Week 10 - Introduction to R

Which programs you should download and install

• R programming language (required) (the link for CRAN project).
• RStudio (Not required but strongly recommended) (the website of RStudio)

What the R part of the course will cover

In the following two weeks you will learn:

• Introduction to R.
• Data types in R.
• How to import data in R.
• Checking summary stats and making exploratory data analysis with R.
• Using dplyr to manipulate data with R.
• Using ggplot for data visualization with R.

What is R?

• R is a suite of software facilities for:
  • Reading and manipulating data
  • Computation
  • Conducting statistical data analysis
  • Application and development of Machine Learning Algorithms
  • Displaying the results
• R is the open-source version (i.e. freely available version - no license fee) of the S programming language, a language for manipulating objects.
• Software and packages can be downloaded from the link for CRAN project

RStudio

• The R Console by itself is not very interesting and useful. A tool named RStudio is designed to use R more efficient and easily.
• RStudio can be installed after installing R.
• RStudio won't work without R. R has to be installed on your computer.
• You can think RStudio as an upgrade of R, on visual and functionality terms. It doesn't add anything that R cannot do.
• RStudio can be downloaded from the website of RStudio

• RStudio have 4 main parts:

Using R and RStudio from Posit Cloud

• If you are a beginner at R and will only use it for the lecture then you may not want to install it into your computer.

• A great alternative for this is to use Posit Cloud.

• Posit is the company that is the founder of RStudio and provides a cloud solution for using R and Rstudio.

• First, go to the website of Posit

• Click Sign Up at the upper right part of the page

• Then, select the free plan and click Learn more.
• The Free Plan gives you 25 hours of computing time per month which should be sufficient for this course.
• If you ever think that you are going to exceed the limit you can always download and install R and RStudio to your computer for free.

• Click Sign Up at the next page.

• Enter your e-mail, a password, your name and your surname to register to posit cloud.

• After registering and possibly confirming your mail address, you can login with your e-mail and password.
• You will encounter the following page when you successfully login to the Posit Cloud.
• Select New Project and New RStudio Project from the upper right buttons.

• After a short period of deployment time, a fresh Rstudio will open as in the following picture.

• Give a new name to your project and open a R script to work on from the menu File $\rightarrow$ New File $\rightarrow$ R Script.

• Now you can start to use RStudio in cloud and make your computations.

Advantages of R

• Open Source:

  • There is no clear difference between user and developer.
  • A unique solution for the given problem can be constructed.
  • You are not limited to pre-defined options by a fixed user interface as is common in proprietary software.
  • Open source also allows to use the program freely without spending any money.

• Flexibility:

  • Gives access to the source code, allows to modify and improve it according to the needs.
  • Ability to further developments and capacity increase with tools like RStudio and Shiny.
  • New packages to solve a certain problem is consistently added to the R repository.
  • Ability to produce reports in PDF and HTML format.

• Community:

  • R has a lot of material in online platforms, in books and in courses.
  • A lot of information can be found via Q&A websites, social media networks, and numerous blogs.

Some useful websites to get help

Typically, a problem you may be encountering is not new and others have faced, solved, and documented the same issue online.

• The following resources can be used to search for online help. Although, I typically just google the problem and find answers relatively quickly.
• RSiteSearch("key phrase"): searches for the key phrase in help manuals and archived mailing lists on the R Project website at http://search.r-project.org/.
• Stack Overflow: a searchable Q&A site oriented toward programming issues. 75 % of my answers typically come from Stack Overflow questions tagged for R at http://stackoverflow.com/questions/tagged/r.
• Cross Validated: a searchable Q&A site oriented toward statistical analysis. Many questions regarding specific statistical functions in R are tagged for R at http://stats.stackexchange.com/questions/tagged/r.
• R--seek: a Google custom search that is focused on R-specific websites. Located at http://rseek.org/
• R -bloggers: a central hub of content collected from over 500 bloggers who provide news and tutorials about R. Located at http://www.r-bloggers.com/
• ChatGPT obviously.

Basic Calculations and Defining Objects in R

• You can either write a code directly into the console or you can use a script.
• Using a script is more efficient. Because it is easier to write modify and save a R Code in a script.
• Open a script with File $\rightarrow$ New File $\rightarrow$ R Script or you can use shortcut Ctrl + Shift + N

Objects

• R works by creating objects and using various functions calls that create and use these objects. For example;
  • Vectors of numbers, logical values (TRUE and FALSE), character strings and even complex numbers.
  • Matrices and general n-way arrays
  • Lists - arbitrary collections of objects of any type; e.g. list of vectors, list of matrices, etc.
  • Data frames - a general data set type
  • functions (yes even functions are objects)

Defining Variables in R

• R is case sensitive !!!

x = 5
y <- 8
z <- x+y
z

13

Basic Math in R

43 + 35 # addition

78

43 - 35 # subtraction

8

12 * 8 # multiplication

96

100 / 8 # division

12.5

2^4 # power

16

100 %% 8 # remainder

4

100 %/% 8 # divident

12

Logical Comparisons in R

5 < 8

TRUE

2 + 2 == 5

FALSE

T == TRUE

TRUE

3 * 3 == 9

TRUE

3 * 3 != 8

TRUE

3 * 3 != 9

FALSE

Functions

• Functions are special commands that are designed for a particular purpose.

