diff --git a/R_examples/EDA.Rmd b/R_examples/EDA.Rmd
index 28f89d0..ec597ab 100644
--- a/R_examples/EDA.Rmd
+++ b/R_examples/EDA.Rmd
@@ -10,9 +10,7 @@ knitr::opts_chunk$set(echo = TRUE)
 library('ggplot2')
 library('scales')
 setwd("~/Research/cdsc_examples_repository/R_examples")
-#load('../processed_data/artDF.RData')
-load('~/Research/taboo/processed_data/EDA.RData') 
-#need vDF, qDF, revDF.clean
+load('~/Research/cdsc_examples_repository/R_examples/data/EDA.RData')
 
 reverselog_trans <- function(base = exp(1)) {
     trans <- function(x) -log(x, base)
@@ -209,14 +207,8 @@ hist(log(artDF$revid))
 #hist(log(artDF$got_reverted))
 #hist(log(artDF$revid))
 
-## try a wilcoxen/mann-whitney u test
-###both are skewed, need to do a rank correlation.
 cor.test(artDF$pct_revert, as.numeric(artDF$taboo))
 cor.test(artDF$revid, artDF$pct_revert)
-##build the R plots of all the visualizations and distributions
-## do the same type of EDA for article-level dataset
-## every time there is skew think through tests/models to match
-## log-log, hists and boxplots, colored by taboo, etc.
 
 #drop Barack Obama
 artDF <- subset(artDF, artDF$taboo!='0.5')