Reproducible Research: Concepts and Ideas (part 3)

Loading...

Do curso por Johns Hopkins University

Reproducible Research

2953 classificações

Johns Hopkins University

Reproducible Research

2953 classificações

Curso 5 de 10 em Specialization Data Science

This course focuses on the concepts and tools behind reporting modern data analyses in a reproducible manner. Reproducible research is the idea that data analyses, and more generally, scientific claims, are published with their data and software code so that others may verify the findings and build upon them. The need for reproducibility is increasing dramatically as data analyses become more complex, involving larger datasets and more sophisticated computations. Reproducibility allows for people to focus on the actual content of a data analysis, rather than on superficial details reported in a written summary. In addition, reproducibility makes an analysis more useful to others because the data and code that actually conducted the analysis are available. This course will focus on literate statistical analysis tools which allow one to publish data analyses in a single document that allows others to easily execute the same analysis to obtain the same results.

Na lição

Week 1: Concepts, Ideas, & Structure

This week will cover the basic ideas of reproducible research since they may be unfamiliar to some of you. We also cover structuring and organizing a data analysis to help make it more reproducible. I recommend that you watch the videos in the order that they are listed on the web page, but watching the videos out of order isn't going to ruin the story.

Introduction1:57

What is Reproducible Research About?8:19

Reproducible Research: Concepts and Ideas (part 1)7:11

Reproducible Research: Concepts and Ideas (part 2) 5:26

Reproducible Research: Concepts and Ideas (part 3) 3:26

Scripting Your Analysis 4:35

Structure of a Data Analysis (part 1)12:29

Structure of a Data Analysis (part 2)17:40

Organizing Your Analysis11:05

Conheça os instrutores

Roger D. Peng, PhD
Associate Professor, Biostatistics
Bloomberg School of Public Health
Jeff Leek, PhD
Associate Professor, Biostatistics
Bloomberg School of Public Health
Brian Caffo, PhD
Professor, Biostatistics
Bloomberg School of Public Health

One basic idea, to kind of, as a tool, to make a

lot of this stuff easier, is what's known as Literate (Statistical) Programing.

And, this comes from the idea of literate programming, in computer programs.

And so the idea is to think of the art-, an article,

or a publication, or a report, as a stream of text and code.

And, so the text is, kind of, readable

by people and the code is readable by computers.

And then the idea is that the analysis is

described in a series of text and code chunks.

And each kind of code chunk will do something.

It'll load some data and compute results.

And each text chunk will kind of relay something in a human readable language.

And so there are going to be presentation

code, that kind of formats tables and figures.

There's article text that explains what's going on around all this code.

And the idea is that this stream of text and code

is a literate statistical program or a liter, literate statistical analysis.

And these programs can be weaved to

produce human readable documents like PDFs or

HTML web pages, and they can tangled

to produce machine-readable documents, so machine readable code.

And so the basic idea behind literate programming; is that if you need a

documentational language, that's human readable and you

need a programming language that's machine readable.

And so one of the original system's in R

that was designed to do this was called Sweave.

And Sweave uses a documentation program called

LaTeX, and a programming language, obviously is R.

And it was developed by Fritz Leisch who was a core member of

R, and is still maintained by R Core and his website is listed here.

Now there are one, there are many limitations to the original Sweave system.

One of which is focused primarily on LaTeX which is not

a language, a documentation language that many, many people are familiar with.

And so it can be difficult to learn this type

of markup language if you're not all ready in the field.

It lacks a lot of features that kind of people want

like caching, and multiple plots per page and mixing programming languages.

And so and it's not as frequently updated or very actively developed.

And so one of the alternative that has come

up in recent times is, is something called knitr.

And the knitr package for R is basically, kind of, takes a lot of

these ideas of literate programming and kind

of updates them and improves upon them.

And there are a lot of kind of features that

were added on to the Sweave original kind of concept.

And so it's still, knitr still uses R as its programming language,

but it allows you to mix a v, other programming languages in.

And you can use a variety of documentation languages now.

You can use LaTex, you can also use

something called Markdown and you can use HTML.

And so knitr was developed by Yihui Xie who was,

who was, while he was a graduate student at Iowa State.

And it's become a very popular package for doing literate statistical programs.

So just to summarize very briefly.

Reproducible research is a I, I think it's an

important minimum standard for,

computationally intensive, types of analysis.

And where replication is very difficult or, if not impossible.

We still need quite a bit of infrastructure and tools to kind

of create and distribute reproducible documents,

and beyond what is currently available.

But that's kind of improving as the, everyday and there

are lots of new tools kind of coming on the scene.

So in the next lecture, I'll talk about some of these tools.

In particular I'll talk about knitr.

And just to show how you can produce reproducible documents.

Explore nosso catálogo

Registre-se gratuitamente e obtenha recomendações, atualizações e ofertas personalizadas.

O Coursera proporciona acesso universal à melhor educação do mundo fazendo parcerias com as melhores universidades e organizações para oferecer cursos on-line.

© 2019 Coursera Inc. Todos os direitos reservados.

Baixar na App Store

Baixar no Google Play