islamic 1
## 0 Kalenjin 10
## 0 Kamba 90
## 0 Kambakikuyu 2
## 0 Kikuyu 255
## 0 Kipsigis 1
## 0 Kisii 38
## 0 Kuria 1
## 0 Luhya 125
## 0 Luhya And Kisii 1
## 0 Luhyakikuyu 2
## 0 Luo 184
## 0 Maasai 5
## 0 Masika 1
## 0 Mbeere 1
## 0 Meru 15
## 0 Mijikenda 3
## 0 Muslim 4
## 0 NA 37
## 0 Na 2
## 0 Nubi 1
## 0 Nuer 1
## 0 Oromo 1
## 0 Pokomo 1
## 0 Rabai 1
## 0 Rwandese 1
## 0 Somali 13
## 0 Suba 1
## 0 Sudanese 3
## 0 Swahili 1
## 0 Taiba Nigerian 1
## 0 Taita 3
## 0 Teso 4
## 0 Turkana 1
## 0 Tutsi 1
## 0 Ugandan 3
## 1 Abagusii 2
## 1 Arab 1
## 1 Bantu 1
## 1 Borana 10
## 1 Bukusu 1
## 1 Burji 5
## 1 Chonyi 1
## 1 Digo 1
## 1 Dinka 1
## 1 Duruma 1
## 1 Embu 4
## 1 Gabra 2
## 1 Giriama 3
## 1 Kalenjin 9
## 1 Kamba 84
## 1 Kikuyu 257
## 1 Kipsigis 1
## 1 Kisii 40
## 1 Kisii Kamba 1
## 1 Kuria 1
## 1 Lamu 1
## 1 Luhya 156
## 1 Luhya And Kisii 1
## 1 Luo 180
## 1 Luyha 1
## 1 Maasai 3
## 1 Mbeere 1
## 1 Meru 13
## 1 Mijikenda 1
## 1 Muslim 3
## 1 Mwilwana 1
## 1 NA 36
## 1 Na 4
## 1 Nandi 2
## 1 Nubi 1
## 1 Nuer 1
## 1 Oromo 2
## 1 Pokomo 1
## 1 Rabai 2
## 1 Rendille 1
## 1 Rwandese 1
## 1 Samburu 1
## 1 Somali 9
## 1 Suba 2
## 1 Sudanese 2
## 1 Swahili 2
## 1 Taita 4
## 1 Taveta 2
## 1 Teso 3
## 1 Ugandakenya 1
## 1 Ugandan 1
## 1 Yoruba 1
## -------------------------------------------stu_county <- rct %>%
group_by(trt,County) %>%
summarise(no_of_students = n_distinct(participant_id)) %>%
arrange(desc(no_of_students))## `summarise()` has grouped output by 'trt'. You can override using the `.groups`
## argument.# Print the summary table with stargazer
stargazer(stu_county, type = "text", summary = FALSE, rownames = FALSE)##
## ===================================
## trt County no_of_students
## -----------------------------------
## 1 Nairobi 173
## 0 Nairobi 163
## 1 Kiambu 145
## 0 Kiambu 122
## 0 Siaya 72
## 1 Siaya 64
## 1 Kakamega 45
## 0 Machakos 39
## 0 NA 36
## 1 Kisumu 35
## 1 Machakos 35
## 0 Kakamega 34
## 0 Kisumu 34
## 0 Muranga 32
## 1 Busia 32
## 1 NA 32
## 0 Vihiga 30
## 1 Muranga 29
## 0 Busia 26
## 1 Kisii 25
## 1 Vihiga 22
## 0 Kisii 21
## 1 Makueni 20
## 0 Kitui 19
## 0 Makueni 19
## 1 Homa Bay 19
## 1 Migori 18
## 1 Nakuru 18
## 0 Nyeri 17
## 1 Kajiado 15
## 1 Kitui 15
## 0 Homa Bay 14
## 1 Na 13
## 0 Nyamira 12
## 0 Bungoma 11
## 0 Meru 11
## 0 Migori 11
## 0 Nakuru 11
## 1 Nyeri 11
## 0 Kajiado 10
## 0 Nyandarua 10
## 0 Mombasa 8
## 0 Na 8
## 1 Meru 8
## 1 Trans Nzoia 8
## 0 Mandera 7
## 1 Bungoma 7
## 1 Marsabit 7
## 0 Marsabit 6
## 0 Narok 6
## 1 Kirinyaga 6
## 1 Nyamira 6
## 1 Nyandarua 6
## 1 Uasin Gishu 6
## 0 Embu 5
## 0 Kilifi 5
## 1 Embu 5
## 0 Laikipia 4
## 0 Moyale 4
## 0 Trans Nzoia 4
## 0 Wajir 4
## 1 Kilifi 4
## 1 Mandera 4
## 1 Taita Taveta 4
## 0 Kenya 3
## 0 Kericho 3
## 0 Kitale 3
## 0 Uasin Gishu 3
## 0 Uganda 3
## 1 Isiolo 3
## 1 Mombasa 3
## 1 Narok 3
## 0 Gatundu 2
## 0 Navaisha 2
## 0 Taita Taveta 2
## 0 Tana River 2
## 1 Garissa 2
## 1 Moyale 2
## 1 Uganda 2
## 0 Baringo 1
## 0 Congo 1
## 0 Eldoret 1
## 0 Kirinyaga 1
## 0 Lamu 1
## 0 Mbooni East 1
## 0 Mwingi 1
## 0 Nandi 1
## 0 Nyahururu 1
## 0 Rwanda 1
## 0 Tharaka Nithi 1
## 0 Turkana 1
## 1 Elgeyomarakwet 1
## 1 Kenya 1
## 1 Kericho 1
## 1 Kwale 1
## 1 Laikipia 1
## 1 Lamu 1
## 1 Nandi 1
## 1 Navaisha 1
## 1 Nueri 1
## 1 Nyahururumandera 1
## 1 Nyakach 1
## 1 Samburu 1
## 1 South Sudan 1
## 1 Tana River 1
## ------------------------------------ Home, Siblings, Religion, Parent’s home, Parent’s Dead, Father’s Education, Mother’s Education
stu_home <- rct %>%
group_by(trt,Home) %>%
summarise(no_of_students = n_distinct(participant_id)) %>%
mutate(Home = case_when(
Home == 1 ~ "Rural Area",
Home == 2 ~ "Small town",
Home == 3 ~ "Big town",
Home == 4 ~ "City",
))## `summarise()` has grouped output by 'trt'. You can override using the `.groups`
## argument.# Print the summary table with stargazer
stargazer(stu_home, type = "text", summary = FALSE, rownames = FALSE)##
## =============================
## trt Home no_of_students
## -----------------------------
## 0 Rural Area 211
## 0 Small town 273
## 0 Big town 193
## 0 City 128
## 0 45
## 1 Rural Area 208
## 1 Small town 304
## 1 Big town 198
## 1 City 115
## 1 41
## -----------------------------stu_sib <- rct %>%
group_by(trt,Siblings) %>%
summarise(no_of_students = n_distinct(participant_id))## `summarise()` has grouped output by 'trt'. You can override using the `.groups`
## argument.# Print the summary table with stargazer
stargazer(stu_sib, type = "text", summary = FALSE, rownames = FALSE)##
## ===========================
## trt Siblings no_of_students
## ---------------------------
## 0 0 26
## 0 1 124
## 0 2 211
## 0 3 177
## 0 4 117
## 0 5 81
## 0 6 41
## 0 7 41
## 0 NA 32
## 1 0 34
## 1 1 143
## 1 2 203
## 1 3 185
## 1 4 123
## 1 5 72
## 1 6 41
## 1 7 31
## 1 NA 34
## ---------------------------stu_rel<- rct %>%
group_by(trt,Religion) %>%
summarise(no_of_students = n_distinct(participant_id)) %>%
mutate(Home = case_when(
Religion == 1 ~ "Christian Protestants",
Religion == 2 ~ "Christian Catholics",
Religion == 3 ~ "Muslim",
Religion == 4 ~ "Hindu",
Religion == 5 ~ "Buddhist",
Religion == 6 ~ "Traditional Africans",
Religion == 7 ~ "No Religion",
Religion == 8 ~ "Other",
))## `summarise()` has grouped output by 'trt'. You can override using the `.groups`
## argument.# Print the summary table with stargazer
stargazer(stu_rel, type = "text", summary = FALSE, rownames = FALSE)##
## =================================================
## trt Religion no_of_students Home
## -------------------------------------------------
## 0 1 507 Christian Protestants
## 0 2 224 Christian Catholics
## 0 3 56 Muslim
## 0 6 6 Traditional Africans
## 0 7 6 No Religion
## 0 8 21 Other
## 0 NA 30
## 1 1 535 Christian Protestants
## 1 2 223 Christian Catholics
## 1 3 45 Muslim
## 1 6 6 Traditional Africans
## 1 7 7 No Religion
## 1 8 23 Other
## 1 NA 27
## -------------------------------------------------# Group by Parents_Home and time, and count the number of students
stu_parentshome <- rct %>%
group_by(trt,time,Parents_Home) %>%
summarise(no_of_students = n(), .groups = 'drop') %>%
mutate(Home = case_when(
Parents_Home == 0 ~ "No Parent",
Parents_Home == 1 ~ "Single Parent",
Parents_Home == 2 ~ "Both Parents"
))
# Pivot the data to a wider format
stu_parentshome_wide <- stu_parentshome %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_parentshome_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ==============================================
## trt Parents_Home Home 0 2 4 8
## ----------------------------------------------
## 0 0 No Parent 27 27 27 27
## 0 1 Single Parent 280 280 280 280
## 0 2 Both Parents 509 509 509 509
## 0 NA 34 34 34 34
## 1 0 No Parent 30 30 30 30
## 1 1 Single Parent 330 330 330 330
## 1 2 Both Parents 491 491 491 491
## 1 NA 15 15 15 15
## ----------------------------------------------# Group by Parents_Dead and time, and count the number of students
stu_parentdead <- rct %>%
group_by(trt, time,Parents_Dead) %>%
summarise(no_of_students = n(), .groups = 'drop') %>%
mutate(Dead = case_when(
Parents_Dead == 1 ~ "Father",
Parents_Dead == 2 ~ "Mother",
Parents_Dead == 3 ~ "Both",
Parents_Dead == 4 ~ "None"
))
# Pivot the data to a wider format
stu_parentdead_wide <- stu_parentdead %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_parentdead_wide, type = "text", summary = FALSE, rownames = FALSE)##
## =======================================
## trt Parents_Dead Dead 0 2 4 8
## ---------------------------------------
## 0 1 Father 84 84 84 84
## 0 2 Mother 27 27 27 27
## 0 3 Both 17 17 17 17
## 0 4 None 688 688 688 688
## 0 NA 34 34 34 34
## 1 1 Father 99 99 99 99
## 1 2 Mother 35 35 35 35
## 1 3 Both 15 15 15 15
## 1 4 None 694 694 694 694
## 1 NA 23 23 23 23
## ---------------------------------------# Group by Fathers_Education and time, and count the number of students
stu_fatheredu <- rct %>%
group_by(trt,time,Fathers_Education) %>%
summarise(no_of_students = n(), .groups = 'drop') %>%
mutate(FEDU = case_when(
Fathers_Education == 1 ~ "University",
Fathers_Education == 2 ~ "Secondary School",
Fathers_Education == 3 ~ "Primary School",
Fathers_Education == 4 ~ "Not Aware"
))
# Pivot the data to a wider format
stu_fatheredu_wide <- stu_fatheredu %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_fatheredu_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ======================================================
## trt Fathers_Education FEDU 0 2 4 8
## ------------------------------------------------------
## 0 1 University 175 175 175 175
## 0 2 Secondary School 226 226 226 226
## 0 3 Primary School 74 74 74 74
## 0 4 Not Aware 333 333 333 333
## 0 NA 42 42 42 42
## 1 1 University 172 172 172 172
## 1 2 Secondary School 240 240 240 240
## 1 3 Primary School 99 99 99 99
## 1 4 Not Aware 313 313 313 313
## 1 NA 42 42 42 42
## ------------------------------------------------------# Group by Mothers_Education and time, and count the number of students
stu_motheredu <- rct %>%
group_by(trt,time,Mothers_Education) %>%
summarise(no_of_students = n(), .groups = 'drop') %>%
mutate(MEDU = case_when(
Mothers_Education == 1 ~ "University",
Mothers_Education == 2 ~ "Secondary School",
Mothers_Education == 3 ~ "Primary School",
Mothers_Education == 4 ~ "Not Aware"
))
# Pivot the data to a wider format
stu_motheredu_wide <- stu_motheredu %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_motheredu_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ======================================================
## trt Mothers_Education MEDU 0 2 4 8
## ------------------------------------------------------
## 0 1 University 163 163 163 163
## 0 2 Secondary School 234 234 234 234
## 0 3 Primary School 171 171 171 171
## 0 4 Not Aware 251 251 251 251
## 0 NA 31 31 31 31
## 1 1 University 146 146 146 146
## 1 2 Secondary School 280 280 280 280
## 1 3 Primary School 184 184 184 184
## 1 4 Not Aware 221 221 221 221
## 1 NA 35 35 35 35
## ------------------------------------------------------- Co-curricular, Sports, Academics
# Group by student gender, Co_Curricular, and time, then count the number of students
stu_cocu_gen <- rct %>%
