Transition to R

R logoThe goal of this class is to help grad students, postdocs, or faculty who have a background in basic statistics and a familiarity with some other statistics package (JMP, SYSTAT, SAS, SPSS) to become comfortable with the R Project as a platform for statistical analyses. It is not meant as a course in statistics, nor does it cover more than a small portion of what is available in R. At the end of the course you should be comfortable managing data in R, making graphs, performing an array of different basic statistical analyses, and be able to use the extensive resources available on line and in books to learn to do just about anything statistical you'd like to in R. The course is offered periodically, when enough UCSC grad students request it, but you can take the course on your own!  The class is primarily based on the handouts below, and videos of the 2015 version lectures are available for UCSC students.
IMPORTANT NOTE ON LINKS TO DATA.  The website has moved since this course was last offered, so the URLs to datasets in the handouts are broken.  Please note, however, that all the datasets are still available, below.  You can download them, or get their current URLs by right-clicking on the links, and then modifying the links in the Rcode.  I will be updating this soon, but in the meantime...

Syllabus, class notes, and class videos

ENVS291 Transition To R Syllabus W2015.pdf

Here are the detailed handouts, and in-class videos of the lectures and conversations from the 2015 version of ENVS291 Transition to R.  Use at your own risk.  All handout and video contents are the 2015 copyright of Gregory S. Gilbert, University of California, Santa Cruz

Class1_GettingStarted.pdf     LectureVideo

Installing R;  R windows; loading libraries; R Types, structures, and objects; R terminology; basic R notation; common operators;  exploring R objects; referencing inside data frames, vectors, and matrices; importing data from a spreadsheet

Class2_Handling Data.PDF          LectureVideo  

Creating, editing, reading, and exporting dataframes; sorting, subseetting, combining data; importing and exporting data; handling dates

Class3_Summarizing Data.PDF           LectureVideo

Summarizing continuous data; functions in R; Creating customized complex functions; Summarize continuous data in groups; Apply functions across rows or columns;  by, aggregate, tapply, apply, sapply, lapply

Classes 4 and 5 Regression, ANOVA, tTest.PDF        LectureVideo4         LectureVideo5

Fitting models; Extractor functions; linear regression using lm; ANOVA using aov or lm; ttest and rank tests; factorial, blocked, and split-plot designs; ANCOVA; Homogeneity of variance; Type I and III sums of squares; predict for fitted lines and confidence intervals; stepwise selection

Class6_BasicPlottingTools.PDF      LectureVideo

Base plotting tools; plot, boxplot, hist;  plot overlays;  par function to control plot attributes; exporting graphs

Class7  Functions and Loops.PDF        LectureVideo

Custom functions; Basics of programming; Conditional statements; Loops;  Writing scripts in R

Class8 GLMs and MixedModels.PDF       LectureVideo

Generalized Linear models (glm); error structure, linear predictors, link functions; Logistic regression; Poisson regression; Survival Analysis; Mixed Models

Class9 Advanced Graphics with ggplot.PDF      LectureVideo

Graphs with ggplot2; data, aesthetic attributes, mapping, geoms, stats, facets, layers, viewports, themes, etc.

Class10a Vegan Community Ecology Analysis.PDF        no class video

Basics of Vegan package for analysis of community structure and diversity data.  Measures of diversity; NMDS   NOT UPDATED SINCE 2013; SOME PARTS ARE OUT OF DATE.

Class10b Picate and Phylomatic  Phylogenetic Ecology tools.PDF       no lecture video

Very brief introduction to tools for phylogenetic community ecology and trait analysis.  NOT UPDATED SINCE 2013  SOME THINGS ARE OUT OF DATE!   Newick tree format; making phylogenetic trees from phylomatic; phylogenetic diversity measures 


 

Plotting resources

plotting symbols

Numbers for plotting symbols; set pch=#

line types

numbers for line type (lty=), line thickness (lwd=) and arrows (code=) for plotting

color numbers

Basic colors called by number for plotting in R (col=)

R colors by names

Select colors in R by defined names

color picker

Use the color guide at http://html-color-codes.info to get 6-character color codes, or RGB codes.  Then plot(y~x, col="#8E2BC3") or mycol1<-rbg(142/255,43/255,195/255,.7); plot(y~x, pch=19,col=mycol1)

