codebook.Rmd

---
title: "SHINE study codebook"
author: "Dani Cosme"
date: "`r Sys.Date()`"
output:
  html_document:
    code_folding: hide
    df_print: paged
    highlight: tango
    theme: united
    toc: yes
    toc_float:
      collapsed: yes
      smooth_scroll: yes
  pdf_document:
    toc: yes
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE, message = FALSE, warning = FALSE)
options(scipen = 999)
```

```{r, include = FALSE}
# prep
# load packages
if (!require(pacman)) {
  install.packages('pacman')
}

pacman::p_load(tidyverse, haven, sjlabelled, knitr)

# load data
baseline = read_sav("surveys/baseline.sav")
prescan = read_sav("surveys/prescan.sav")
postscan = read_sav("surveys/postscan.sav")
covid = read_sav("surveys/covid.sav")
followup_6months = read_sav("surveys/followup_6months.sav")
followup_12months = read_sav("surveys/followup_12months.sav")

ignore_items = read.csv("ignore_items.csv", stringsAsFactors = FALSE)
item_info = read.csv("items_info.csv", stringsAsFactors = FALSE)
missing = read.csv("missing.csv", stringsAsFactors = FALSE) %>%
  mutate(item = as.character(item))

# tidy data
choice_options = enframe(get_labels(baseline)) %>% mutate(session = "baseline") %>%
  bind_rows(enframe(get_labels(prescan)) %>% mutate(session = "prescan")) %>%
  bind_rows(enframe(get_labels(postscan)) %>% mutate(session = "postscan")) %>%
  bind_rows(enframe(get_labels(covid)) %>% mutate(session = "covid")) %>%
  bind_rows(enframe(get_labels(followup_6months)) %>% mutate(session = "followup_6months")) %>%
  bind_rows(enframe(get_labels(followup_12months)) %>% mutate(session = "followup_12months")) %>%
  rename("choice_options" = value)

codebook_all = enframe(get_label(baseline)) %>% mutate(session = "baseline") %>%
  bind_rows(enframe(get_label(prescan)) %>% mutate(session = "prescan")) %>%
  bind_rows(enframe(get_label(postscan)) %>% mutate(session = "postscan")) %>%
  bind_rows(enframe(get_label(covid)) %>% mutate(session = "covid")) %>%
  bind_rows(enframe(get_label(followup_6months)) %>% mutate(session = "followup_6months")) %>%
  bind_rows(enframe(get_label(followup_12months)) %>% mutate(session = "followup_12months")) %>%
  left_join(., choice_options) %>%
  extract(value, c("instruction", "item_text"), "(.*) - (.*)", remove = FALSE) %>%
  mutate(instruction = ifelse(is.na(instruction), value, instruction),
         name = gsub("Q64", "ERQ", name),
         name = gsub("erq_1", "ERQ", name),
         name = gsub("Q75", "fs", name),
         name = gsub("Q31", "drink_intent", name),
         name = gsub("Q33", "alc_change", name),
         name = gsub("Q38|purpose_in_life1", "PILS_1", name),
         name = gsub("psychDrugs", "psych_drugs", name),
         name = gsub("loneliness_UCLA|ucla", "ULS", name),
         name = gsub("flourishing|fs|fs_1", "FS", name),
         name = gsub("cesd|Q70", "CES_D", name),
         name = gsub("irq", "IRQ", name),
         name = gsub("lotr", "LOTR", name),
         name = gsub("ftp", "FTP", name),
         name = gsub("sise", "SISE", name),
         name = gsub("srhi", "SRHI", name),
         name = gsub("ipaq", "IPAQ", name),
         name = gsub("pls", "PILS", name),
         name = gsub("stai", "STAI_state", name),
         name = gsub("intolerance", "IUS", name),
         name = gsub("autonomy", "IAF", name),
         name = gsub("IPAQ", "IPAQ_", name),
         name = gsub("tipi", "TIPI", name),
         item_text = gsub("Alpha Phi Omega", "[group]", item_text),
         item_text = gsub("Field-group", "group", item_text),
         instruction = gsub("from 1 (not at all) to 7 (very much).", "from 1(not at all) to 7(very much)", instruction),
         instruction = gsub("Alpha Phi Omega", "[group]", instruction),
         instruction = gsub("Field-group", "group", instruction),
         instruction = ifelse(name == "spirits_freq", "On how many days per week do you drink spirits (whisky, vodka, gin, rum, etc - but not beer or wine)?", instruction),
         instruction = ifelse(name == "CES_D", "Below is a list of some of the ways you may have felt or behaved. Please indicate how often you have felt this way during the past week by checking the appropriate box for each question.", instruction),
         choice_options = ifelse(name %in% c("grm_role", "eyes_rx2", "grm_year", "grm_memberStat", "housing_city",
                                             "housing_country", "perspective1", "mindful1", "react1", "physdis",
                                             "spirits_brands", "wine_brands", "beer_brands", "weight", "height",
                                             "psych_drugs_names", "psych_drugs_dose", "psych_drugs_length"), NA, choice_options),
         choice_options = ifelse(grepl("IPAQ|other_drugs|mhealth_|drugs_today_", name), NA, choice_options),
         choice_options = ifelse(grepl("ladder", name), list(c("Off", "On")), choice_options),
         choice_options = ifelse(name == "physdis_3", "text entry", choice_options),
         choice_options = ifelse(name == "housing_member_fut", "text entry", choice_options),
         choice_options = ifelse(instruction == "When did you join [group]?", "text entry", choice_options),
         choice_options = as.character(choice_options),
         choice_options = gsub("c\\(", "", choice_options),
         choice_options = gsub("\\)$", "", choice_options),
         choice_options = gsub('\\"', "", choice_options)) %>%
  filter(!name %in% ignore_items$name & !grepl("TEXT|photo", name)) %>%
  extract(name, c("survey", "item"), c("(.*)_([0-9]+)$"), remove = FALSE) %>%
  mutate(survey = ifelse(is.na(survey), name, survey)) %>%
  select(-name, -value)