group_by(trt,time,Gender, Co_Curricular) %>%
summarise(no_of_students = n(), .groups = 'drop') %>%
mutate(Cocu = case_when(
Co_Curricular == 1 ~ "Not",
Co_Curricular == 2 ~ "Quite",
Co_Curricular == 3 ~ "Extremely"
))
# Pivot the data to a wider format
stu_cocu_gen_wide <- stu_cocu_gen %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_cocu_gen_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ==================================================
## trt Gender Co_Curricular Cocu 0 2 4 8
## --------------------------------------------------
## 0 Female 1 Not 118 118 118 118
## 0 Female 2 Quite 161 161 161 161
## 0 Female 3 Extremely 124 124 124 124
## 0 Female NA 14 14 14 14
## 0 Male 1 Not 107 107 107 107
## 0 Male 2 Quite 162 162 162 162
## 0 Male 3 Extremely 116 116 116 116
## 0 Male NA 20 20 20 20
## 0 NA 1 Not 4 4 4 4
## 0 NA 2 Quite 8 8 8 8
## 0 NA 3 Extremely 9 9 9 9
## 0 NA NA 7 7 7 7
## 1 Female 1 Not 128 128 128 128
## 1 Female 2 Quite 164 164 164 164
## 1 Female 3 Extremely 112 112 112 112
## 1 Female NA 15 15 15 15
## 1 Male 1 Not 117 117 117 117
## 1 Male 2 Quite 175 175 175 175
## 1 Male 3 Extremely 113 113 113 113
## 1 Male NA 12 12 12 12
## 1 NA 1 Not 11 11 11 11
## 1 NA 2 Quite 9 9 9 9
## 1 NA 3 Extremely 7 7 7 7
## 1 NA NA 3 3 3 3
## --------------------------------------------------# Group by student age, Co_Curricular, and time, then count the number of students
stu_cocu_age <- rct %>%
group_by(trt,Age, Co_Curricular, time) %>%
summarise(no_of_students = n(), .groups = 'drop') %>%
mutate(Cocu = case_when(
Co_Curricular == 1 ~ "Not",
Co_Curricular == 2 ~ "Quite",
Co_Curricular == 3 ~ "Extremely"
))
# Pivot the data to a wider format
stu_cocu_age_wide <- stu_cocu_age %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_cocu_age_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ===============================================
## trt Age Co_Curricular Cocu 0 2 4 8
## -----------------------------------------------
## 0 12 2 Quite 1 1 1 1
## 0 13 1 Not 5 5 5 5
## 0 13 2 Quite 2 2 2 2
## 0 13 3 Extremely 6 6 6 6
## 0 14 1 Not 28 28 28 28
## 0 14 2 Quite 43 43 43 43
## 0 14 3 Extremely 16 16 16 16
## 0 14 NA 4 4 4 4
## 0 15 1 Not 53 53 53 53
## 0 15 2 Quite 80 80 80 80
## 0 15 3 Extremely 54 54 54 54
## 0 15 NA 13 13 13 13
## 0 16 1 Not 56 56 56 56
## 0 16 2 Quite 101 101 101 101
## 0 16 3 Extremely 58 58 58 58
## 0 16 NA 7 7 7 7
## 0 17 1 Not 39 39 39 39
## 0 17 2 Quite 49 49 49 49
## 0 17 3 Extremely 57 57 57 57
## 0 17 NA 8 8 8 8
## 0 18 1 Not 24 24 24 24
## 0 18 2 Quite 30 30 30 30
## 0 18 3 Extremely 30 30 30 30
## 0 18 NA 3 3 3 3
## 0 19 1 Not 7 7 7 7
## 0 19 2 Quite 11 11 11 11
## 0 19 3 Extremely 9 9 9 9
## 0 20 1 Not 3 3 3 3
## 0 20 2 Quite 1 1 1 1
## 0 20 3 Extremely 2 2 2 2
## 0 22 1 Not 2 2 2 2
## 0 22 2 Quite 1 1 1 1
## 0 22 NA 1 1 1 1
## 0 NA 1 Not 12 12 12 12
## 0 NA 2 Quite 12 12 12 12
## 0 NA 3 Extremely 17 17 17 17
## 0 NA NA 5 5 5 5
## 1 13 1 Not 8 8 8 8
## 1 13 2 Quite 5 5 5 5
## 1 13 3 Extremely 4 4 4 4
## 1 14 1 Not 28 28 28 28
## 1 14 2 Quite 56 56 56 56
## 1 14 3 Extremely 24 24 24 24
## 1 14 NA 4 4 4 4
## 1 15 1 Not 56 56 56 56
## 1 15 2 Quite 86 86 86 86
## 1 15 3 Extremely 66 66 66 66
## 1 15 NA 1 1 1 1
## 1 16 1 Not 65 65 65 65
## 1 16 2 Quite 69 69 69 69
## 1 16 3 Extremely 62 62 62 62
## 1 16 NA 8 8 8 8
## 1 17 1 Not 52 52 52 52
## 1 17 2 Quite 73 73 73 73
## 1 17 3 Extremely 51 51 51 51
## 1 17 NA 8 8 8 8
## 1 18 1 Not 21 21 21 21
## 1 18 2 Quite 31 31 31 31
## 1 18 3 Extremely 13 13 13 13
## 1 18 NA 2 2 2 2
## 1 19 1 Not 7 7 7 7
## 1 19 2 Quite 7 7 7 7
## 1 19 3 Extremely 4 4 4 4
## 1 19 NA 1 1 1 1
## 1 20 1 Not 1 1 1 1
## 1 20 2 Quite 1 1 1 1
## 1 20 3 Extremely 1 1 1 1
## 1 20 NA 1 1 1 1
## 1 NA 1 Not 18 18 18 18
## 1 NA 2 Quite 20 20 20 20
## 1 NA 3 Extremely 7 7 7 7
## 1 NA NA 5 5 5 5
## -----------------------------------------------# Group by student gender, Sports, and time, then count the number of students
stu_sports_gen <- rct %>%
group_by(trt,Gender, Sports, time) %>%
summarise(no_of_students = n(), .groups = 'drop') %>%
mutate(spo = case_when(
Sports == 1 ~ "Yes",
Sports == 2 ~ "No"
))
# Pivot the data to a wider format
stu_sports_gen_wide <- stu_sports_gen %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_sports_gen_wide, type = "text", summary = FALSE, rownames = FALSE)##
## =====================================
## trt Gender Sports spo 0 2 4 8
## -------------------------------------
## 0 Female 1 Yes 126 126 126 126
## 0 Female 2 No 280 280 280 280
## 0 Female NA 11 11 11 11
## 0 Male 1 Yes 131 131 131 131
## 0 Male 2 No 259 259 259 259
## 0 Male NA 15 15 15 15
## 0 NA 1 Yes 10 10 10 10
## 0 NA 2 No 14 14 14 14
## 0 NA NA 4 4 4 4
## 1 Female 1 Yes 113 113 113 113
## 1 Female 2 No 297 297 297 297
## 1 Female NA 9 9 9 9
## 1 Male 1 Yes 123 123 123 123
## 1 Male 2 No 281 281 281 281
## 1 Male NA 13 13 13 13
## 1 NA 1 Yes 6 6 6 6
## 1 NA 2 No 20 20 20 20
## 1 NA NA 4 4 4 4
## -------------------------------------# Group by student age, Sports, and time, then count the number of students
stu_sports_age <- rct %>%
group_by(trt,Age, Sports, time) %>%
summarise(no_of_students = n(), .groups = 'drop') %>%
mutate(spo = case_when(
Sports == 1 ~ "Yes",
Sports == 2 ~ "No"
))
# Pivot the data to a wider format
stu_sports_age_wide <- stu_sports_age %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_sports_age_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ==================================
## trt Age Sports spo 0 2 4 8
## ----------------------------------
## 0 12 1 Yes 1 1 1 1
## 0 13 1 Yes 3 3 3 3
## 0 13 2 No 10 10 10 10
## 0 14 1 Yes 16 16 16 16
## 0 14 2 No 71 71 71 71
## 0 14 NA 4 4 4 4
## 0 15 1 Yes 57 57 57 57
## 0 15 2 No 138 138 138 138
## 0 15 NA 5 5 5 5
## 0 16 1 Yes 64 64 64 64
## 0 16 2 No 154 154 154 154
## 0 16 NA 4 4 4 4
## 0 17 1 Yes 49 49 49 49
## 0 17 2 No 95 95 95 95
## 0 17 NA 9 9 9 9
## 0 18 1 Yes 48 48 48 48
## 0 18 2 No 35 35 35 35
## 0 18 NA 4 4 4 4
## 0 19 1 Yes 9 9 9 9
## 0 19 2 No 17 17 17 17
## 0 19 NA 1 1 1 1
## 0 20 1 Yes 1 1 1 1
## 0 20 2 No 4 4 4 4
## 0 20 NA 1 1 1 1
## 0 22 1 Yes 2 2 2 2
## 0 22 2 No 2 2 2 2
## 0 NA 1 Yes 17 17 17 17
## 0 NA 2 No 27 27 27 27
## 0 NA NA 2 2 2 2
## 1 13 1 Yes 3 3 3 3
## 1 13 2 No 13 13 13 13
## 1 13 NA 1 1 1 1
## 1 14 1 Yes 25 25 25 25
## 1 14 2 No 85 85 85 85
## 1 14 NA 2 2 2 2
## 1 15 1 Yes 51 51 51 51
## 1 15 2 No 156 156 156 156
## 1 15 NA 2 2 2 2
## 1 16 1 Yes 60 60 60 60
## 1 16 2 No 138 138 138 138
## 1 16 NA 6 6 6 6
## 1 17 1 Yes 54 54 54 54
## 1 17 2 No 122 122 122 122
## 1 17 NA 8 8 8 8
## 1 18 1 Yes 21 21 21 21
## 1 18 2 No 43 43 43 43
## 1 18 NA 3 3 3 3
## 1 19 1 Yes 9 9 9 9
## 1 19 2 No 10 10 10 10
## 1 20 1 Yes 3 3 3 3
## 1 20 2 No 1 1 1 1
## 1 NA 1 Yes 16 16 16 16
## 1 NA 2 No 30 30 30 30
## 1 NA NA 4 4 4 4
## ----------------------------------# Group by student gender, Percieved_Academic_Abilities, and time, then count the number of students
stu_acad_gen <- rct %>%
group_by(trt,Gender, Percieved_Academic_Abilities, time) %>%
summarise(no_of_students = n(), .groups = 'drop') %>%
mutate(acad = case_when(
Percieved_Academic_Abilities == 1 ~ "Not Satisfactory",
Percieved_Academic_Abilities == 2 ~ "Satisfactory",
Percieved_Academic_Abilities == 3 ~ "Good",
Percieved_Academic_Abilities == 4 ~ "Very Good",
Percieved_Academic_Abilities == 5 ~ "Excellent"
))
# Pivot the data to a wider format
stu_acad_gen_wide <- stu_acad_gen %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_acad_gen_wide, type = "text", summary = FALSE, rownames = FALSE)##
## ========================================================================
## trt Gender Percieved_Academic_Abilities acad 0 2 4 8
## ------------------------------------------------------------------------
## 0 Female 1 Not Satisfactory 92 92 92 92
## 0 Female 2 Satisfactory 57 57 57 57
## 0 Female 3 Good 130 130 130 130
## 0 Female 4 Very Good 59 59 59 59
## 0 Female 5 Excellent 65 65 65 65
## 0 Female NA 14 14 14 14
## 0 Male 1 Not Satisfactory 53 53 53 53
## 0 Male 2 Satisfactory 68 68 68 68
## 0 Male 3 Good 140 140 140 140
## 0 Male 4 Very Good 64 64 64 64
## 0 Male 5 Excellent 63 63 63 63
## 0 Male NA 17 17 17 17
## 0 NA 1 Not Satisfactory 1 1 1 1
## 0 NA 2 Satisfactory 4 4 4 4
## 0 NA 3 Good 5 5 5 5
## 0 NA 4 Very Good 6 6 6 6
## 0 NA 5 Excellent 6 6 6 6
## 0 NA NA 6 6 6 6
## 1 Female 1 Not Satisfactory 99 99 99 99
## 1 Female 2 Satisfactory 66 66 66 66
## 1 Female 3 Good 121 121 121 121
## 1 Female 4 Very Good 67 67 67 67
## 1 Female 5 Excellent 59 59 59 59
## 1 Female NA 7 7 7 7
## 1 Male 1 Not Satisfactory 50 50 50 50
## 1 Male 2 Satisfactory 70 70 70 70
## 1 Male 3 Good 158 158 158 158
## 1 Male 4 Very Good 81 81 81 81
## 1 Male 5 Excellent 47 47 47 47
## 1 Male NA 11 11 11 11
## 1 NA 1 Not Satisfactory 4 4 4 4
## 1 NA 2 Satisfactory 4 4 4 4
## 1 NA 3 Good 12 12 12 12
## 1 NA 4 Very Good 1 1 1 1
## 1 NA 5 Excellent 6 6 6 6
## 1 NA NA 3 3 3 3
## ------------------------------------------------------------------------# Group by student age, Percieved_Academic_Abilities, and time, then count the number of students
stu_acad_age <- rct %>%
group_by(trt, Age,Percieved_Academic_Abilities, time) %>%
summarise(no_of_students = n(), .groups = 'drop') %>%
mutate(acad = case_when(
Percieved_Academic_Abilities == 1 ~ "Not Satisfactory",
Percieved_Academic_Abilities == 2 ~ "Satisfactory",
Percieved_Academic_Abilities == 3 ~ "Good",
Percieved_Academic_Abilities == 4 ~ "Very Good",
Percieved_Academic_Abilities == 5 ~ "Excellent"
))
# Pivot the data to a wider format
stu_acad_age_wide <- stu_acad_age %>%
pivot_wider(names_from = time, values_from = no_of_students, values_fill = list(no_of_students = 0))
# Display the table using stargazer
stargazer(stu_acad_age_wide, type = "text", summary = FALSE, rownames = FALSE)##
