sample description.Rmd

---
title: "analysis data summary table"
author: "Sina Özdemir"
date: "9/17/2021"
output: word_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = F,message = F)
library(tidyverse)
library(gtsummary)
data<- readRDS(file = "C:/Users/sinaoz/OneDrive - NTNU/Projects/communication_quality_repo/data/AnalyticData_AllSamples.RDS")

```

```{r io_personal}

io_data<- data %>% filter(tweetsample == "IO")


io_screennames<- io_data %>% pull(screen_name) %>% unique()

io_indv<- c("JHPedraza","AUBinetaDiop","GuyCyrille","AU_Chergui","NEPAD_Mayaki","AUC_MoussaFaki","AmbFMadeira")


io_data$tweetsample<- ifelse(io_data$screen_name%in%io_indv,"IO (pers. account)","IO (inst. account)")

data<- anti_join(data, io_data, by = "screen_name")

data <- rbind(data,io_data)

```


```{r descriptive_table}


data$pers<- grepl(pattern = "(pers. account)",x = data$tweetsample,fixed = T)
data$pers<- ifelse(data$pers == T, "Personal account","Institutional Account")

data$sample<- ifelse(str_detect(data$tweetsample,"EU"),"EU",
                     ifelse(str_detect(data$tweetsample,"IO"),"IO",
                            ifelse(str_detect(data$tweetsample,"Random"),"Random tweets",
                                   ifelse(str_detect(data$tweetsample,"UK"),"UK","Unidentified"))))


data %>% select(pers,sample) %>% tbl_summary(by = pers)
```

```{r account_type_summary}
data %>%
  select(screen_name,pers,sample) %>%
  distinct(screen_name,.keep_all = T) %>%
  filter(sample != "Random tweets") %>%
  select(-screen_name) %>% 
  tbl_summary(by = pers)


```