# fix individual survey information
rpi = read.csv("rpi.csv", stringsAsFactors = FALSE)

prc_if = codebook_all %>%
  filter(grepl("prc|^if_", survey)) %>%
  mutate(item_text = "[name]",
         item = NA) %>%
  unique()

friendly_universe = codebook_all %>%
  filter(grepl("fu_", survey)) %>%
  mutate(item_text = "[name]",
         item = NA,
         choice_options = ifelse(grepl("age|category|networkname", survey), "text entry", choice_options)) %>%
  filter(!grepl("fu_channel_[2-9]|fu_channel_[1]+", survey)) %>%
  unique() %>%
  select(-item)

full_network_survey = read.csv("fns.csv", stringsAsFactors = FALSE)

# full_network_survey = codebook_all %>%
#   filter(grepl("fns|alcMost|closest|alcLeast|close_risk|risk_notclose|partnerGroup", survey)) %>%
#   mutate(instruction = gsub("\\?.*", "", instruction),
#          item_text = "[name]",
#          item = NA,
#          choice_options = ifelse(grepl("^alc|amount|freq|adjust|face|phone|text|closest_interact|partnerGroup", survey), choice_options, NA)) %>%
#   unique() %>%
#   select(-item)

codebook = codebook_all %>%
  filter(!grepl("RPI|prc|^if_|fns|alcMost|closest|alcLeast|close_risk|risk_notclose|fu_|partnerGroup", survey)) %>%
  filter(!survey %in% missing$survey) %>%
  bind_rows(., rpi, prc_if, missing) %>%
  filter(!grepl("close_risk|risk_notclose", survey)) %>%
  filter(!(grepl("prc", survey) & grepl("NULL", choice_options))) %>%
  mutate(single_item = ifelse((is.na(item_text) | item_text == "Selected Choice" | survey == "alc_change"), 1, 0),
         item_text = ifelse(single_item == 1, instruction, item_text),
         item_text = gsub("Selected Choice ", "", item_text),
         item = ifelse(item_text == "I lead a purposeful and meaningful life", 1, item),
         single_item = ifelse(grepl("IPAQ", survey), 0, single_item), 
         item = ifelse(single_item == 1, NA, item),
         instruction = ifelse(single_item == 1, "", instruction)) %>%
  select(-single_item) %>%
  left_join(., item_info) %>%
  group_by(session, survey) %>%
  mutate(item = ifelse(!is.na(item) & !survey == "RPI", row_number(), NA)) %>%
  select(survey_name, survey_citation, survey, everything()) %>%
  ungroup()

```

<br><br>

# Full network survey 
Codebook for items in the full network survey. Participants saw members of their campus social group and made nominations. 

<br><br>