## =================================================================
## trt Age Percieved_Academic_Abilities acad 0 2 4 8
## -----------------------------------------------------------------
## 0 12 1 Not Satisfactory 1 1 1 1
## 0 13 1 Not Satisfactory 1 1 1 1
## 0 13 2 Satisfactory 1 1 1 1
## 0 13 3 Good 5 5 5 5
## 0 13 4 Very Good 1 1 1 1
## 0 13 5 Excellent 4 4 4 4
## 0 13 NA 1 1 1 1
## 0 14 1 Not Satisfactory 11 11 11 11
## 0 14 2 Satisfactory 10 10 10 10
## 0 14 3 Good 28 28 28 28
## 0 14 4 Very Good 22 22 22 22
## 0 14 5 Excellent 16 16 16 16
## 0 14 NA 4 4 4 4
## 0 15 1 Not Satisfactory 39 39 39 39
## 0 15 2 Satisfactory 30 30 30 30
## 0 15 3 Good 57 57 57 57
## 0 15 4 Very Good 32 32 32 32
## 0 15 5 Excellent 35 35 35 35
## 0 15 NA 7 7 7 7
## 0 16 1 Not Satisfactory 38 38 38 38
## 0 16 2 Satisfactory 38 38 38 38
## 0 16 3 Good 70 70 70 70
## 0 16 4 Very Good 33 33 33 33
## 0 16 5 Excellent 32 32 32 32
## 0 16 NA 11 11 11 11
## 0 17 1 Not Satisfactory 30 30 30 30
## 0 17 2 Satisfactory 24 24 24 24
## 0 17 3 Good 50 50 50 50
## 0 17 4 Very Good 21 21 21 21
## 0 17 5 Excellent 20 20 20 20
## 0 17 NA 8 8 8 8
## 0 18 1 Not Satisfactory 17 17 17 17
## 0 18 2 Satisfactory 16 16 16 16
## 0 18 3 Good 31 31 31 31
## 0 18 4 Very Good 8 8 8 8
## 0 18 5 Excellent 13 13 13 13
## 0 18 NA 2 2 2 2
## 0 19 1 Not Satisfactory 3 3 3 3
## 0 19 2 Satisfactory 6 6 6 6
## 0 19 3 Good 11 11 11 11
## 0 19 4 Very Good 2 2 2 2
## 0 19 5 Excellent 5 5 5 5
## 0 20 1 Not Satisfactory 1 1 1 1
## 0 20 3 Good 4 4 4 4
## 0 20 NA 1 1 1 1
## 0 22 1 Not Satisfactory 2 2 2 2
## 0 22 3 Good 1 1 1 1
## 0 22 5 Excellent 1 1 1 1
## 0 NA 1 Not Satisfactory 3 3 3 3
## 0 NA 2 Satisfactory 4 4 4 4
## 0 NA 3 Good 18 18 18 18
## 0 NA 4 Very Good 10 10 10 10
## 0 NA 5 Excellent 8 8 8 8
## 0 NA NA 3 3 3 3
## 1 13 2 Satisfactory 2 2 2 2
## 1 13 3 Good 8 8 8 8
## 1 13 4 Very Good 4 4 4 4
## 1 13 5 Excellent 3 3 3 3
## 1 14 1 Not Satisfactory 14 14 14 14
## 1 14 2 Satisfactory 15 15 15 15
## 1 14 3 Good 38 38 38 38
## 1 14 4 Very Good 27 27 27 27
## 1 14 5 Excellent 16 16 16 16
## 1 14 NA 2 2 2 2
## 1 15 1 Not Satisfactory 37 37 37 37
## 1 15 2 Satisfactory 24 24 24 24
## 1 15 3 Good 75 75 75 75
## 1 15 4 Very Good 43 43 43 43
## 1 15 5 Excellent 29 29 29 29
## 1 15 NA 1 1 1 1
## 1 16 1 Not Satisfactory 31 31 31 31
## 1 16 2 Satisfactory 45 45 45 45
## 1 16 3 Good 63 63 63 63
## 1 16 4 Very Good 31 31 31 31
## 1 16 5 Excellent 26 26 26 26
## 1 16 NA 8 8 8 8
## 1 17 1 Not Satisfactory 46 46 46 46
## 1 17 2 Satisfactory 31 31 31 31
## 1 17 3 Good 55 55 55 55
## 1 17 4 Very Good 24 24 24 24
## 1 17 5 Excellent 24 24 24 24
## 1 17 NA 4 4 4 4
## 1 18 1 Not Satisfactory 15 15 15 15
## 1 18 2 Satisfactory 11 11 11 11
## 1 18 3 Good 28 28 28 28
## 1 18 4 Very Good 10 10 10 10
## 1 18 5 Excellent 1 1 1 1
## 1 18 NA 2 2 2 2
## 1 19 1 Not Satisfactory 5 5 5 5
## 1 19 2 Satisfactory 4 4 4 4
## 1 19 3 Good 7 7 7 7
## 1 19 4 Very Good 1 1 1 1
## 1 19 5 Excellent 2 2 2 2
## 1 20 1 Not Satisfactory 1 1 1 1
## 1 20 3 Good 1 1 1 1
## 1 20 4 Very Good 2 2 2 2
## 1 NA 1 Not Satisfactory 4 4 4 4
## 1 NA 2 Satisfactory 8 8 8 8
## 1 NA 3 Good 16 16 16 16
## 1 NA 4 Very Good 7 7 7 7
## 1 NA 5 Excellent 11 11 11 11
## 1 NA NA 4 4 4 4
## -----------------------------------------------------------------Part 3: Frequency of responses to Mental health questions in each scale
- PCS, PHQ, GAD scores
# Function to get the question text based on PCS column name
get_question_text <- function(pcs_col) {
question_texts <- list(
PCS_1 = "If I try to behave, I can keep myself out of trouble.",
PCS_2 = "I can not succeed at school no matter how hard I try.",
PCS_3 = "I can stay out of trouble if I really try.",
PCS_4 = "I can not get good grades no matter how hard I try.",
PCS_5 = "I can not do well on tests at school even if I study hard.",
PCS_6 = "I can not get good marks for my homework, even if I work hard at it."
)
return(question_texts[[pcs_col]])
}
# Function to extract the legend from a ggplot
get_legend <- function(myggplot) {
tmp <- ggplotGrob(myggplot)
legend <- tmp$grobs[which(sapply(tmp$grobs, function(x) x$name) == "guide-box")]
return(legend[[1]])
}
# Function to summarize a PCS column by time and create a frequency plot
create_summary_and_plot <- function(data, pcs_col) {
# Summarize the PCS column by time and treatment group
freq_data <- data %>%
select(time, !!sym(pcs_col), trt) %>%
mutate(score = ifelse(is.na(!!sym(pcs_col)), "NA", as.character(!!sym(pcs_col)))) %>%
group_by(score, time, trt) %>%
summarise(Frequency = n(), .groups = 'drop')
# Add score labels
freq_data <- freq_data %>%
mutate(score_label = case_when(
score == "0" ~ "Very False",
score == "1" ~ "Somewhat False",
score == "2" ~ "Somewhat True",
score == "3" ~ "Very True",
score == "NA" ~ "NA",
TRUE ~ as.character(score) # For any other values
))
# Pivot the summarized data to wide format
freq_wide <- freq_data %>%
pivot_wider(names_from = time, values_from = Frequency, values_fill = list(Frequency = 0))
# Get the question text for the title
question_text <- get_question_text(pcs_col)
# Create plots for each treatment group
plots <- freq_data %>%
split(.$trt) %>%
map(~ ggplot(.x, aes(x = time, y = Frequency, color = score_label, group = score_label)) +
geom_line() +
geom_point() +
labs(title = paste("Treatment Group:", unique(.x$trt)),
x = "Time",
y = "Frequency") +
theme_minimal() +
theme(
plot.title = element_text(hjust = 0.5, size = 12),
axis.title = element_text(size = 10),
axis.text = element_text(size = 8),
legend.position = "none"
) +
scale_x_continuous(breaks = c(0, 2, 4, 8))) # Specify the breaks on x-axis
# Extract the legend from one plot
legend <- get_legend(plots[[1]] + labs(color = "Score") + theme(legend.position = "bottom"))
# Combine the plots and add the common title
combined_plot <- arrangeGrob(grobs = lapply(plots, ggplotGrob), ncol = 2, top = textGrob(
paste("Frequency of Scores for", pcs_col, ":", question_text),
gp = gpar(fontsize = 14), hjust = 0.5
))
# Print stargazer table
cat("\n", pcs_col, " summary by time and treatment group\n")
stargazer(freq_wide, type = "text", summary = FALSE, rownames = FALSE)
# Combine the plots and legend
final_plot <- arrangeGrob(combined_plot, legend, ncol = 1, heights = c(10, 1))
# Print the final plot
grid.newpage()
grid.draw(final_plot)
}
# List of PCS columns
pcs_columns <- paste0("PCS_", 1:6)
# Apply the function to each PCS column
for (pcs_col in pcs_columns) {
create_summary_and_plot(rct, pcs_col)
}##
## PCS_1 summary by time and treatment group
##
## ========================================
## score trt score_label 0 2 4 8
## ----------------------------------------
## 0 0 Very False 31 27 48 56
## 0 1 Very False 42 29 49 55
## 1 0 Somewhat False 20 7 13 21
## 1 1 Somewhat False 28 21 13 23
## 2 0 Somewhat True 71 87 82 86
## 2 1 Somewhat True 116 88 53 88
## 3 0 Very True 701 572 584 552
## 3 1 Very True 654 501 468 529
## NA 0 NA 27 157 123 135
## NA 1 NA 26 227 283 171
## ----------------------------------------
##
## PCS_2 summary by time and treatment group
##
## ========================================
## score trt score_label 0 2 4 8
## ----------------------------------------
## 0 0 Very False 44 33 32 39
## 0 1 Very False 55 18 29 25
## 1 0 Somewhat False 62 52 53 53
## 1 1 Somewhat False 78 49 37 53
## 2 0 Somewhat True 71 97 90 101
## 2 1 Somewhat True 99 94 68 94
## 3 0 Very True 635 509 545 520
## 3 1 Very True 605 471 440 518
## NA 0 NA 38 159 130 137
## NA 1 NA 29 234 292 176
## ----------------------------------------
##
## PCS_3 summary by time and treatment group
##
## ========================================
## score trt score_label 0 2 4 8
## ----------------------------------------
## 0 0 Very False 54 32 49 46
## 0 1 Very False 58 38 39 57
## 1 0 Somewhat False 21 15 23 24
## 1 1 Somewhat False 46 25 24 37
## 2 0 Somewhat True 143 125 107 129
## 2 1 Somewhat True 145 124 103 117
## 3 0 Very True 599 521 530 506
## 3 1 Very True 572 443 405 477
## NA 0 NA 33 157 141 145
## NA 1 NA 45 236 295 178
## ----------------------------------------
##
## PCS_4 summary by time and treatment group
##
## ========================================
## score trt score_label 0 2 4 8
## ----------------------------------------
## 0 0 Very False 51 43 40 50
## 0 1 Very False 61 41 28 30
## 1 0 Somewhat False 72 63 51 50
## 1 1 Somewhat False 82 53 32 49
## 2 0 Somewhat True 92 81 89 88
## 2 1 Somewhat True 101 75 75 91
## 3 0 Very True 603 505 538 519
## 3 1 Very True 581 458 437 514
## NA 0 NA 32 158 132 143
## NA 1 NA 41 239 294 182
## ----------------------------------------
##
## PCS_5 summary by time and treatment group
##
## ========================================
## score trt score_label 0 2 4 8
## ----------------------------------------
## 0 0 Very False 53 42 47 53
## 0 1 Very False 57 37 36 37
## 1 0 Somewhat False 71 62 58 61
## 1 1 Somewhat False 104 69 46 71
## 2 0 Somewhat True 116 84 89 94
## 2 1 Somewhat True 129 104 66 86
## 3 0 Very True 575 506 520 500
## 3 1 Very True 546 418 419 501
## NA 0 NA 35 156 136 142
## NA 1 NA 30 238 299 171
## ----------------------------------------
##
## PCS_6 summary by time and treatment group
##
## ========================================
## score trt score_label 0 2 4 8
## ----------------------------------------
## 0 0 Very False 58 36 52 58
## 0 1 Very False 49 41 27 41
## 1 0 Somewhat False 59 43 52 53
## 1 1 Somewhat False 81 46 37 53
## 2 0 Somewhat True 116 115 81 91
## 2 1 Somewhat True 151 88 77 93
## 3 0 Very True 586 499 533 508
## 3 1 Very True 561 459 431 512
## NA 0 NA 31 157 132 140
## NA 1 NA 24 232 294 167
## ----------------------------------------
We can note the following:
- The trends for each question remain roughly the same. Including how students have responded to NA values in each of the time periods. The exact frequency of responses might differ but the trend is consistent.
