R Statistics commands

citation()

To get some information about how to cite R

+-*l^

Basic arithmetic operations in R

← or →

Variable assignment

if, else, repeat, while, function, for, in, next, break, TRUE, FALSE, NULL, Inf, NaN, NA, NA_integer_, NA_real_, NA_compleX_, NA_character_

Reserved keywords that R needs to keep “safe” so that you can’t use them as variable names

sqrt()

Square root function

abs()

Absolute value function

round(x,# of decimal places) or round(x, digits = )

Round some value

c()

Combine function

Example: sales.by.month ← c(0, 100, 200, 50, 0, 0, 0, 0, 0, 0, 0, 0)

[ ]

To get information out of a vector

Example: sales.by.month[2]

february.sales ← sales.by.month[2]

edit() or fix()

Edit variables

length(x=, )

To find out how many elements there are in a vector

Example: length(x=sales.by.month)

Output = 12

“word”

Storing text data

Example: greeting ← “hello”

Output: hello

nchar(x=)

To count the number of individual characters that make up a string

For example: nchar(x=greeting)

Output = 5

==

Equality operator to force R to make a “true or false” judgment

Example: 2+2==4

Output = 4

<, <=, >, >=, ==, !=

Operators, whereby < is less than, <= is less than or equal to, > is greater than, >= is greater than or equal to, == is equal to, and != is not equal to

!

not

|

or

&

and

TRUE (or T)

True

FALSE (or F)

False

[( )]

To extract multiple elements of a vector

For example: sales.by.month[ c(2,3,4) ]

x:x

Shorthand for c()

For example: 2:8

Output = 2 3 4 5 6 7 8

q()

To quit R

#

To tell R to ignore everything else you’ve written on this line. Handy for commenting

print()

To get R to display information

library(“your package of choice”)

To load a desired package

exists(“function within a package”)

To see if R knows about the existence of a function within a package

detach(“package:package of choice”, unload=TRUE)

To get rid of / to unload a package

install.packages()

To install packages

Example: install.packages(“psych”)

update.packages()

To update packages

lsr, psych, car, gplots, sciplot, foreign, effects, R.matlab, gdata, lmtest, reschape, compute.es, HistData, and multcomp, sem, ez, nlme, and lm34

Useful packages

objects() or ls() or ls.str() or who()

To view the objects in your data environment

Preferable to use the who() command

Example: objects(“packages:stats”)

rm( ,)

To remove variables

getwd()

To get the current working directory

setwd(“C:/Users/liam”)

To specify the working directory

list.files()

To list the contents of the directory

load(file=”the file you are looking for”)

To load a data file

Example: load(“chapek9.Rdata)

setwd(“../data”) and load(“the file you want”)

Easier way is to change the working directory first and then load the file

read.csv

To read a csv file

(data file) ← read.csv(file = “data file.csv”)

To import a CSV file

Example: books ← read.csv(file = ”booksales.csv”)

save.image(file = “desired file”)

To save all of the variables in your workspace into the data file

Example: save.image(file = ”myfile.Rdata”)

save(variable1, variable2, file = “file name”)

To save some (but not all) of your variables

Example: save(data, handy, file = “myfile.Rdata”)

The data for the variable junky is not being saved here

save(file = “file name”, list = list of variables you intend to save)

To save variables. Perhaps easier than the other save option

Inf, NaN, NA, NULL

Special values that may appear in situations where you were expecting a number

Inf

Infinity

Example: 1 / 0

Output = Inf

NaN

“Not a number”

Example: 0 / 0

Output = NaN

NA

“Not available”. Usually when there are missing values

NULL

The variable genuinely has no value whatsoever

names(variable name) ← c(“value1”, “value2”, value3”, “value4”)

To assign names to each element of a variable

Example: names(profit) ← c(“Q1”, “Q2”, “Q3”, “Q4”)

class(x)

A “high level” classification to capture psychologically or statistically meaningful distinctions

Example: x ← “hello world”

class(x)

Output = “character”

Example 2: x ← TRUE

class(x)

Output = “logical”

Example 3: x ← 100

class(x)

Output = “numeric”

mode(x)

The format of the information that a variable stores

type(x)

To see the distinction between integer data, double precision numeric, etc.)

group ← as.factor(group)

To group a variable by its factors

Example: group ← c(1, 1, 1, 2, 2, 2, 3, 3, 3)

group ← as.factor(group)

group

Output: 1 1 1 2 2 2 3 3 3

Levels: 1 2 3

data.frame(var1, var2, var3, var4)

To convert your variables into a data frame

Example: experiment1 ← data.frame(age, gender, group, score)

$

To extract / pull out the contents of a data frame you’re interested in

Example: experiment1$score

Outpurt = 12 10 11 15 16 14 25 21 29

who(expand = TRUE)

To expand any data frames that you’ve got in the workspace

~

To specify a formula

Example1: formula1 ← out ~ pred

Example2: formula2 ← out ~ pred1 + pred2

Example3: formula3 ← out ~ pred1 * pred2

Example4: formula 4 ← ~ var1 + var2

print.default(x)

(Come back to)

summary(object=variable)

To get a summary of your data

Example1: summary(object=afl.margins)

Example2: summary(chapek9)

plot()

To plot your data

?

