04_omegaPLoR.Rmd

---
title: "04_omegaPLoR"
output: html_document
---

__Code written:__ 2019-12-02  
__Last run:__ `r Sys.Date()`   
__Authors:__ Navona Calarco 
__Website:__ https://rpubs.com/navona/Mar_categoricalOmegaReading
__Git repo:__ https://github.com/navonacalarco/NRC_Mar 

```{r setup}

#clean environment
rm(list = ls(all.names = TRUE))

#set working directory
knitr::opts_knit$set(root.dir = '../data/out/PLoR')

```

```{r}

#libraries
library('MBESS')
library('dplyr')

#read in all PLoR csvs
all_dfs = list.files(pattern="*.csv")
for (i in 1:length(all_dfs)) assign(all_dfs[i], read.csv(all_dfs[i]))

#fn for omega -- hierarchical as 7 levels
omega_fn <- function(df){
  ci.reliability(df,  type='hierarchical', interval.type = 'bca', B=1000, conf.level=.95 )
}

#apply fn to all PLoR dfs -- results will have same name
cCmd  <- paste(all_dfs , "<- omega_fn(" ,all_dfs,")", sep="")
eCmd  <- parse(text=cCmd)
eval(eCmd)

```

```{r calculations for table}

#write fn to find missing values -- and replace with average of participant's scores  score
replaceMissing_fn <- function(df){
  df <- as.matrix(df)
  missing <- which(is.na(df), arr.ind=TRUE)
  df[missing] <- rowMeans(df, na.rm=TRUE)[missing[,1]]
  df <- as.data.frame(df)
}

#apply fn to all PLoR dfs -- results will have same name
cCmd  <- paste(all_dfs , "<- replaceMissing_fn(" ,all_dfs,")", sep="")
eCmd  <- parse(text=cCmd)
eval(eCmd)

#row sums function
rowSums_fn = function(df){
    df = mutate(df, total=rowSums(df))
    return(df)
}

#apply fn to all PLoR dfs -- results will have same name
cCmd  <- paste(all_dfs , "<- rowSums_fn(" ,all_dfs,")", sep="")
eCmd  <- parse(text=cCmd)
eval(eCmd)

#turn all dataframes into a list
df.list <- Filter(function(x) is(x, "data.frame"), mget(ls()))

#for each dataframe in the list:
means <- t(as.data.frame(lapply(df.list, function (x) mean(x[,'total'], na.rm= TRUE)))) 
sds   <- t(as.data.frame(lapply(df.list, function (x) sd(x[,'total'], na.rm= TRUE))))
min   <- t(as.data.frame(lapply(df.list, function (x) min(x[,'total'], na.rm= TRUE))))
max   <- t(as.data.frame(lapply(df.list, function (x) max(x[,'total'], na.rm= TRUE))))

```