These are the data used for all the scatterplot examples.  Copy and paste these into the R console, and hit return.  Then copy and paste the code from any of the graphs to reproduce the graphs.

x<-c(1,2,3,4,5,6,7,8)
y1<-c(2,4,5,7,8,7,9,10)
y2<-c(1,3,2,4,6,5,7,7)

Simple scatterplotplot(x,y1,xlab="arrival order",ylab="hat size (cm)", ylim=c(0,10),xlim=c(0,8))

Scatterplot with overlay pointsplot(x,y1,xlab="arrival order",ylab="hat size (cm)",ylim=c(0,10),xlim=c(0,8),pch=1,col="black")
points(x,y2,pch=19,col="blue")
legend(0,10,c("male","female"),pch=c(1,19),col=c("black","blue"))

plot(x,y1,xlab="arrival order",ylab="hat size (cm)",ylim=c(0,10),xlim=c(0,8),pch=1,col="black",type="b") lines(x,y2,pch=19,col="blue",type="o")    #type"p"=points, "l"=lines, "b"= both,"c" lines alone of "b", #"o" =overplotted,"h"=histogram-like,"s"=stair steps

plot(x,y1,xlab="arrival order",ylab="hat size (cm)",ylim=c(0,10),xlim=c(0,8),pch=1,col="black",type="b")
lines(x,y2,pch=19,col="blue",type="o")

#type"p"=points, "l"=lines, "b"= both,"c" lines alone of "b",
#"o" =overplotted,"h"=histogram-like,"s"=stair steps

Simple linesplot(x,y1,xlab="arrival order",ylab="hat size (cm)",ylim=c(0,10),xlim=c(0,8),pch=1,col="black",type="l",lwd=2,lty=2)

lines(x,y2,pch=19,col="blue",type="l",lwd=1,lty=1)

legend("topleft",c("male","female"),lty=c(2,1),col=c("black","blue"),lwd=c(2,1))

 

Lowess Curves#add smooth lowess curves to  each set of points in the scatterplot
plot(x,y1,xlab="arrival order",ylab="hat size (cm)",ylim=c(0,10),xlim=c(0,8),col="dark green",pch=1,lwd=2)
lines(lowess(x,y1),lwd=2,lty=3,col="dark green")
points(x,y2,pch=19,col="dark blue")
lines(lowess(x,y2),lwd=2,lty=2,col="dark blue")
legend("topleft",c("male","female"),lty=c(3,2),pch=c(1,19),col=c("dark green

 

Fitted lines#Use abline to add linear regression lines to each set of points in the scatterplot

plot(x,y1,xlab="arrival order",ylab="hat size (cm)",ylim=c(0,10),xlim=c(0,8),col="black",pch=1,lwd=1)
abline(lm(y1~x),lwd=1,lty=1,col="black")
points(x,y2,pch=19,col="blue")
abline(lm(y2~x),lwd=1,lty=2,col="blue")
legend("topleft",c("male","female"),lty=c(1,2),pch=c(1,19),col=c("black","blue"),lwd=c(1,1))

 

fitted line with text#get the relevant statistics for the regression line, then put on the graph as text
a<-summary(lm(y2~x))  #this puts summary stats of the linear regression of y2 on x into list a
R2<-signif(a$adj.r.squared,3)  #adjusted R squared
F<-signif(a$fstatistic[1],3) #F statistic
ndf<-signif(a$fstatistic[2],1)  #degrees of freedom numerator
ddf<- signif(a$fstatistic[3],1)  #degress of freedom denominator
P<-signif(a$coefficients[2,4],4)   #P value for significant slope

plot(x,y2,xlab="arrival order",ylab="hat size (cm)",ylim=c(0,10),xlim=c(0,8),col="blue",pch=19,lwd=1)
abline(lm(y2~x),lwd=1,lty=1,col="blue")  #puts in the regression line
text(0,9,paste("F=",F,", df=",ndf,",",ddf,"n","R^2=",R2,", P=",P,sep=""),pos=4)  #adds the statistics