Moving Around in R

What we will see:

• console, and script window: 4 basic panes
• auto-suggest, code highlighting
• projects and working directories (use getwd())

# assign a value, = is possible, but  '<-'  is R-ish
a <- 1
is.vector(a)

[1] TRUE

# a vector with multiple elements, 
# !! watch out 'c' is a base R function!!
b <- c(1,2,3)
# careful! c is a bad name for a custom object !
d <- 1:10
d

 [1]  1  2  3  4  5  6  7  8  9 10

• vector
• matrix (columns have same length AND same class)
• data.frame (columns have the same length)
• list (elements can have arbitrary length and classes, lists can also be nested)

# what's in my R Session 
ls()
# look at the first few lines of dataset
data(swiss)
head(swiss) # what might tail() do ?
# what type of data am I dealing with
str(swiss) 

# get 2nd element of a vector
d[2]

[1] 2

# get 2nd line (row) of a matrix or data.frame
swiss[2,] # guess what second column would be ?

         Fertility Agriculture Examination Education Catholic
Delemont      83.1        45.1           6         9    84.84
         Infant.Mortality
Delemont             22.2

# all but first element: negative indexing
d[-1]

[1]  2  3  4  5  6  7  8  9 10

1. Assign the 2nd and 3rd column to a an R Object called sec_third.

2. Find out the class of the newly created object.

3. Concatenate both columns of sec_third into a single R object called sng.

head(swiss[,c("Education","Catholic")])

             Education Catholic
Courtelary          12     9.96
Delemont             9    84.84
Franches-Mnt         5    93.40
Moutier              7    33.77
Neuveville          15     5.16
Porrentruy           7    90.57

# conveniently extract single vectors
swiss$Fertility[1:3]

[1] 80.2 83.1 92.5

na <- c(1,2,3,4,NA,6)
# sum(na) # NA
sum(na,na.rm=T) # 16

[1] 16

# set NA
is.na(na) <- 1
na[is.na(na)] <- c(1,5)
na

[1] 1 2 3 4 5 6

• native R binaries: .RData (load())
• comma separated values: .csv (seperator often: “;”) (read.table(), read.csv())
• third party formats: .xlsx, .xls, .dta (stata), SPSS, …
• databases (RPostgreSQL,ROracle, RMySQL, …)

-> Hint: Most read/write functions can also handle data from a remote server.

# let's save our work
save(d,swiss,file="session1.RData")
# clear the entire memory
rm(list=ls())
ls()

character(0)

# load 
load("session1.RData")
ls()

[1] "d"     "swiss"

# find our working directory
getwd()

[1] "/Users/mbannert/Phd/teaching/exploring_stats_with_R/course_slides"

# read narcissistic personality inventory into R
# previously downloaded from http://personality-testing.info/_rawdata/
# 
npi <- read.csv2("../data/data-session-1.csv",sep=",")
# how would you see what you have got?

# using indexing to subset
children <- npi[npi$age < 18,]
# use subset to subset
children2 <- subset(npi,age < 18)
identical(children,children2)

[1] TRUE

1. Create a new R Studio project called 'tasks' and create an Rmarkdown file task-1.Rmd. Use this file to document what you are doing and to present your results.

2. Name 3 potentially useful packages for pyschological measurement that don't ship with a standard R installation.

3. Create a matrix called m with 4 rows and 3 columns. Create a data.frame d with 3 rows and 4 columns. Find out what the function t() can do for you.

4. Use the example dataset swiss. a) What's the average education? Explain the measurement. b) Display the last 10 rows of the entire dataset.

5. What's the minimal Agriculture? Which Province has it?