Project

Replication and Extension Project Description

NEW DESCRIPTION OF PAPER

In this project, you will work to reproduce and extend a published research article. The primary goal of this project is to take a published research paper and to replicate all of the tables and statistical models, then to extend these models for your own research question.

The secondary goals of this project are to

  1. Give you practice learning how to evaluate published research.
  2. Give you practice interpreting a research question into statistical methods.
  3. Give you practice learning apply statistical methods and computational techniques.
  4. Display the true complexities to working with real data and and an open ended question.

Project Requirements

For the project, you must follow the schedule and meet certian technical requirements.

Follow the Calendar Below for Deadlines. Note: these are in addition to regular homework

Deadline Task Description
2/22/2019 Pick Data that you will be using Choose a partner and submit a document on canvas with 3 potential papers to replicate
3/8/2019 Get your plan approved We will ask you to verify that you have the data by this point. It may not be easy to gather the data
4/3/2019 First Steps Replication Project including:
RMarkdown file of characteristic tables and figures
Data which you are using
PDF of original Paper
You will need to give the 2 other groups whose projects you were assigned to a zipped copy of this.
4/10/2019 Peer Review 1 Each group will be peer reviewing 2 others. E-mail zipped folder to that team and the teaching staff including:
All of the original files
"Files with comments on the replication status
4/17/2019 2nd Steps Create an outline of how you plan to extend the work in the paper as well as finish up replicating parts you were unable to previously.
Outline the tables and figures you wish to include into your proposal.
Submit a zipped folder to peer review groups and teaching staff including:
Rmarkdown file with tables and figures including outline and description of new tables and figures.
All of the previous files as well.
4/24/2019 Peer Review 2 Each group will be peer reviewing 2 others. E-mail zipped folder to that team and the teaching staff including:
All of the original files
"Files with comments on the replication status
5/1/2019 Final Steps Complete the rest of your coding as well as your additional models.
Submit a zipped folder to peer review groups and teaching staff including:
Rmarkdown file with tables and figures.
All of the previous files as well.
5/8/2019 Rough Draft Send your Rough draft to the peer review groups.
Paper Description
5/10/2019 Final Peer Review Due
5/11/2019 Final Paper Due Final Review of the Papers
Be sure to comment on changes made from past iterations.
Comment on completeness and quality.

Pick a Paper or Data That you Will be Using

You should pick a paper or data that is interesting to you. There are many data sources that you can consider.

If you want to replicate a paper, you should try to find a paper that has data available. Make sure the methods used in the paper are what is being used in this class.

If you are looking for papers to replicate here are some strategies that you could use:

Feel free to browse examples of papers that have been replicated by students in the past:

Get Data and Have Plan Approved

Before you get too involved in this project you will submit a one page summary of:

  • What Project you plan to do.
  • What paper/data you plan to use.
  • Question of interest.
  • Ideas of what types of Analysis you will need to perform

Include the following in this proposal:

  • Citation information for the paper/data.
  • List of main statistical methods used
  • List of datasets used Were these datasets collected via complex sample designs?
  • Summary of data availability and data access plan
  • Summary of code availability (may be helpful for replication paper)
  • Questions that you wish to answer above and beyond research
  • Short explanation for why you picked this paper/data

Replication of Paper

Your goal is to reproduce a paper. You need to try reproduce the results exactly. You should be able to re-create every table and graph in the paper. Focus on the replicating the content of the tables and not the layout.

Begin your replication by creating a document with images of each table and each figure. Next, add the parts of the text where the authors describe how the results were generated. Finally, create code that reproduces the results. This document will be highly structured. For example,

  • Image of Table 1
  • Text describing how Table 1 was created
  • Your code to reproduce Table 1

Use the structure described below:

Remember you should somehow extend or improve the paper you are reproducing. Here are some examples of how you could do that:

  • create different visualizations of their data and results.
  • extend the analysis by using different methods or asking different questions.
  • re-run the same analysis on completely different data.
  • take an old paper and update the analysis with newer data.

For more help you can read this paper by Gary King

Suggested Replication further reading:

▲