- There are two types of questions about perceived control here - staying out of trouble and academic performances - both based on efforts from the students.
- For both normally coded and reverse coded questions, we see that the response frequencies are similar. This means that the positives and the negatives have the same trends - there may be a measurement error (most probably by the respondents).
- The issue here is - both the questions on staying out of trouble are normally coded and all the four questions on academic performance are negatively coded and are framed in the similar manner which is focused on test scores/grades/marks - which makes it difficult to isolate if the measurement error is actually present.
- Number of NA responses increase every two weeks - except for the last week: possible explanation is that there was response fatigue or the NAs reduced for the endline because it was the last survey and there may be a push from the organisation to encourage students to engage with the survey.
# Function to get the question text based on PHQ column name
get_phq_question_text <- function(phq_col) {
question_texts <- list(
PHQ_1 = "Little interest or pleasure in doing things?",
PHQ_2 = "Feeling sad, depressed or hopeless?",
PHQ_3 = "Trouble falling or staying asleep, or sleeping too much?",
PHQ_4 = "Feeling tired or having little energy?",
PHQ_5 = "Poor appetite or eating too much (overeating)?",
PHQ_6 = "Feeling bad about yourself– or that you are a failure or have let yourself or your family down?",
PHQ_7 = "Trouble concentrating on things, such as reading a book?",
PHQ_8 = "Moving or speaking so slowly that other people could have noticed. Or the opposite — being so energetic or nervous that you have been moving around a lot more than usual?",
PHQ_Functioning = "If you experienced any of the problems listed above, how difficult have they made it for you to do your work or get along with other people?"
)
return(question_texts[[phq_col]])
}
# Function to extract the legend from a ggplot
get_legend <- function(myggplot) {
tmp <- ggplotGrob(myggplot)
legend <- tmp$grobs[which(sapply(tmp$grobs, function(x) x$name) == "guide-box")]
return(legend[[1]])
}
# Function to summarize a PHQ column by time and create a frequency plot
create_summary_and_plot_phq <- function(data, phq_col) {
# Summarize the PHQ column by time and treatment group
freq_data <- data %>%
select(time, !!sym(phq_col), trt) %>%
mutate(score = ifelse(is.na(!!sym(phq_col)), "NA", as.character(!!sym(phq_col)))) %>%
group_by(score, time, trt) %>%
summarise(Frequency = n(), .groups = 'drop')
# Add score labels
freq_data <- freq_data %>%
mutate(score_label = case_when(
score == "0" ~ ifelse(phq_col == "PHQ_Functioning", "Not difficult at all", "Not at all"),
score == "1" ~ ifelse(phq_col == "PHQ_Functioning", "Somewhat difficult", "Several days (between 2 and 7 days)"),
score == "2" ~ ifelse(phq_col == "PHQ_Functioning", "Very difficult", "Over half the days (more than 7 days)"),
score == "3" ~ ifelse(phq_col == "PHQ_Functioning", "Extremely difficult", "Nearly/almost every day"),
score == "NA" ~ "NA",
TRUE ~ as.character(score) # For any other values
))
# Pivot the summarized data to wide format
freq_wide <- freq_data %>%
pivot_wider(names_from = time, values_from = Frequency, values_fill = list(Frequency = 0))
# Get the question text for the title
question_text <- get_phq_question_text(phq_col)
# Create plots for each treatment group
plots <- freq_data %>%
split(.$trt) %>%
map(~ ggplot(.x, aes(x = time, y = Frequency, color = score_label, group = score_label)) +
geom_line() +
geom_point() +
labs(title = paste("Treatment Group:", unique(.x$trt)),
x = "Time",
y = "Frequency") +
theme_minimal() +
theme(
plot.title = element_text(hjust = 0.5, size = 12),
axis.title = element_text(size = 10),
axis.text = element_text(size = 8),
legend.position = "none"
) +
scale_x_continuous(breaks = c(0, 2, 4, 8))) # Specify the breaks on x-axis
# Extract the legend from one plot
legend <- get_legend(plots[[1]] + labs(color = "Score") + theme(legend.position = "bottom"))
# Combine the plots and add the common title
combined_plot <- arrangeGrob(grobs = lapply(plots, ggplotGrob), ncol = 2, top = textGrob(
paste("Frequency of Scores for", phq_col, ":", question_text),
gp = gpar(fontsize = 14), hjust = 0.5
))
# Print stargazer table
cat("\n", phq_col, " summary by time and treatment group\n")
stargazer(freq_wide, type = "text", summary = FALSE, rownames = FALSE)
# Combine the plots and legend
final_plot <- arrangeGrob(combined_plot, legend, ncol = 1, heights = c(10, 1))
# Print the final plot
grid.newpage()
grid.draw(final_plot)
}
# List of PHQ columns (assuming PHQ_1 to PHQ_8 and PHQ_Functioning)
phq_columns <- c(paste0("PHQ_", 1:8), "PHQ_Functioning")
# Apply the function to each PHQ column
for (phq_col in phq_columns) {
create_summary_and_plot_phq(rct, phq_col)
}##
## PHQ_1 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 258 205 254 252
## 0 1 Not at all 251 168 202 253
## 1 0 Several days (between 2 and 7 days) 249 272 265 242
## 1 1 Several days (between 2 and 7 days) 286 245 211 260
## 2 0 Over half the days (more than 7 days) 77 89 100 106
## 2 1 Over half the days (more than 7 days) 107 108 97 86
## 3 0 Nearly/almost every day 201 101 109 101
## 3 1 Nearly/almost every day 150 92 57 91
## NA 0 NA 65 183 122 149
## NA 1 NA 72 253 299 176
## ---------------------------------------------------------------
##
## PHQ_2 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 371 316 368 345
## 0 1 Not at all 356 266 285 342
## 1 0 Several days (between 2 and 7 days) 221 202 205 202
## 1 1 Several days (between 2 and 7 days) 286 232 190 222
## 2 0 Over half the days (more than 7 days) 107 92 97 101
## 2 1 Over half the days (more than 7 days) 96 75 66 78
## 3 0 Nearly/almost every day 101 74 58 59
## 3 1 Nearly/almost every day 105 51 23 50
## NA 0 NA 50 166 122 143
## NA 1 NA 23 242 302 174
## ---------------------------------------------------------------
##
## PHQ_3 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 405 345 401 382
## 0 1 Not at all 428 286 311 366
## 1 0 Several days (between 2 and 7 days) 173 171 176 182
## 1 1 Several days (between 2 and 7 days) 189 169 147 186
## 2 0 Over half the days (more than 7 days) 107 87 74 71
## 2 1 Over half the days (more than 7 days) 94 99 60 73
## 3 0 Nearly/almost every day 106 78 69 67
## 3 1 Nearly/almost every day 121 68 36 63
## NA 0 NA 59 169 130 148
## NA 1 NA 34 244 312 178
## ---------------------------------------------------------------
##
## PHQ_4 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 354 280 303 324
## 0 1 Not at all 344 232 228 308
## 1 0 Several days (between 2 and 7 days) 247 240 240 217
## 1 1 Several days (between 2 and 7 days) 250 262 202 229
## 2 0 Over half the days (more than 7 days) 118 103 121 91
## 2 1 Over half the days (more than 7 days) 149 90 100 84
## 3 0 Nearly/almost every day 77 54 56 67
## 3 1 Nearly/almost every day 84 38 32 62
## NA 0 NA 54 173 130 151
## NA 1 NA 39 244 304 183
## ---------------------------------------------------------------
##
## PHQ_5 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 430 368 416 380
## 0 1 Not at all 460 341 319 390
## 1 0 Several days (between 2 and 7 days) 155 153 157 161
## 1 1 Several days (between 2 and 7 days) 182 143 132 144
## 2 0 Over half the days (more than 7 days) 94 68 78 68
## 2 1 Over half the days (more than 7 days) 81 71 57 78
## 3 0 Nearly/almost every day 120 96 69 93
## 3 1 Nearly/almost every day 98 69 47 75
## NA 0 NA 51 165 130 148
## NA 1 NA 45 242 311 179
## ---------------------------------------------------------------
##
## PHQ_6 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 380 343 380 359
## 0 1 Not at all 360 283 300 362
## 1 0 Several days (between 2 and 7 days) 194 168 197 169
## 1 1 Several days (between 2 and 7 days) 228 190 149 173
## 2 0 Over half the days (more than 7 days) 78 77 61 85
## 2 1 Over half the days (more than 7 days) 87 69 73 81
## 3 0 Nearly/almost every day 150 92 84 87
## 3 1 Nearly/almost every day 158 84 40 73
## NA 0 NA 48 170 128 150
## NA 1 NA 33 240 304 177
## ---------------------------------------------------------------
##
## PHQ_7 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 296 265 316 305
## 0 1 Not at all 309 227 255 310
## 1 0 Several days (between 2 and 7 days) 224 229 240 207
## 1 1 Several days (between 2 and 7 days) 228 212 173 207
## 2 0 Over half the days (more than 7 days) 112 105 108 106
## 2 1 Over half the days (more than 7 days) 113 91 83 90
## 3 0 Nearly/almost every day 170 86 58 87
## 3 1 Nearly/almost every day 184 93 55 84
## NA 0 NA 48 165 128 145
## NA 1 NA 32 243 300 175
## ---------------------------------------------------------------
##
## PHQ_8 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 449 382 405 406
## 0 1 Not at all 486 361 346 414
## 1 0 Several days (between 2 and 7 days) 163 160 169 167
## 1 1 Several days (between 2 and 7 days) 158 126 117 152
## 2 0 Over half the days (more than 7 days) 85 70 84 63
## 2 1 Over half the days (more than 7 days) 86 62 55 63
## 3 0 Nearly/almost every day 95 70 65 65
## 3 1 Nearly/almost every day 89 71 42 63
## NA 0 NA 58 168 127 149
## NA 1 NA 47 246 306 174
## ---------------------------------------------------------------
##
## PHQ_Functioning summary by time and treatment group
##
## ==============================================
## score trt score_label 0 2 4 8
## ----------------------------------------------
## 0 0 Not difficult at all 226 155 175 202
## 0 1 Not difficult at all 229 145 168 226
## 1 0 Somewhat difficult 339 305 302 317
## 1 1 Somewhat difficult 378 305 203 299
## 2 0 Very difficult 104 81 69 73
## 2 1 Very difficult 113 54 55 72
## 3 0 Extremely difficult 59 45 26 37
## 3 1 Extremely difficult 55 24 14 28
## NA 0 NA 122 264 278 221
## NA 1 NA 91 338 426 241
## ----------------------------------------------
We can note the following:
- NA values follow exactly the same trend pattern. Also, the number of people responding NA to the questions in various time periods are also quite similar.
- For every question, the frequency of reporting “several days”, “half the days” and “nearly every day” seems to be reducing in comparison to the baseline.
- The frequency trend of “Not at all” remains consistent with all questions - decreases substantially from the baseline to week 2 and 4 and then rises up in the 8th week. This is exact opposite of the NA trend.
Possible explanation for this trend?
# Function to get the question text based on GAD column name
get_gad_question_text <- function(gad_col) {
question_texts <- list(
GAD_1 = "Feeling nervous, anxious, restless, or uneasy?",
GAD_2 = "Not being able to stop or control worrying about things?",
GAD_3 = "Worrying too much about different things?",
GAD_4 = "Trouble relaxing?",
GAD_5 = "Being so restless or uneasy that it's hard to sit still?",
GAD_6 = "Becoming easily annoyed or irritable?",
GAD_7 = "Feeling afraid as if something bad might happen?",
GAD_Check = "If you read this question, respond 'Not at all'",
GAD_Functioning = "If you experienced any of the problems listed above, how difficult have they made it for you to do your work or get along with other people?"
