---
title: "Summary Statistics Tables for Abalone Clean"
author: "Melinda Higgins"
date: "4/22/2022"
output:
  html_document: default
  word_document: default
editor_options: 
  chunk_output_type: console
---

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

load(file = "abalone_clean.RData")
```

# Summary Statistics Tables for Cleaned Abalone Dataset

## Summary stats using the `arsenal` package

This is a quick summary statistics table for the cleaned abalone dataset made using the `arsenal package.

NOTE: You need to add `results = "asis"` in the R chunk option in order for this table to render correctly when "knitted".

Learn more at [https://cran.r-project.org/web/packages/arsenal/vignettes/tableby.html](https://cran.r-project.org/web/packages/arsenal/vignettes/tableby.html).

### Look at sex, length, height, diameter by adult category

Note: By default categorical data will perform a chi-square test for differences in the groups and ANOVA (or t-test for 2 groups) is run for the continuous data.

```{r results="asis"}
library(arsenal)
tab1 <- tableby(adult ~ sex + length + diameter + height,
                data = abalone_clean)
summary(tab1, pfootnote = TRUE)
```

### Look at whole, shucked, viscera, shell weights by adult category

```{r results="asis"}
tab1 <- tableby(adult ~ wholeWeight + shuckedWeight +
                  visceraWeight + shellWeight,
                data = abalone_clean)
summary(tab1, pfootnote = TRUE)
```

## Another option with the `gtsummary` package

I really like the `arsenal` package since it works really well for knitting to WORD, but the `gtsummary` also works ok with WORD and HTML and other formats also. And there seems to be more active development happening for the `gtsummary` package so it may be the future.

Learn more at [https://www.danieldsjoberg.com/gtsummary/index.html](https://www.danieldsjoberg.com/gtsummary/index.html).

### Look at sex, length, height, diameter by adult category

Note: In the code below, I specifically designated that all categorical data would perform a chi-square test and for the continuous data the t-test is performed, see [https://www.danieldsjoberg.com/gtsummary/reference/add_p.tbl_summary.html](https://www.danieldsjoberg.com/gtsummary/reference/add_p.tbl_summary.html).

I also changed the default statistics. The default is to report the median and IQR, I changed it to mean and SD - see [https://www.danieldsjoberg.com/gtsummary/reference/tbl_summary.html](https://www.danieldsjoberg.com/gtsummary/reference/tbl_summary.html).

```{r}
library(gtsummary)

# make dataset with a few variables to summarize
ab2 <- abalone_clean %>% 
  select(sex, adult, length, diameter, height)

table2 <- 
  tbl_summary(
    ab2,
    by = adult, 
    statistic = list(all_continuous() ~ "{mean} ({sd})"),
    missing = "no" 
  ) %>%
  add_n() %>% 
  add_p(test = list(all_continuous() ~ "t.test",
                    all_categorical() ~ "chisq.test")) %>% 
  modify_header(label = "**Variable**") %>% 
  bold_labels() 

table2
```

### Look at whole, shucked, viscera, shell weights by adult category

In the code below, I changed the test to the non-parametrc wilcox.test.

```{r}
# make dataset with a few variables to summarize
ab3 <- abalone_clean %>% 
  select(adult, wholeWeight, shuckedWeight, 
         visceraWeight, shellWeight)

table3 <- 
  tbl_summary(
    ab3,
    by = adult, 
    missing = "no" 
  ) %>%
  add_n() %>% 
  add_p(test = list(all_continuous() ~ "wilcox.test")) %>% 
  modify_header(label = "**Variable**") %>% 
  bold_labels() 

table3
```