```{r, echo = FALSE}
full_network_survey %>%
  filter(!survey %in% missing$survey) %>%
  DT::datatable(filter = "top")
```

<br><br><br><br>


# Friendly Universe ego-network
Codebook for items in the self-generated [Friendly Universe ego-network survey](https://github.com/ruipei1/friendlyUniverse). 

<br><br>

```{r, echo = FALSE}
friendly_universe %>%
  DT::datatable(filter = "top")
```

<br><br><br><br>

# Non-network questionnaires {.tabset}
Codebook for questionnaire items. Items rated using slider scales are noted and anchor points are provided.

<br><br>

```{r, include = FALSE}
print_chunk = function(scales) {
  for (variable in scales){
    
    data_scale = data %>%
      filter(survey_name == !!variable)
    
    surveys = unique(data_scale$survey)
    
    citation = as.array(unique(data_scale$survey_citation)[!unique(data_scale$survey_citation) == ""])
    
    instructions_all = as.array(unique(data_scale$instruction))
    
    items = as.array(unique(data_scale$item)[!unique(data_scale$item) == ""])
    
    same_scale = unique(data_scale$same_instructions)
    
    data_print = data_scale %>%
      select(-survey_name, -survey_citation, -session, -same_instructions)
    
    cat(paste0('\n\n### ', variable, '{.tabset}\n\n')) 
    
    cat('<br>')
    
    if (length(citation) == 1) {
      
      cat(paste0("**Citation**\n\n"), citation, "\n\n")
      
      cat('<br>')
      
      }
    
    
    if (!is.na(same_scale) & length(instructions_all) == 1) {
      cat(paste0("**Instructions**\n\n"), instructions_all, "\n\n")
      
      cat('<br>')
      
      cat(knitr::knit_print(DT::datatable(data_print %>%
                                          select(-instruction),
                                          width = "100%")))
      
    } else if (length(instructions_all) == 1 & !instructions_all == "" & length(items) == 1) {
      cat(paste0("**Instructions**\n\n"), instructions_all, "\n\n")
      
      cat('<br>')
      
      cat(knitr::knit_print(DT::datatable(data_print %>%
                                          select(-instruction, -item),
                                          width = "100%")))
      
    } else if (instructions_all == "" & length(items) > 1) {
      cat(knitr::knit_print(DT::datatable(data_print %>%
                                          select(-instruction),
                                          width = "100%")))
    } else if (instructions_all == "" & length(items) == 1) {
      cat(knitr::knit_print(DT::datatable(data_print %>%
                                          select(-instruction, -item),
                                          width = "100%")))      
    } else {
      cat(knitr::knit_print(DT::datatable(data_print,
                                          width = "100%")))
    }
  }
}
```

## baseline {.tabset}

<br><br>

```{r, results='asis', echo = FALSE}
data = codebook %>%
  filter(session == "baseline")

scales = unique(data$survey_name)

print_chunk(scales)
```

## prescan {.tabset}

<br>

```{r, results='asis', echo = FALSE}
data = codebook %>%
  filter(session == "prescan")

scales = unique(data$survey_name)

print_chunk(scales)
```

## COVID-19 {.tabset}

<br>

```{r, results='asis', echo = FALSE}
data = codebook %>%
  filter(session == "covid")

scales = unique(data$survey_name)

print_chunk(scales)
```

## postscan {.tabset}

<br>

```{r, results='asis', echo = FALSE}
data = codebook %>%
  filter(session == "postscan")

scales = unique(data$survey_name)

print_chunk(scales)
```

## followup: 6 months {.tabset}

<br>

```{r, results='asis', echo = FALSE}
data = codebook %>%
  filter(session == "followup_6months")

scales = unique(data$survey_name)

print_chunk(scales)
```

## followup: 12 months {.tabset}

<br>

```{r, results='asis', echo = FALSE}
data = codebook %>%
  filter(session == "followup_12months")

scales = unique(data$survey_name)

print_chunk(scales)
```