)
return(question_texts[[gad_col]])
}
# Function to extract the legend from a ggplot
get_legend <- function(myggplot) {
tmp <- ggplotGrob(myggplot)
legend <- tmp$grobs[which(sapply(tmp$grobs, function(x) x$name) == "guide-box")]
return(legend[[1]])
}
# Function to summarize a GAD column by time and create a frequency plot
create_summary_and_plot_gad <- function(data, gad_col) {
# Summarize the GAD column by time and treatment group
freq_data <- data %>%
select(time, !!sym(gad_col), trt) %>%
mutate(score = ifelse(is.na(!!sym(gad_col)), "NA", as.character(!!sym(gad_col)))) %>%
group_by(score, time, trt) %>%
summarise(Frequency = n(), .groups = 'drop')
# Add score labels
freq_data <- freq_data %>%
mutate(score_label = case_when(
score == "0" ~ ifelse(gad_col == "GAD_Functioning", "Not difficult at all", "Not at all"),
score == "1" ~ ifelse(gad_col == "GAD_Functioning", "Somewhat difficult", "Several days (between 2 and 7 days)"),
score == "2" ~ ifelse(gad_col == "GAD_Functioning", "Very difficult", "Over half the days (more than 7 days)"),
score == "3" ~ ifelse(gad_col == "GAD_Functioning", "Extremely difficult", "Nearly/almost every day"),
score == "NA" ~ "NA",
TRUE ~ as.character(score) # For any other values
))
# Pivot the summarized data to wide format
freq_wide <- freq_data %>%
pivot_wider(names_from = time, values_from = Frequency, values_fill = list(Frequency = 0))
# Get the question text for the title
question_text <- get_gad_question_text(gad_col)
# Create plots for each treatment group
plots <- freq_data %>%
split(.$trt) %>%
map(~ ggplot(.x, aes(x = time, y = Frequency, color = score_label, group = score_label)) +
geom_line() +
geom_point() +
labs(title = paste("Treatment Group:", unique(.x$trt)),
x = "Time",
y = "Frequency") +
theme_minimal() +
theme(
plot.title = element_text(hjust = 0.5, size = 12),
axis.title = element_text(size = 10),
axis.text = element_text(size = 8),
legend.position = "none"
) +
scale_x_continuous(breaks = c(0, 2, 4, 8))) # Specify the breaks on x-axis
# Extract the legend from one plot
legend <- get_legend(plots[[1]] + labs(color = "Score") + theme(legend.position = "bottom"))
# Combine the plots and add the common title
combined_plot <- arrangeGrob(grobs = lapply(plots, ggplotGrob), ncol = 2, top = textGrob(
paste("Frequency of Scores for", gad_col, ":", question_text),
gp = gpar(fontsize = 14), hjust = 0.5
))
# Print stargazer table
cat("\n", gad_col, " summary by time and treatment group\n")
stargazer(freq_wide, type = "text", summary = FALSE, rownames = FALSE)
# Combine the plots and legend
final_plot <- arrangeGrob(combined_plot, legend, ncol = 1, heights = c(10, 1))
# Print the final plot
grid.newpage()
grid.draw(final_plot)
}
# List of GAD columns (assuming GAD_1 to GAD_7, GAD_Check, and GAD_Functioning)
gad_columns <- c(paste0("GAD_", 1:7), "GAD_Check", "GAD_Functioning")
# Apply the function to each GAD column
for (gad_col in gad_columns) {
create_summary_and_plot_gad(rct, gad_col)
}##
## GAD_1 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 384 321 377 380
## 0 1 Not at all 412 275 307 376
## 1 0 Several days (between 2 and 7 days) 246 219 235 212
## 1 1 Several days (between 2 and 7 days) 270 235 203 228
## 2 0 Over half the days (more than 7 days) 84 82 70 48
## 2 1 Over half the days (more than 7 days) 83 75 45 45
## 3 0 Nearly/almost every day 90 57 41 67
## 3 1 Nearly/almost every day 62 38 13 45
## NA 0 NA 46 171 127 143
## NA 1 NA 39 243 298 172
## ---------------------------------------------------------------
##
## GAD_2 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 284 267 296 309
## 0 1 Not at all 307 236 232 302
## 1 0 Several days (between 2 and 7 days) 259 226 243 212
## 1 1 Several days (between 2 and 7 days) 269 228 221 226
## 2 0 Over half the days (more than 7 days) 95 86 99 87
## 2 1 Over half the days (more than 7 days) 93 90 66 92
## 3 0 Nearly/almost every day 172 105 83 95
## 3 1 Nearly/almost every day 153 70 42 69
## NA 0 NA 40 166 129 147
## NA 1 NA 44 242 305 177
## ---------------------------------------------------------------
##
## GAD_3 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 248 230 263 282
## 0 1 Not at all 261 183 214 283
## 1 0 Several days (between 2 and 7 days) 260 226 268 208
## 1 1 Several days (between 2 and 7 days) 272 251 200 237
## 2 0 Over half the days (more than 7 days) 118 102 102 103
## 2 1 Over half the days (more than 7 days) 105 100 90 73
## 3 0 Nearly/almost every day 175 122 86 107
## 3 1 Nearly/almost every day 188 88 52 95
## NA 0 NA 49 170 131 150
## NA 1 NA 40 244 310 178
## ---------------------------------------------------------------
##
## GAD_4 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 444 361 389 381
## 0 1 Not at all 461 342 323 382
## 1 0 Several days (between 2 and 7 days) 167 156 170 169
## 1 1 Several days (between 2 and 7 days) 194 146 145 170
## 2 0 Over half the days (more than 7 days) 101 87 86 71
## 2 1 Over half the days (more than 7 days) 85 75 58 76
## 3 0 Nearly/almost every day 91 74 68 79
## 3 1 Nearly/almost every day 76 54 27 54
## NA 0 NA 47 172 137 150
## NA 1 NA 50 249 313 184
## ---------------------------------------------------------------
##
## GAD_5 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 506 423 426 410
## 0 1 Not at all 558 383 351 412
## 1 0 Several days (between 2 and 7 days) 149 138 163 160
## 1 1 Several days (between 2 and 7 days) 139 136 127 171
## 2 0 Over half the days (more than 7 days) 67 67 79 80
## 2 1 Over half the days (more than 7 days) 78 64 54 61
## 3 0 Nearly/almost every day 71 49 43 50
## 3 1 Nearly/almost every day 46 40 22 40
## NA 0 NA 57 173 139 150
## NA 1 NA 45 243 312 182
## ---------------------------------------------------------------
##
## GAD_6 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 336 266 318 292
## 0 1 Not at all 316 241 247 293
## 1 0 Several days (between 2 and 7 days) 223 202 215 213
## 1 1 Several days (between 2 and 7 days) 256 200 178 232
## 2 0 Over half the days (more than 7 days) 88 98 84 82
## 2 1 Over half the days (more than 7 days) 108 84 70 85
## 3 0 Nearly/almost every day 150 101 96 114
## 3 1 Nearly/almost every day 133 96 59 79
## NA 0 NA 53 183 137 149
## NA 1 NA 53 245 312 177
## ---------------------------------------------------------------
##
## GAD_7 summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 283 284 300 296
## 0 1 Not at all 317 227 240 311
## 1 0 Several days (between 2 and 7 days) 278 209 232 214
## 1 1 Several days (between 2 and 7 days) 266 224 174 237
## 2 0 Over half the days (more than 7 days) 94 96 87 106
## 2 1 Over half the days (more than 7 days) 96 89 85 69
## 3 0 Nearly/almost every day 153 94 95 87
## 3 1 Nearly/almost every day 147 87 49 71
## NA 0 NA 42 167 136 147
## NA 1 NA 40 239 318 178
## ---------------------------------------------------------------
##
## GAD_Check summary by time and treatment group
##
## ===============================================================
## score trt score_label 0 2 4 8
## ---------------------------------------------------------------
## 0 0 Not at all 654 594 622 597
## 0 1 Not at all 658 542 480 581
## 1 0 Several days (between 2 and 7 days) 25 23 33 37
## 1 1 Several days (between 2 and 7 days) 41 32 34 49
## 2 0 Over half the days (more than 7 days) 22 12 18 20
## 2 1 Over half the days (more than 7 days) 17 14 9 27
## 3 0 Nearly/almost every day 26 13 19 24
## 3 1 Nearly/almost every day 22 18 16 14
## NA 0 NA 123 208 158 172
## NA 1 NA 128 260 327 195
## ---------------------------------------------------------------
##
## GAD_Functioning summary by time and treatment group
##
## ==============================================
## score trt score_label 0 2 4 8
## ----------------------------------------------
## 0 0 Not difficult at all 209 161 165 223
## 0 1 Not difficult at all 220 153 136 245
## 1 0 Somewhat difficult 306 245 249 286
## 1 1 Somewhat difficult 349 228 190 271
## 2 0 Very difficult 123 69 62 76
## 2 1 Very difficult 109 53 45 72
## 3 0 Extremely difficult 52 39 24 31
## 3 1 Extremely difficult 47 20 8 25
## NA 0 NA 160 336 350 234
## NA 1 NA 141 412 487 253
## ----------------------------------------------
We can note the following:
- NA trends remain consistent among all GAD_1 to GAD_7 questions. And this trend is consistent with PHQ questions as well.
# Function to get the question text based on MSPSS column name
get_mspss_question_text <- function(mspss_col) {
question_texts <- list(
MSPSS_1 = "My family really tries to help me",
MSPSS_2 = "I get the emotional help and support I need from my family",
MSPSS_3 = "My friends really try to help me",
MSPSS_4 = "I can count on my friends when things go wrong",
MSPSS_5 = "I can talk about my problems with my family",
MSPSS_6 = "I have friends with whom I can share my joys and sorrows",
MSPSS_7 = "My family is willing to help me make decisions",
MSPSS_8 = "I can talk about my problems with my friends"
)
return(question_texts[[mspss_col]])
}
# Function to extract the legend from a ggplot
get_legend <- function(myggplot) {
tmp <- ggplotGrob(myggplot)
legend <- tmp$grobs[which(sapply(tmp$grobs, function(x) x$name) == "guide-box")]
return(legend[[1]])
}
# Function to summarize a MSPSS column by time and create a frequency plot
create_summary_and_plot_mspss <- function(data, mspss_col) {
# Summarize the MSPSS column by time and treatment group
freq_data <- data %>%
select(time, !!sym(mspss_col), trt) %>%
mutate(score = ifelse(is.na(!!sym(mspss_col)), "NA", as.character(!!sym(mspss_col)))) %>%
group_by(score, time, trt) %>%
summarise(Frequency = n(), .groups = 'drop')
# Add score labels
freq_data <- freq_data %>%
mutate(score_label = case_when(
score == "1" ~ "I very strongly disagree",
score == "2" ~ "I strongly disagree",
score == "3" ~ "I mildly disagree",
score == "4" ~ "I don’t agree or disagree (neutral)",
score == "5" ~ "I mildly agree",
score == "6" ~ "I strongly agree",
score == "7" ~ "I very strongly agree",
score == "NA" ~ "NA",
TRUE ~ as.character(score) # For any other values
))
# Pivot the summarized data to wide format
freq_wide <- freq_data %>%
pivot_wider(names_from = time, values_from = Frequency, values_fill = list(Frequency = 0))
# Get the question text for the title
question_text <- get_mspss_question_text(mspss_col)
# Create plots for each treatment group
plots <- freq_data %>%
split(.$trt) %>%
map(~ ggplot(.x, aes(x = time, y = Frequency, color = score_label, group = score_label)) +
geom_line() +
geom_point() +
labs(title = paste("Treatment Group:", unique(.x$trt)),
x = "Time",
y = "Frequency") +
theme_minimal() +
theme(
plot.title = element_text(hjust = 0.5, size = 12),
axis.title = element_text(size = 10),
axis.text = element_text(size = 8),
legend.position = "none"
) +
scale_x_continuous(breaks = c(0, 2, 4, 8))) # Specify the breaks on x-axis
# Extract the legend from one plot
legend <- get_legend(plots[[1]] + labs(color = "Score") + theme(legend.position = "bottom"))
# Combine the plots and add the common title
combined_plot <- arrangeGrob(grobs = lapply(plots, ggplotGrob), ncol = 2, top = textGrob(
paste("Frequency of Scores for", mspss_col, ":", question_text),
gp = gpar(fontsize = 14), hjust = 0.5
))
# Print stargazer table
cat("\n", mspss_col, " summary by time and treatment group\n")
stargazer(freq_wide, type = "text", summary = FALSE, rownames = FALSE)
# Combine the plots and legend
final_plot <- arrangeGrob(combined_plot, legend, ncol = 1, heights = c(10, 1))
