R Studio Lab Summary Sheet

R Studio Lab Summary Sheet
Getting R Studio Started Collect variables together into a dataset using

Select from the Start menu: dataset<-data.frame(variable1,variable2,…)
 RStudio > RStudio Working with Datasets

Dataset names are case sensitive and cannot have
spaces.
Creating Variables
A variable name can be a letter, word, or names(dataset) gives the variable
letter/numeric/punctuation combination. No names
spaces, case sensitive. Values are assigned using head(dataset) gives the first 5
“<-“or “=”. observations
tail(dataset) gives the last 5
eg, a=5 b=-3 c=15 are numeric variables observations
animal=’dog’ is a character variable
Calling a variable within a dataset:
z=c(1,2,3,4) is a combination of numeric values dataset$variable
belonging to the variable z.
Saving a dataset as an .RData dataset:
animals=c(‘dog’,cat’,’bird’) is a combination of save(dataset,file=’filename.RData’)
character values belonging to the variable animals.
Categorical Data
Calculating using R Studio table(variable) one-way table
+ addition table(variable1,variable2) two-way table
- subtraction percent=round(100*table(variable/sum(variable)))
* multiplication
/ division pie(table(variable)) for a pie
^ powers chart
() to ensure the BEDMAS rules are applied barplot(table(variable)) for a bar
correctly chart
eg. a=5 b=-3 c=15

(a*b)/c will give -1 Numerical Data
min(variable)
max(variable)
Datasets mean(variable)
A dataset can have numeric or character variables. median(variable)
IQR(variable)
Existing datasets within R Studio: sd(variable)
library(help=’datasets’)
dataset summary(variable) will give you the minimum, Q1,
help(dataset) median, mean, Q3, maximum for a variable.
Reading textfiles: by(numeric,category,function)

Within R Studio, choose function can be any of min, max, mean, median,
Environment >Import Dataset>Text file… IQR, sd or summary.
Reading RData files: dotplot(variable)

In the Files/Plots window go to the folder stripchart(variable)
containing the dataset. Double click on the stem(variable)
dataset. A dialogue box will ask to load the .RData hist(variable)
file. Click Yes. boxplot(variable)
Creating your own data:

library(aplpack) to obtain back to back stem pnorm(lower,mean=,sd=,lower.tail=FALSE) for
plots P(X>a)
stem.leaf.backback(variable1,variable2)
pnorm(c(lower,upper),mean=,sd=,lower.tail=TRUE)
to find P(a<X<b)
Copying a graph from R Studio to Word qnorm(c(probability), mean=,sd=,lower.tail=TRUE)

In the Files/Plots window use the Export pull down for P(X<?)=probability
menu, and Copy to Clipboard. A dialogue box will
open, click Copy Plot, and paste in your Word file. qnorm(c(probability), mean=,sd=,lower.tail=FALSE)
for P(X>?)=probability
Commands for both Categorical and Numerical qnorm(c(probability1,probability2),

Data mean=,sd=,lower.tail=TRUE)
plot(dataset) gives scatterplots for for P(?<X<?)=central probability
each pair of variables (both categorical and
numeric)
plot(category) gives a bar chart
plot(numeric) gives an index plot Sampling
plot(numeric~category) gives boxplots by sample(1:population size, sample size)
category
boxplot(numeric~category) gives boxplots by
category Experimental Design
library(lattice)
summary(category) will give you the number in dotplot(numeric~category,data=dataset)
each category.
stripchart(numeric~category,data=dataset)
summary(numeric) will give you the minimum, Q1,
median, mean, Q3, maximum for a variable. boxplot(numeric~category,data=dataset)
summary(dataset) will give you a numerical or library(gplots)

categorical summary for variables within a dataset. plotmeans(numeric~category,data=dataset)
Creating titles, axes labels Regression

The following labelling options can be used in any plot(response~explanatory,data=dataset)
graphing command: cor(dataset)
model=lm(response~explanatory,data=dataset)
windows() to create a bigger graph
window coef(model) gives the coefficient for
main=’main title ’ the regression equation
xlab=’x axis label ’ summary(model)
ylab=’y axis label‘ abline(model)
res=residuals(model) gives the residuals for the

model
Normal Distribution plot(res~explanatory,data=dataset)
qqnorm(variable) gives a normal quantile plot abline(0,0)
qqline(variable) puts a line through the quantile

plot to see if the values fit a normal distribution.
Customising your graphs
pnorm(upper,mean=,sd=,lower.tail=TRUE) for col=’red’ yellow, green, blue, etc
P(X<a) cex=3 point size, 1 (default), 2, 3, etc
pch=16 plotting character – google Quick-R
A good graph should have…
 A meaningful title
 Labels for axes with measurements or times
 Be easy to read
 If data is only 2-dimensional, then we shouldn’t
represent it as 3-D
 To compare graphs they need the same scale
 Do the numbers “add-up”?
 Does the picture “match” the data?
 Do measurements or times given on axes have
the same interval width, or the same order?
 Do variables need definitions or a legend (key)?

R Studio Lab Summary Sheet

Uploaded by

Copyright:

Available Formats

R Studio Lab Summary Sheet

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

R Studio Lab Summary Sheet

Uploaded by

Copyright:

Available Formats

R Studio Lab Summary Sheet

Getting R Studio Started Collect variables together into a dataset using

 RStudio > RStudio Working with Datasets

eg. a=5 b=-3 c=15

Reading textfiles: by(numeric,category,function)

Reading RData files: dotplot(variable)

Creating your own data:

Copying a graph from R Studio to Word qnorm(c(probability), mean=,sd=,lower.tail=TRUE)

Commands for both Categorical and Numerical qnorm(c(probability1,probability2),

summary(dataset) will give you a numerical or library(gplots)

Creating titles, axes labels Regression

res=residuals(model) gives the residuals for the

qqline(variable) puts a line through the quantile

You might also like