[Statistical Analysis with R] Overview

Overview and help for R

R introduction: http://r-tutorial.nl

R download and overview: https://cran.r-project.org

R packages: https://cran.r-project.org/web/packages

R journal: https://journal.r-project.org

RStudio cheatsheets: www.rstudio.com/resources/cheatsheets  

Overview: www.rstudio.com/online-learning

*For publicly available datasets for research: https://datasetsearch.research.google.com/

 

Download and installation of R

Download and install most recent, archived, and development versions from https://cran.r-project.org/.

RStudio: Download and install from www.rstudio.com/products/rstudio/download 

Help for the installation can be found e.g. at http://r-tutorial.nl/.

Update R e.g. using the updateR() function (and other functions) in the install r package (https://cran.r-project.org/web/packages/installr).

 

Background 

1. Empirical research process; analysis & interpretation

Theoretical Q <--------R------> Empirical observations

-Operationalization of Constructs

-Data collection

-Data organization & digitalization

-Data check & manipulation

-Data analysis --> Description of sample (Descriptive Statistics) / Use sample to make conclusion/draw inference about general population

-Interpretation of results --> Is the sample representative? / Are the results convincing?

 

2. R & RStudio history

-1980s: S, S-PLUS developed by Becker & Chambers.

-Ross Ihaka & Rober Gentleman (University of Auckland): development of reduced version of S:R.

-New version updated approx. every 2~3 months ; current version: 4.0.3

R: ~ Engine

-Statistical computer program

-Complete programming language

-Environment to perform statistical analyses, produce graphics

>It's free, used a lot in life sciences, public health, etc.

>Very good people working on the development and adding new things (e.g. Hadley Wickham) – e.g. work in the cloud (https://rstudio.cloud)

 

RStudio: ~ Program

-Nice graphical interface, with additional options and functionalities

 

Other programs:

SPSS: can do all analysis by "clicking" without programming/with programming, not an open source

SAS: can do some analysis by clicking/generally made for programming. High costs

Python: Similar to R: increasing user base and applications, especially in machine learning

>used more in computer science, machine learning

 

Introduction, overview:

R for Data Science: http://r4ds.had.co.nz/index.html

Dalgaard (2008). Introductory Statistics with R.

Crawley (2012). The R book.

 

Advanced R, Graphics in R, and further references:

Advanced R: https://adv-r.hadley.nz/

Matloff (2011). The art of R programming.

ggplot2: Elegant Graphics for Data Analysis: http://had.co.nz/ggplot2/

Dynamic Documents with R and knitr: https://github.com/yihui/knitr- book

Practical Regression and Anova using R: https://cran.r- project.org/doc/contrib/Faraway- PRA.pdf

 

Packages contains different category of functions: search after packages/what you are looking for

         >> eg. https://cran.r-project.org/web/packages/ggplot2/index.html

ggplot2: Create Elegant Data Visualisations Using the Grammar of Graphics

 

In order to remove an assigned object or dataset, need to use rm(object or dataset) command.