# Print the final plot
grid.newpage()
grid.draw(final_plot)
}
# List of MSPSS columns (assuming MSPSS_1 to MSPSS_8)
mspss_columns <- paste0("MSPSS_", 1:8)
# Apply the function to each MSPSS column
for (mspss_col in mspss_columns) {
create_summary_and_plot_mspss(rct, mspss_col)
}##
## MSPSS_1 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I very strongly disagree 43 35 37 52
## 1 1 I very strongly disagree 35 31 25 39
## 2 0 I strongly disagree 15 11 17 18
## 2 1 I strongly disagree 12 11 18 12
## 3 0 I mildly disagree 13 9 18 21
## 3 1 I mildly disagree 16 14 12 18
## 4 0 I don’t agree or disagree (neutral) 32 23 23 25
## 4 1 I don’t agree or disagree (neutral) 28 23 19 33
## 5 0 I mildly agree 54 45 49 44
## 5 1 I mildly agree 58 48 35 48
## 6 0 I strongly agree 193 152 156 149
## 6 1 I strongly agree 201 131 128 143
## 7 0 I very strongly agree 476 412 422 398
## 7 1 I very strongly agree 489 372 336 393
## NA 0 NA 24 163 128 143
## NA 1 NA 27 236 293 180
## -------------------------------------------------------------
##
## MSPSS_2 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I very strongly disagree 52 30 34 43
## 1 1 I very strongly disagree 43 27 20 43
## 2 0 I strongly disagree 37 23 29 32
## 2 1 I strongly disagree 35 25 22 24
## 3 0 I mildly disagree 30 19 23 31
## 3 1 I mildly disagree 48 23 19 25
## 4 0 I don’t agree or disagree (neutral) 47 40 31 51
## 4 1 I don’t agree or disagree (neutral) 35 28 17 37
## 5 0 I mildly agree 88 86 67 75
## 5 1 I mildly agree 113 89 59 70
## 6 0 I strongly agree 215 164 173 143
## 6 1 I strongly agree 199 156 152 150
## 7 0 I very strongly agree 343 323 369 329
## 7 1 I very strongly agree 360 280 280 339
## NA 0 NA 38 165 124 146
## NA 1 NA 33 238 297 178
## -------------------------------------------------------------
##
## MSPSS_3 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I very strongly disagree 75 70 64 86
## 1 1 I very strongly disagree 93 64 39 71
## 2 0 I strongly disagree 54 35 42 42
## 2 1 I strongly disagree 46 34 19 32
## 3 0 I mildly disagree 71 50 67 50
## 3 1 I mildly disagree 67 34 42 54
## 4 0 I don’t agree or disagree (neutral) 90 87 78 77
## 4 1 I don’t agree or disagree (neutral) 110 67 54 77
## 5 0 I mildly agree 201 209 179 169
## 5 1 I mildly agree 205 182 164 162
## 6 0 I strongly agree 197 141 158 150
## 6 1 I strongly agree 182 133 133 155
## 7 0 I very strongly agree 127 96 131 133
## 7 1 I very strongly agree 132 109 108 137
## NA 0 NA 35 162 131 143
## NA 1 NA 31 243 307 178
## -------------------------------------------------------------
##
## MSPSS_4 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I very strongly disagree 127 109 109 107
## 1 1 I very strongly disagree 145 93 61 100
## 2 0 I strongly disagree 76 57 55 63
## 2 1 I strongly disagree 72 41 35 46
## 3 0 I mildly disagree 60 75 62 62
## 3 1 I mildly disagree 80 68 50 67
## 4 0 I don’t agree or disagree (neutral) 97 85 96 91
## 4 1 I don’t agree or disagree (neutral) 92 89 77 83
## 5 0 I mildly agree 180 167 159 147
## 5 1 I mildly agree 177 150 127 146
## 6 0 I strongly agree 129 107 118 114
## 6 1 I strongly agree 134 74 112 125
## 7 0 I very strongly agree 138 85 113 120
## 7 1 I very strongly agree 121 105 96 116
## NA 0 NA 43 165 138 146
## NA 1 NA 45 246 308 183
## -------------------------------------------------------------
##
## MSPSS_5 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I very strongly disagree 88 58 55 66
## 1 1 I very strongly disagree 121 54 34 61
## 2 0 I strongly disagree 48 36 39 35
## 2 1 I strongly disagree 55 26 28 37
## 3 0 I mildly disagree 56 39 45 36
## 3 1 I mildly disagree 51 33 28 53
## 4 0 I don’t agree or disagree (neutral) 71 45 53 62
## 4 1 I don’t agree or disagree (neutral) 69 47 44 58
## 5 0 I mildly agree 102 99 90 100
## 5 1 I mildly agree 118 115 89 102
## 6 0 I strongly agree 143 141 152 148
## 6 1 I strongly agree 129 115 118 124
## 7 0 I very strongly agree 287 262 280 253
## 7 1 I very strongly agree 285 226 215 239
## NA 0 NA 55 170 136 150
## NA 1 NA 38 250 310 192
## -------------------------------------------------------------
##
## MSPSS_6 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I very strongly disagree 105 70 92 90
## 1 1 I very strongly disagree 103 75 45 79
## 2 0 I strongly disagree 50 46 34 48
## 2 1 I strongly disagree 51 29 31 31
## 3 0 I mildly disagree 53 42 53 50
## 3 1 I mildly disagree 52 44 39 48
## 4 0 I don’t agree or disagree (neutral) 79 71 74 73
## 4 1 I don’t agree or disagree (neutral) 80 65 67 64
## 5 0 I mildly agree 141 155 152 135
## 5 1 I mildly agree 165 137 118 123
## 6 0 I strongly agree 171 143 153 155
## 6 1 I strongly agree 181 119 121 150
## 7 0 I very strongly agree 203 151 153 154
## 7 1 I very strongly agree 196 153 140 178
## NA 0 NA 48 172 139 145
## NA 1 NA 38 244 305 193
## -------------------------------------------------------------
##
## MSPSS_7 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I very strongly disagree 50 35 34 43
## 1 1 I very strongly disagree 49 25 30 45
## 2 0 I strongly disagree 33 21 29 19
## 2 1 I strongly disagree 28 22 13 20
## 3 0 I mildly disagree 34 17 20 26
## 3 1 I mildly disagree 27 19 15 24
## 4 0 I don’t agree or disagree (neutral) 35 46 42 54
## 4 1 I don’t agree or disagree (neutral) 49 30 32 41
## 5 0 I mildly agree 88 73 70 71
## 5 1 I mildly agree 103 75 69 61
## 6 0 I strongly agree 189 142 162 145
## 6 1 I strongly agree 154 139 132 155
## 7 0 I very strongly agree 382 347 361 345
## 7 1 I very strongly agree 416 310 270 331
## NA 0 NA 39 169 132 147
## NA 1 NA 40 246 305 189
## -------------------------------------------------------------
##
## MSPSS_8 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I very strongly disagree 167 113 106 119
## 1 1 I very strongly disagree 183 100 76 105
## 2 0 I strongly disagree 73 65 60 60
## 2 1 I strongly disagree 75 40 31 50
## 3 0 I mildly disagree 66 58 70 52
## 3 1 I mildly disagree 67 45 36 56
## 4 0 I don’t agree or disagree (neutral) 86 84 95 91
## 4 1 I don’t agree or disagree (neutral) 114 92 73 95
## 5 0 I mildly agree 181 173 161 150
## 5 1 I mildly agree 167 152 139 137
## 6 0 I strongly agree 110 104 113 116
## 6 1 I strongly agree 117 91 102 109
## 7 0 I very strongly agree 130 86 114 116
## 7 1 I very strongly agree 116 106 104 136
## NA 0 NA 37 167 131 146
## NA 1 NA 27 240 305 178
## -------------------------------------------------------------
# Function to get the question text based on GQ column name
get_gq_question_text <- function(gq_col) {
question_texts <- list(
GQ_1 = "I have a lot to be thankful for in my life",
GQ_2 = "I do not see a lot to be thankful for in my community and in the world",
GQ_3 = "I am grateful to many different people",
GQ_4 = "I don’t feel grateful very often",
GQ_5 = "As I get older, I am able to appreciate things that happened in the past",
GQ_6 = "I can write down a long list of things to be grateful for"
)
return(question_texts[[gq_col]])
}
# Function to extract the legend from a ggplot
get_legend <- function(myggplot) {
tmp <- ggplotGrob(myggplot)
legend <- tmp$grobs[which(sapply(tmp$grobs, function(x) x$name) == "guide-box")]
return(legend[[1]])
}
# Function to summarize a GQ column by time and create a frequency plot
create_summary_and_plot_gq <- function(data, gq_col) {
# Summarize the GQ column by time and treatment group
freq_data <- data %>%
select(time, !!sym(gq_col), trt) %>%
mutate(score = ifelse(is.na(!!sym(gq_col)), "NA", as.character(!!sym(gq_col)))) %>%
group_by(score, time, trt) %>%
summarise(Frequency = n(), .groups = 'drop')
# Add score labels
freq_data <- freq_data %>%
mutate(score_label = case_when(
score == "1" ~ "I disagree very much",
score == "2" ~ "I disagree",
score == "3" ~ "I disagree a little",
score == "4" ~ "I don’t agree or disagree (neutral)",
score == "5" ~ "I agree a little",
score == "6" ~ "I agree",
score == "7" ~ "I agree very much",
score == "NA" ~ "NA",
TRUE ~ as.character(score) # For any other values
))
# Pivot the summarized data to wide format
freq_wide <- freq_data %>%
pivot_wider(names_from = time, values_from = Frequency, values_fill = list(Frequency = 0))
# Get the question text for the title
question_text <- get_gq_question_text(gq_col)
# Create plots for each treatment group
plots <- freq_data %>%
split(.$trt) %>%
map(~ ggplot(.x, aes(x = time, y = Frequency, color = score_label, group = score_label)) +
geom_line() +
geom_point() +
labs(title = paste("Treatment Group:", unique(.x$trt)),
x = "Time",
y = "Frequency") +
theme_minimal() +
theme(
plot.title = element_text(hjust = 0.5, size = 12),
axis.title = element_text(size = 10),
axis.text = element_text(size = 8),
legend.position = "none"
) +
scale_x_continuous(breaks = c(0, 2, 4, 8))) # Specify the breaks on x-axis
# Extract the legend from one plot
legend <- get_legend(plots[[1]] + labs(color = "Score") + theme(legend.position = "bottom"))
# Combine the plots and add the common title
combined_plot <- arrangeGrob(grobs = lapply(plots, ggplotGrob), ncol = 2, top = textGrob(
paste("Frequency of Scores for", gq_col, ":", question_text),
gp = gpar(fontsize = 14), hjust = 0.5
))
# Print stargazer table
cat("\n", gq_col, " summary by time and treatment group\n")
stargazer(freq_wide, type = "text", summary = FALSE, rownames = FALSE)
# Combine the plots and legend
final_plot <- arrangeGrob(combined_plot, legend, ncol = 1, heights = c(10, 1))
# Print the final plot
grid.newpage()
grid.draw(final_plot)
}
# List of GQ columns (assuming GQ_1 to GQ_6)
gq_columns <- paste0("GQ_", 1:6)
# Apply the function to each GQ column
for (gq_col in gq_columns) {
create_summary_and_plot_gq(rct, gq_col)
}##
## GQ_1 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I disagree very much 23 14 16 24
## 1 1 I disagree very much 18 9 7 13
## 2 0 I disagree 7 4 5 9
## 2 1 I disagree 16 2 6 4
## 3 0 I disagree a little 8 2 3 4
## 3 1 I disagree a little 10 5 5 7
## 4 0 I don’t agree or disagree (neutral) 15 14 14 17
## 4 1 I don’t agree or disagree (neutral) 13 8 11 12
## 5 0 I agree a little 18 27 26 31
## 5 1 I agree a little 43 32 26 38
## 6 0 I agree 220 157 168 170
## 6 1 I agree 234 183 137 176
## 7 0 I agree very much 522 463 487 450
## 7 1 I agree very much 501 386 381 434
## NA 0 NA 37 169 131 145
## NA 1 NA 31 241 293 182
## -------------------------------------------------------------
##
## GQ_2 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I disagree very much 49 34 47 39
## 1 1 I disagree very much 42 34 27 50
## 2 0 I disagree 76 38 58 57
## 2 1 I disagree 81 38 46 53
## 3 0 I disagree a little 102 84 79 60
## 3 1 I disagree a little 108 90 49 82
## 4 0 I don’t agree or disagree (neutral) 98 74 74 91
## 4 1 I don’t agree or disagree (neutral) 92 71 64 67
## 5 0 I agree a little 45 53 40 40
## 5 1 I agree a little 68 43 44 39
## 6 0 I agree 179 147 154 134
## 6 1 I agree 198 147 132 151
## 7 0 I agree very much 252 249 263 277
## 7 1 I agree very much 239 197 204 245
## NA 0 NA 49 171 135 152
## NA 1 NA 38 246 300 179
## -------------------------------------------------------------
##
## GQ_3 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I disagree very much 24 15 24 25
## 1 1 I disagree very much 21 13 10 18
## 2 0 I disagree 27 19 10 23
## 2 1 I disagree 29 13 6 23
## 3 0 I disagree a little 14 18 19 11
## 3 1 I disagree a little 27 15 15 10
## 4 0 I don’t agree or disagree (neutral) 35 39 39 36
## 4 1 I don’t agree or