Title

An Exploration of Multiple Tools for Creating Reproducible Research

Presenter Information/ Coauthors Information

Adam Sullivan, Brown UniversityFollow

Presentation Type

Event

Track

Tools

Abstract

In a great deal of statistical methods we employ the Central Limit Theorem which we assume to be true if we replicated our research over and over again. In the traditional research world, this assumption was made but rarely was there a way to truly replicate work.

The world of Data Science changes this paradigm. All areas of work from business to health care collect more types of data on a more regular basis. This should lend us the ability to reproduce results and test our assumptions. How do we adequately set up a data workflow to allow for things to be more reproducible? How do we encourage sharing of findings and code so that others in our fields or companies can replicate and reproduce our work?

This presentation will discuss many strategies to creating reproducible research, both from the lense of a researcher preparing their data workflow to allow them to reproduce but also tools for sharing this research to allow others to reproduce it as well.

We will look further into tools like RStudio, RCloud Social, Jupyterhub and Github. All provide platforms for sharing and collaborating on projects while making reproducible research easy to employ.

Start Date

5-2-2019 10:00 AM

End Date

5-2-2019 10:50 AM

This document is currently not available here.

Share

Import Event to Google Calendar

COinS
 
Feb 5th, 10:00 AM Feb 5th, 10:50 AM

An Exploration of Multiple Tools for Creating Reproducible Research

Pasque 255

In a great deal of statistical methods we employ the Central Limit Theorem which we assume to be true if we replicated our research over and over again. In the traditional research world, this assumption was made but rarely was there a way to truly replicate work.

The world of Data Science changes this paradigm. All areas of work from business to health care collect more types of data on a more regular basis. This should lend us the ability to reproduce results and test our assumptions. How do we adequately set up a data workflow to allow for things to be more reproducible? How do we encourage sharing of findings and code so that others in our fields or companies can replicate and reproduce our work?

This presentation will discuss many strategies to creating reproducible research, both from the lense of a researcher preparing their data workflow to allow them to reproduce but also tools for sharing this research to allow others to reproduce it as well.

We will look further into tools like RStudio, RCloud Social, Jupyterhub and Github. All provide platforms for sharing and collaborating on projects while making reproducible research easy to employ.