To get help for a function

Example1: ?load

help(“load”)

Example2: ??load

help.search(“load”)

help.request

To request help from the R help mailing list

sum(variable)

To calculate the sum of a variable

Example1: sum(afl.margins)

Output = 6213

Example2: sum(afl.margins[1:5]) / 5

Output = 183

mean(x=variable)

To calculate the mean of a variable

Example1: mean(x=afl.margins)

Output = 35.30

Example2: mean(afl.margins[1:5])

Output = 36.6

sort(x=variable)

Puts the data points in numerical order

median(x=variable)

To calculate the median of a variable

Example: median(x=afl.margins)

mean(x=variable, trim=number to be entered)

To trim/”drop” the extreme values (outliers) on either side of the mean

Example1: mean(x=dataset, trim=.1)

Output = 5.5

Example2: mean(x=afl.margins, trim = .05)

Output = 33.75

table(dataset)

To create a frequency table of the dataset

Example: xtab.3d ← table(speaker, utterance, dan.awake)

xtab.3d

modeOf(variable)

To calculate the mode of a variable

Example: modeOf(x=afl.finalists)

Output = “Geelong”

maxFreq()

To calculate the modal frequency

Example: maxFreq(x=afl.finalists)

Output = 39

max(variable)

To get the maximum value of a variable

Example: max(afl.margins)

min(variable)

To get the minimum value of a variable

Example: min(afl.margins)

range(variable)

To get the range of a variable

Example: range(afl.margins)

Output = 0 116

quantile(x=variable, probs=number to be entered)

To calculate the percentiles/quantiles of a variable. The median of a set is its 50th quantile/percentile

Example1: quantile(x=afl.margins, probs=.5)

Output = 50%

30.5

Example2: quantile(x=afl.margins, probs=c(.25, .75))

Output = 25% 75%

12.75 50.50

IQR(x=variable)

To calculate the interquartile range (25%-75%) of a variable

Example: IQR(x = afl.margins)

aad(x)

To calculate the mean (average) absolute deviation of a variable

Example: aad(x)

var(x)

To calculate the variance of a variable

sd(variable)

To calculate the standard deviation of a variable

mad(x=variable, constant=1)

To calculate the median absolute deviation

Example: mad(x=afl.margins, constant=1)

skew(x=variable)

To calculate the skewness of a variable

Example: skew(x = afl.margins)

kurtosi(x=variable)

To calculate the kurtosis of a variable

Example: kurtosi(x=afl.margins)

as.character(variable)

To convert something to a character vector

Example: f2 ← as.character(afl.finalists)

summary(object=f2)

Output = Length 400 Class character Mode character

var, n, mean, sd, median, trimmed, mad, min, max, range, skew, kurtosis, se

The descriptive statistics included in the summary()

describeBy(x=variable1, group=variable1$variable2)

Very similar to the describe() function, except that it has an additional argument called group which specifies a grouping variable

Example: describeBy(x = clin.trial, group = clin.trial$therapy)

by(data=dataset, INDICES=dataset$variable, FUN=summary)

The command groups the dataset by the variable and applies the summary function to each subset of data

Example: by(data=clin.trial, INDICES=clin.trial$therapy, FUN=summary)

aggregate(formula = var1 ~ var2 + var3, data = dataset, FUN = mean)

This command calculates the mean of var1 for each combination of var2 and var3 groups in the dataset

Example1: aggregate(formula = mood.gain ~ drug + therapy, data = clin.trial, FUN = mean)

Example2: aggregate(mood.gain ~ drug, clin.trial, sd)

cor(x = dataset$var1, y = dataset$var2))

To calculate correlations

cor(x = dataset)

To calculate all correlations of a dataset

Example: cor(parenthood)

cor(dataset$var1, dataset$var2, method = “spearman”)

This command calculates the Spearman correlation coefficient between var1 and var2 in the dataset to measure their monotonic relationship

Example: cor(effort$hours, effort$grade, method = “spearman”)

correlate(dataset)

Automatically ignores variables in a data frame, making it more convenient for computing correlations only between numeric variables without needing manual data preprocessing

Example: correlate(work)

correlate(x, corr.method = “spearman”)

The same as correlate(x), but now species which correlation method to use

na.rm=TRUE

To ignore missing (NA) values when performing calculations

Example: mean(x = partial, na.rm = TRUE)

cor(datatset, use “complete.obs”)

To ensure that only rows with no missing values across all variables are used in the correlation calculation

Example: cor(parenthood2, use = “complete.obs”)

cor(dataset, use = “pairwise.complete.obs”)

To compute correlations using all available data for each pair of variables, ignoring missing values only for the specific variables being correlated rather than removing entire rows

Example: cor(parenthood2, use = “pairwise.complete.obs”)

plot.default()

To make a basic plot

main = “title of plot”

To specify title of plot

sub = “subtitle of plot”

To specify subtitle of plot

xlab = “x-axis label”

To specify x-axis label

ylab = “y-axis label”

To specify y-axis

font.main, font.sub, font.lab, font.axis

To specify font style of plot, where 1 = plain text, 2 = boldface, 3 = italic, 4 = bold italic