disagree (neutral) 61 39 21 43
## 5 0 I agree a little 99 86 95 99
## 5 1 I agree a little 122 88 60 84
## 6 0 I agree 296 256 235 222
## 6 1 I agree 295 209 197 214
## 7 0 I agree very much 308 248 288 285
## 7 1 I agree very much 273 241 251 291
## NA 0 NA 47 169 140 149
## NA 1 NA 38 248 306 183
## -------------------------------------------------------------
##
## GQ_4 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I disagree very much 51 41 52 62
## 1 1 I disagree very much 62 48 38 52
## 2 0 I disagree 122 92 75 79
## 2 1 I disagree 105 81 53 61
## 3 0 I disagree a little 144 100 91 87
## 3 1 I disagree a little 122 89 81 102
## 4 0 I don’t agree or disagree (neutral) 77 82 104 94
## 4 1 I don’t agree or disagree (neutral) 113 60 65 75
## 5 0 I agree a little 69 60 47 56
## 5 1 I agree a little 78 66 48 55
## 6 0 I agree 180 147 166 135
## 6 1 I agree 182 135 140 146
## 7 0 I agree very much 158 154 174 190
## 7 1 I agree very much 147 133 138 190
## NA 0 NA 49 174 141 147
## NA 1 NA 57 254 303 185
## -------------------------------------------------------------
##
## GQ_5 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I disagree very much 43 20 29 41
## 1 1 I disagree very much 49 24 14 30
## 2 0 I disagree 44 20 19 22
## 2 1 I disagree 55 28 18 20
## 3 0 I disagree a little 22 11 15 15
## 3 1 I disagree a little 29 13 16 20
## 4 0 I don’t agree or disagree (neutral) 50 39 29 52
## 4 1 I don’t agree or disagree (neutral) 57 35 31 49
## 5 0 I agree a little 85 68 62 61
## 5 1 I agree a little 86 68 56 69
## 6 0 I agree 234 211 201 188
## 6 1 I agree 221 183 140 162
## 7 0 I agree very much 333 315 365 323
## 7 1 I agree very much 331 261 294 336
## NA 0 NA 39 166 130 148
## NA 1 NA 38 254 297 180
## -------------------------------------------------------------
##
## GQ_6 summary by time and treatment group
##
## =============================================================
## score trt score_label 0 2 4 8
## -------------------------------------------------------------
## 1 0 I disagree very much 40 23 31 41
## 1 1 I disagree very much 40 25 9 36
## 2 0 I disagree 26 15 20 25
## 2 1 I disagree 39 18 15 19
## 3 0 I disagree a little 17 9 19 16
## 3 1 I disagree a little 40 19 19 18
## 4 0 I don’t agree or disagree (neutral) 55 43 35 36
## 4 1 I don’t agree or disagree (neutral) 42 35 26 37
## 5 0 I agree a little 106 75 81 81
## 5 1 I agree a little 115 85 61 66
## 6 0 I agree 237 183 191 168
## 6 1 I agree 227 176 146 167
## 7 0 I agree very much 324 339 344 338
## 7 1 I agree very much 331 267 293 347
## NA 0 NA 45 163 129 145
## NA 1 NA 32 241 297 176
## -------------------------------------------------------------
# Function to get the question text based on SWEMWBS column name
get_swemwbs_question_text <- function(swemwbs_col) {
question_texts <- list(
SWEMWBS_1 = "Over the last two weeks, I’ve been feeling optimistic about the future",
SWEMWBS_2 = "Over the last two weeks, I’ve been feeling useful",
SWEMWBS_3 = "Over the last two weeks, I’ve been feeling relaxed",
SWEMWBS_4 = "Over the last two weeks, I’ve been dealing with problems well",
SWEMWBS_5 = "Over the last two weeks, I’ve been thinking clearly",
SWEMWBS_6 = "Over the last two weeks, I’ve been feeling close to other people",
SWEMWBS_7 = "Over the last two weeks, I’ve been able to make up my own mind about things"
)
return(question_texts[[swemwbs_col]])
}
# Function to extract the legend from a ggplot
get_legend <- function(myggplot) {
tmp <- ggplotGrob(myggplot)
legend <- tmp$grobs[which(sapply(tmp$grobs, function(x) x$name) == "guide-box")]
return(legend[[1]])
}
# Function to summarize a SWEMWBS column by time and create a frequency plot
create_summary_and_plot_swemwbs <- function(data, swemwbs_col) {
# Summarize the SWEMWBS column by time and treatment group
freq_data <- data %>%
select(time, !!sym(swemwbs_col), trt) %>%
mutate(score = ifelse(is.na(!!sym(swemwbs_col)), "NA", as.character(!!sym(swemwbs_col)))) %>%
group_by(score, time, trt) %>%
summarise(Frequency = n(), .groups = 'drop')
# Add score labels
freq_data <- freq_data %>%
mutate(score_label = case_when(
score == "1" ~ "None of the time",
score == "2" ~ "Rarely",
score == "3" ~ "Some of the time",
score == "4" ~ "Often",
score == "5" ~ "All of the time",
score == "NA" ~ "NA",
TRUE ~ as.character(score) # For any other values
))
# Pivot the summarized data to wide format
freq_wide <- freq_data %>%
pivot_wider(names_from = time, values_from = Frequency, values_fill = list(Frequency = 0))
# Get the question text for the title
question_text <- get_swemwbs_question_text(swemwbs_col)
# Create plots for each treatment group
plots <- freq_data %>%
split(.$trt) %>%
map(~ ggplot(.x, aes(x = time, y = Frequency, color = score_label, group = score_label)) +
geom_line() +
geom_point() +
labs(title = paste("Treatment Group:", unique(.x$trt)),
x = "Time",
y = "Frequency") +
theme_minimal() +
theme(
plot.title = element_text(hjust = 0.5, size = 12),
axis.title = element_text(size = 10),
axis.text = element_text(size = 8),
legend.position = "none"
) +
scale_x_continuous(breaks = c(0, 2, 4, 8))) # Specify the breaks on x-axis
# Extract the legend from one plot
legend <- get_legend(plots[[1]] + labs(color = "Score") + theme(legend.position = "bottom"))
# Combine the plots and add the common title
combined_plot <- arrangeGrob(grobs = lapply(plots, ggplotGrob), ncol = 2, top = textGrob(
paste("Frequency of Scores for", swemwbs_col, ":", question_text),
gp = gpar(fontsize = 14), hjust = 0.5
))
# Print stargazer table
cat("\n", swemwbs_col, " summary by time and treatment group\n")
stargazer(freq_wide, type = "text", summary = FALSE, rownames = FALSE)
# Combine the plots and legend
final_plot <- arrangeGrob(combined_plot, legend, ncol = 1, heights = c(10, 1))
# Print the final plot
grid.newpage()
grid.draw(final_plot)
}
# List of SWEMWBS columns (assuming SWEMWBS_1 to SWEMWBS_7)
swemwbs_columns <- paste0("SWEMWBS_", 1:7)
# Apply the function to each SWEMWBS column
for (swemwbs_col in swemwbs_columns) {
create_summary_and_plot_swemwbs(rct, swemwbs_col)
}##
## SWEMWBS_1 summary by time and treatment group
##
## ==========================================
## score trt score_label 0 2 4 8
## ------------------------------------------
## 1 0 None of the time 110 98 133 143
## 1 1 None of the time 93 96 110 120
## 2 0 Rarely 111 119 130 102
## 2 1 Rarely 129 118 132 137
## 3 0 Some of the time 227 166 156 152
## 3 1 Some of the time 263 157 120 162
## 4 0 Often 145 131 132 118
## 4 1 Often 134 127 102 134
## 5 0 All of the time 198 160 158 176
## 5 1 All of the time 181 114 97 123
## NA 0 NA 59 176 141 159
## NA 1 NA 66 254 305 190
## ------------------------------------------
##
## SWEMWBS_2 summary by time and treatment group
##
## ==========================================
## score trt score_label 0 2 4 8
## ------------------------------------------
## 1 0 None of the time 131 91 99 97
## 1 1 None of the time 130 68 64 88
## 2 0 Rarely 121 95 87 98
## 2 1 Rarely 124 82 69 96
## 3 0 Some of the time 196 174 154 152
## 3 1 Some of the time 217 167 112 148
## 4 0 Often 165 140 166 157
## 4 1 Often 168 146 135 139
## 5 0 All of the time 184 185 204 193
## 5 1 All of the time 184 155 180 213
## NA 0 NA 53 165 140 153
## NA 1 NA 43 248 306 182
## ------------------------------------------
##
## SWEMWBS_3 summary by time and treatment group
##
## ==========================================
## score trt score_label 0 2 4 8
## ------------------------------------------
## 1 0 None of the time 138 104 93 89
## 1 1 None of the time 139 71 56 111
## 2 0 Rarely 164 131 148 134
## 2 1 Rarely 195 131 111 111
## 3 0 Some of the time 258 210 187 189
## 3 1 Some of the time 235 174 157 159
## 4 0 Often 118 125 126 152
## 4 1 Often 142 123 119 136
## 5 0 All of the time 112 107 149 132
## 5 1 All of the time 99 118 120 158
## NA 0 NA 60 173 147 154
## NA 1 NA 56 249 303 191
## ------------------------------------------
##
## SWEMWBS_4 summary by time and treatment group
##
## ==========================================
## score trt score_label 0 2 4 8
## ------------------------------------------
## 1 0 None of the time 130 77 79 94
## 1 1 None of the time 116 56 57 88
## 2 0 Rarely 158 112 110 106
## 2 1 Rarely 145 118 77 93
## 3 0 Some of the time 252 191 191 167
## 3 1 Some of the time 235 142 127 172
## 4 0 Often 138 139 141 158
## 4 1 Often 182 143 141 122
## 5 0 All of the time 119 156 180 174
## 5 1 All of the time 129 154 157 199
## NA 0 NA 53 175 149 151
## NA 1 NA 59 253 307 192
## ------------------------------------------
##
## SWEMWBS_5 summary by time and treatment group
##
## ==========================================
## score trt score_label 0 2 4 8
## ------------------------------------------
## 1 0 None of the time 64 35 44 59
## 1 1 None of the time 63 40 26 59
## 2 0 Rarely 109 82 78 79
## 2 1 Rarely 112 69 68 73
## 3 0 Some of the time 240 179 183 164
## 3 1 Some of the time 241 152 124 156
## 4 0 Often 153 151 144 166
## 4 1 Often 159 158 138 133
## 5 0 All of the time 233 233 248 225
## 5 1 All of the time 235 198 202 251
## NA 0 NA 51 170 153 157
## NA 1 NA 56 249 308 194
## ------------------------------------------
##
## SWEMWBS_6 summary by time and treatment group
##
## ==========================================
## score trt score_label 0 2 4 8
## ------------------------------------------
## 1 0 None of the time 94 82 74 79
## 1 1 None of the time 96 52 49 69
## 2 0 Rarely 143 99 109 114
## 2 1 Rarely 134 107 79 104
## 3 0 Some of the time 217 165 163 176
## 3 1 Some of the time 228 148 121 154
## 4 0 Often 151 150 158 145
## 4 1 Often 168 128 130 152
## 5 0 All of the time 193 186 200 183
## 5 1 All of the time 192 182 180 203
## NA 0 NA 52 168 146 153
## NA 1 NA 48 249 307 184
## ------------------------------------------
##
## SWEMWBS_7 summary by time and treatment group
##
## ==========================================
## score trt score_label 0 2 4 8
## ------------------------------------------
## 1 0 None of the time 61 39 44 53
## 1 1 None of the time 64 34 36 48
## 2 0 Rarely 58 71 64 57
## 2 1 Rarely 70 53 49 49
## 3 0 Some of the time 195 153 146 155
## 3 1 Some of the time 209 124 109 162
## 4 0 Often 192 137 154 153
## 4 1 Often 175 143 126 147
## 5 0 All of the time 300 288 301 281
## 5 1 All of the time 309 271 246 278
## NA 0 NA 44 162 141 151
## NA 1 NA 39 241 300 182
## ------------------------------------------
# Function to get the question text based on SES column name
get_ses_question_text <- function(ses_col) {
question_texts <- c(
SES_1 = "I follow the rules at school",
SES_2 = "I get in trouble at school",
SES_3 = "When I am in class, I just pretend as if I am working",
SES_4 = "I pay attention in class",
SES_5 = "I complete my work on time",
SES_6 = "I like being at school",
SES_7 = "I feel excited by my work at school",
SES_8 = "My classroom is a fun place to be",
SES_9 = "I am interested in the work at school",
SES_10 = "I feel happy in school",
SES_11 = "I feel bored in school",
SES_12 = "I check my schoolwork for mistakes",
SES_13 = "I study at home even when I don’t have a test",
SES_14 = "I try to watch TV shows about things we do in school",
SES_15 = "When I read a book, I ask myself questions to make sure I understand what it is about",
SES_16 = "I read extra books to learn more about things we do in school",
SES_17 = "If I don’t know what a word means when I am reading, I do something to figure it out",
