This repository contains files for the BST 260 final project by Lauren Flynn, Anna Lai, Kehuan Lin, and Carolin Schulte, completed in Fall 2021.
All main analyses are included in BST260_final. The supplementary information in the different appendices is referenced as appropriate.
-
BST260_final.Rmd: R Markdown file containing an overview of project background and motivation, all main analyses (code and figures) and a discussion of our findings.
-
BST260_final.html: BST260_final.Rmd knitted to html
-
AppendixA.Rmd: Code for merging NHANES data for 2007 - 2018 and re-coding variables of interest; all .XPT files were obtained from the CDC website for NHANES and are not included as part of this repository (file will not knit without those data). The output is 'masterDF.Rda', which is included in the Data folder.
-
AppendixA.html: AppendixA.Rmd knitted to html
-
AppendixB.Rmd: Code for creating graphs of NHANES data for exploratory data analysis (in addition to analyses performed in BST260_final) and assessing missingness for covariates of interest
-
AppendixB.html: AppendixB.Rmd knitted to html
-
AppendixC.Rmd: Code for fitting all machine learning models described in this study
-
AppendixC.html: AppendixC.Rmd knitted to html
-
AppendixD.Rmd: Code for running sensitivity analyses for the inclusion of survey weights in the logistic regression models
-
AppendixD.html: AppendixD.Rmd knitted to html
-
GlobalMortalityData2014.csv: Data on global country-specific mortality due to eating disorders in 2014; based on data published by the WHO
-
masterDF.Rda: R data object containing combined NHANES data for 2007 - 2018; required for all data analyses using NHANES data. The data was originally accessed through the CDC website for NHANES.
-
app.R: Code for the Shiny app; in order to download this app, please download the data file filtered_for_shiny.Rda. The app is also available online at https://laurenflynn.shinyapps.io/project/
-
filtered_for_shiny.Rda: NHANES data which has been filtered to include the variables necessary for the Shiny app (app.R).
-
filtering_for_shiny.Rmd: Selects variables of interest and filters out missing values in order to create filtered_for_shiny.Rda
Our video is available at https://youtu.be/TFpKMJgDkFU
Our website is available at https://sites.google.com/view/bst260-nhanes