SES_18 = "If I don’t understand what I read, I go back and read it over again",
SES_19 = "I talk with people outside of school about what I am learning in class"
)
return(question_texts[[ses_col]])
}
# Function to extract the legend from a ggplot
get_legend <- function(myggplot) {
tmp <- ggplotGrob(myggplot)
legend <- tmp$grobs[which(sapply(tmp$grobs, function(x) x$name) == "guide-box")]
return(legend[[1]])
}
# Function to summarize an SES column by time and create a frequency plot
create_summary_and_plot_ses <- function(data, ses_col) {
# Summarize the SES column by time and treatment group
freq_data <- data %>%
select(time, !!sym(ses_col), trt) %>%
mutate(score = ifelse(is.na(!!sym(ses_col)), "NA", as.character(!!sym(ses_col)))) %>%
group_by(score, time, trt) %>%
summarise(Frequency = n(), .groups = 'drop')
# Add score labels
freq_data <- freq_data %>%
mutate(score_label = case_when(
score == "1" ~ "Not true",
score == "2" ~ "(Blank)",
score == "3" ~ "Somewhat true",
score == "4" ~ "(Blank)",
score == "5" ~ "Very true",
score == "NA" ~ "NA",
TRUE ~ as.character(score) # For any other values
))
# Pivot the summarized data to wide format
freq_wide <- freq_data %>%
pivot_wider(names_from = time, values_from = Frequency, values_fill = list(Frequency = 0))
# Get the question text for the title
question_text <- get_ses_question_text(ses_col)
# Create plots for each treatment group
plots <- freq_data %>%
split(.$trt) %>%
map(~ ggplot(.x, aes(x = time, y = Frequency, color = score_label, group = score_label)) +
geom_line() +
geom_point() +
labs(title = paste("Treatment Group:", unique(.x$trt)),
x = "Time",
y = "Frequency") +
theme_minimal() +
theme(
plot.title = element_text(hjust = 0.5, size = 12),
axis.title = element_text(size = 10),
axis.text = element_text(size = 8),
legend.position = "none"
) +
scale_x_continuous(breaks = c(0, 2, 4, 8))) # Specify the breaks on x-axis
# Extract the legend from one plot
legend <- get_legend(plots[[1]] + labs(color = "Score") + theme(legend.position = "bottom"))
# Combine the plots and add the common title
combined_plot <- arrangeGrob(grobs = lapply(plots, ggplotGrob), ncol = 2, top = textGrob(
paste("Frequency of Scores for", ses_col, ":", question_text),
gp = gpar(fontsize = 14), hjust = 0.5
))
# Print stargazer table
cat("\n", ses_col, " summary by time and treatment group\n")
stargazer(freq_wide, type = "text", summary = FALSE, rownames = FALSE)
# Combine the plots and legend
final_plot <- arrangeGrob(combined_plot, legend, ncol = 1, heights = c(10, 1))
# Print the final plot
grid.newpage()
grid.draw(final_plot)
}
# List of SES columns (assuming SES_1 to SES_19)
ses_columns <- paste0("SES_", 1:19)
# Apply the function to each SES column
for (ses_col in ses_columns) {
create_summary_and_plot_ses(rct, ses_col)
}##
## SES_1 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 29 32 43 48
## 1 1 Not true 40 18 21 42
## 2 0 (Blank) 6 5 13 16
## 2 1 (Blank) 6 6 5 7
## 3 0 Somewhat true 168 145 149 155
## 3 1 Somewhat true 201 155 125 143
## 4 0 (Blank) 20 12 15 19
## 4 1 (Blank) 17 14 17 22
## 5 0 Very true 589 496 503 469
## 5 1 Very true 577 434 399 470
## NA 0 NA 38 160 127 143
## NA 1 NA 25 239 299 182
## ---------------------------------------
##
## SES_2 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 49 37 37 55
## 1 1 Not true 57 36 31 49
## 2 0 (Blank) 11 8 10 13
## 2 1 (Blank) 10 2 7 11
## 3 0 Somewhat true 123 128 132 141
## 3 1 Somewhat true 152 125 103 137
## 4 0 (Blank) 25 26 33 34
## 4 1 (Blank) 20 24 13 30
## 5 0 Very true 600 486 500 460
## 5 1 Very true 590 437 407 455
## NA 0 NA 42 165 138 147
## NA 1 NA 37 242 305 184
## ---------------------------------------
##
## SES_3 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 89 52 60 50
## 1 1 Not true 74 45 51 68
## 2 0 (Blank) 9 11 14 18
## 2 1 (Blank) 8 11 6 4
## 3 0 Somewhat true 125 107 136 123
## 3 1 Somewhat true 162 120 93 119
## 4 0 (Blank) 21 20 23 28
## 4 1 (Blank) 15 17 14 24
## 5 0 Very true 561 496 471 480
## 5 1 Very true 571 429 391 459
## NA 0 NA 45 164 146 151
## NA 1 NA 36 244 311 192
## ---------------------------------------
##
## SES_4 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 29 25 28 28
## 1 1 Not true 45 23 17 39
## 2 0 (Blank) 12 9 18 13
## 2 1 (Blank) 6 11 9 9
## 3 0 Somewhat true 237 188 162 188
## 3 1 Somewhat true 274 165 128 187
## 4 0 (Blank) 23 31 24 37
## 4 1 (Blank) 24 17 22 13
## 5 0 Very true 503 434 486 435
## 5 1 Very true 487 409 381 430
## NA 0 NA 46 163 132 149
## NA 1 NA 30 241 309 188
## ---------------------------------------
##
## SES_5 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 99 70 67 59
## 1 1 Not true 120 57 51 62
## 2 0 (Blank) 12 11 11 18
## 2 1 (Blank) 8 12 8 18
## 3 0 Somewhat true 272 232 217 212
## 3 1 Somewhat true 328 220 191 216
## 4 0 (Blank) 18 30 27 40
## 4 1 (Blank) 21 30 18 24
## 5 0 Very true 403 347 388 369
## 5 1 Very true 352 308 294 355
## NA 0 NA 46 160 140 152
## NA 1 NA 37 239 304 191
## ---------------------------------------
##
## SES_6 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 77 70 69 88
## 1 1 Not true 89 75 52 81
## 2 0 (Blank) 17 15 20 26
## 2 1 (Blank) 14 15 8 8
## 3 0 Somewhat true 148 145 136 136
## 3 1 Somewhat true 170 112 120 142
## 4 0 (Blank) 17 26 29 15
## 4 1 (Blank) 18 28 16 19
## 5 0 Very true 547 432 464 436
## 5 1 Very true 540 394 366 429
## NA 0 NA 44 162 132 149
## NA 1 NA 35 242 304 187
## ---------------------------------------
##
## SES_7 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 69 72 70 69
## 1 1 Not true 97 66 58 72
## 2 0 (Blank) 18 20 14 17
## 2 1 (Blank) 17 19 10 14
## 3 0 Somewhat true 204 165 176 172
## 3 1 Somewhat true 220 163 123 168
## 4 0 (Blank) 24 24 26 34
## 4 1 (Blank) 22 21 19 23
## 5 0 Very true 479 403 428 407
## 5 1 Very true 458 352 347 401
## NA 0 NA 56 166 136 151
## NA 1 NA 52 245 309 188
## ---------------------------------------
##
## SES_8 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 129 113 119 94
## 1 1 Not true 142 90 85 96
## 2 0 (Blank) 23 21 14 17
## 2 1 (Blank) 6 12 10 12
## 3 0 Somewhat true 160 148 144 164
## 3 1 Somewhat true 181 126 112 146
## 4 0 (Blank) 15 16 19 25
## 4 1 (Blank) 26 28 14 25
## 5 0 Very true 469 382 418 397
## 5 1 Very true 466 372 335 393
## NA 0 NA 54 170 136 153
## NA 1 NA 45 238 310 194
## ---------------------------------------
##
## SES_9 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 67 65 65 81
## 1 1 Not true 85 67 50 72
## 2 0 (Blank) 15 15 17 23
## 2 1 (Blank) 11 11 11 16
## 3 0 Somewhat true 211 181 175 154
## 3 1 Somewhat true 228 163 119 171
## 4 0 (Blank) 14 24 26 38
## 4 1 (Blank) 23 32 21 31
## 5 0 Very true 485 401 421 402
## 5 1 Very true 470 344 358 383
## NA 0 NA 58 164 146 152
## NA 1 NA 49 249 307 193
## ---------------------------------------
##
## SES_10 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 65 65 73 76
## 1 1 Not true 73 71 40 82
## 2 0 (Blank) 9 15 24 17
## 2 1 (Blank) 10 11 11 9
## 3 0 Somewhat true 184 150 128 146
## 3 1 Somewhat true 190 126 100 140
## 4 0 (Blank) 21 19 19 36
## 4 1 (Blank) 20 16 15 25
## 5 0 Very true 516 438 470 422
## 5 1 Very true 529 397 387 420
## NA 0 NA 55 163 136 153
## NA 1 NA 44 245 313 190
## ---------------------------------------
##
## SES_11 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 103 75 71 94
## 1 1 Not true 85 81 71 110
## 2 0 (Blank) 17 11 18 28
## 2 1 (Blank) 15 8 10 15
## 3 0 Somewhat true 167 144 171 177
## 3 1 Somewhat true 174 114 120 132
## 4 0 (Blank) 22 29 26 33
## 4 1 (Blank) 23 16 19 32
## 5 0 Very true 480 420 423 364
## 5 1 Very true 522 400 340 382
## NA 0 NA 61 171 141 154
## NA 1 NA 47 247 306 195
## ---------------------------------------
##
## SES_12 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 212 171 158 152
## 1 1 Not true 259 173 152 174
## 2 0 (Blank) 30 32 25 22
## 2 1 (Blank) 17 26 15 22
## 3 0 Somewhat true 237 225 232 221
## 3 1 Somewhat true 277 193 164 191
## 4 0 (Blank) 22 21 21 39
## 4 1 (Blank) 23 19 29 29
## 5 0 Very true 272 238 271 259
## 5 1 Very true 229 212 197 256
## NA 0 NA 77 163 143 157
## NA 1 NA 61 243 309 194
## ---------------------------------------
##
## SES_13 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 166 127 114 112
## 1 1 Not true 215 118 97 120
## 2 0 (Blank) 22 17 28 19
## 2 1 (Blank) 19 18 12 20
## 3 0 Somewhat true 243 212 226 219
## 3 1 Somewhat true 256 179 184 207
## 4 0 (Blank) 13 20 27 32
## 4 1 (Blank) 16 23 23 21
## 5 0 Very true 351 296 310 315
## 5 1 Very true 303 278 238 311
## NA 0 NA 55 178 145 153
## NA 1 NA 57 250 312 187
## ---------------------------------------
##
## SES_14 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 280 212 186 183
## 1 1 Not true 343 207 155 174
## 2 0 (Blank) 20 21 28 27
## 2 1 (Blank) 18 17 16 28
## 3 0 Somewhat true 191 197 209 209
## 3 1 Somewhat true 197 153 148 184
## 4 0 (Blank) 18 24 22 35
## 4 1 (Blank) 13 23 18 23
## 5 0 Very true 272 223 254 243
## 5 1 Very true 251 224 219 267
## NA 0 NA 69 173 151 153
## NA 1 NA 44 242 310 190
## ---------------------------------------
##
## SES_15 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 90 71 79 70
## 1 1 Not true 115 75 55 67
## 2 0 (Blank) 22 24 16 16
## 2 1 (Blank) 13 13 14 14
## 3 0 Somewhat true 162 171 193 185
## 3 1 Somewhat true 211 168 158 169
## 4 0 (Blank) 30 20 24 36
## 4 1 (Blank) 26 14 19 27
## 5 0 Very true 493 387 393 389
## 5 1 Very true 444 349 307 395
## NA 0 NA 53 177 145 154
## NA 1 NA 57 247 313 194
## ---------------------------------------
##
## SES_16 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 135 102 110 86
## 1 1 Not true 193 103 80 90
## 2 0 (Blank) 25 25 20 23
## 2 1 (Blank) 22 12 11 16
## 3 0 Somewhat true 229 210 188 228
## 3 1 Somewhat true 232 172 169 194
## 4 0 (Blank) 30 36 40 35
## 4 1 (Blank) 19 23 29 25
## 5 0 Very true 370 311 350 330
## 5 1 Very true 352 307 269 355
## NA 0 NA 61 166 142 148
## NA 1 NA 48 249 308 186
## ---------------------------------------
##
## SES_17 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 112 68 94 99
## 1 1 Not true 136 92 71 93
## 2 0 (Blank) 12 24 17 14
## 2 1 (Blank) 8 10 10 15
## 3 0 Somewhat true 163 190 176 173
## 3 1 Somewhat true 197 156 152 183
## 4 0 (Blank) 23 20 32 33
## 4 1 (Blank) 29 22 17 17
## 5 0 Very true 471 378 388 379
## 5 1 Very true 430 326 304 374
## NA 0 NA 69 170 143 152
## NA 1 NA 66 260 312 184
## ---------------------------------------
##
## SES_18 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 66 58 63 63
## 1 1 Not true 83 45 51 62
## 2 0 (Blank) 13 16 17 16
## 2 1 (Blank) 9 14 5 15
## 3 0 Somewhat true 157 154 142 163
## 3 1 Somewhat true 179 144 119 175
## 4 0 (Blank) 18 24 32 35
## 4 1 (Blank) 17 18 27 23
## 5 0 Very true 541 432 450 423
## 5 1 Very true 535 400 352 410
## NA 0 NA 55 166 146 150
## NA 1 NA 43 245 312 181
## ---------------------------------------
##
## SES_19 summary by time and treatment group
##
## =======================================
## score trt score_label 0 2 4 8
## ---------------------------------------
## 1 0 Not true 223 159 148 148
## 1 1 Not true 263 170 115 145
## 2 0 (Blank) 11 16 19 20
## 2 1 (Blank) 14 6 9 16
## 3 0 Somewhat true 194 218 205 198
## 3 1 Somewhat true 220 147 159 186
## 4 0 (Blank) 27 23 31 32
## 4 1 (Blank) 21 25 24 21
## 5 0 Very true 348 273 312 303
## 5 1 Very true 311 281 255 322
## NA 0 NA 47 161 135 149
## NA 1 NA 37 237 304 176
## ---------------------------------------