• For example sum() gives the sum of a numerical values, sqrt() takes root of a number etc..

• Functions are always followed by a (). Inside the () most of the functions take some special values called arguments.

• Lets look at the help page for the sqrt() function.

?sqrt

{r,
knitr::include_graphics("./figures/help_sqrt.png")

• sqrt() function only takes one argument x which is either a single number, or arrays of numbers.

sqrt(8)

2.82842712474619

sqrt(c(1,4,9,16,25))

1. 1
2. 2
3. 3
4. 4
5. 5

• Lets look at the help page for the sum() function.

?sum

{r,
knitr::include_graphics("./figures/help_sum.png")

• According to the help file the usage of function is sum(..., na.rm = FALSE)
• sum() function takes two arguments.
  • ... numeric or complex or logical vectors.
  • na.rm logical. Should missing values (including NaN) be removed?
• The second argument na.rm has a default value of FALSE. A default value means that if you don't specify a value, it will take the default value, here FALSE.

x<- c(6, 8, 10, 12, 14)

sum(x)

50

sum(x, na.rm = FALSE)

50

sum(x, na.rm = TRUE)

50

y<- c(6, 8, 10, 12, NA)

sum(y)

<NA>

sum(y, na.rm = FALSE)

<NA>

sum(y, na.rm = TRUE)

36

z <- c(T, T, F, F, F, T, T)
sum(z)

4

Packages

• Some functions are not contained locally in R. They are called packages and they should be installed when needed.
• R contains one or more libraries of packages. A package contain various functions and data sets for numerous purposes, e.g. e1071 package, spatstat package and DandEFA package, etc.
• Some packages are a part of the R. Others should be downloaded from the Comphrensive R Archive Network.
• To access all of the functions and data sets of a particular package; for example, DandEFA; it must be loaded to the workspace:

# install.packages('DandEFA')

library(DandEFA) # Buy you have to call and load a package every new R session.

• You can also use Rstudio to download multiple packages easily and at once.
• Tools -> Install Packages

An Example for Packages: DandEFA Package

• Using packages to utilize various methods and algorithms.
• DandEFA package contains functions for a particular analysis called factor analysis.
• Factor Analysis is a method for categorize variables into groups to find the relationship between the variables in the same group.
• The package contains functions:
  • factload: A method for producing the factor loadings
  • dandelion: A method for visualizing the factor loadings

help(package="DandEFA")

• Alternatively you can also use the bottom-right panel in RStudio to get info on a specific function:

#packageDescription("DandEFA")

• You don't have to understand the following code, but understand that the following code is taken from the documentation from the DandEFA package and can be applied directly.

library(DandEFA) # loading the package

data(timss2011) # loading the dataset

timss2011 <- na.omit(timss2011) # removing the rows with missing values

dandpal <- rev(rainbow(100, start = 0, end = 0.2)) # Choose colors for visualisation

facl <- factload(timss2011,nfac=5,method="prax",cormeth="spearman") # Find the factor loadings
facl # Show the factor loadings

dandelion(facl,bound=0,mcex=c(1,1.2),palet=dandpal) # Visualise

Warning message in sqrt(val):
"NaNs produced"Warning message in sqrt(val):
"NaNs produced"Warning message in sqrt(val):
"NaNs produced"Warning message in sqrt(val):
"NaNs produced"Warning message in sqrt(val):
"NaNs produced"Warning message in sqrt(val):
"NaNs produced"Warning message in sqrt(val):
"NaNs produced"Warning message in sqrt(val):
"NaNs produced"

Loadings:
     [,1]   [,2]   [,3]   [,4]   [,5]  
X10A                0.103 -0.101 -0.224
X10B                                   
X10C         0.106               -0.129
X11A                      -0.544 -0.130
X11B                      -0.514       
X11C -0.129 -0.105        -0.500       
X11D                      -0.475       
X12A -0.116        -0.152 -0.338  0.318
X12B        -0.254 -0.133 -0.328  0.256
X12C        -0.149 -0.136 -0.298  0.249
X13A         0.549                     
X13B         0.504                     
X13C         0.583                     
X13D         0.398                     
X13E         0.595                     
X13F         0.458                     
X17A -0.539        -0.419 -0.140       
X17B  0.633         0.156        -0.164
X17C -0.350        -0.450 -0.185       
X17D  0.727         0.222        -0.173
X17E -0.325        -0.337 -0.164       
X17F -0.611        -0.445 -0.143       
X17G -0.252        -0.481 -0.145  0.157
X18A -0.303        -0.420 -0.267 -0.138
X18B  0.537  0.146         0.152 -0.152
X18C -0.353        -0.326 -0.192       
X18D -0.416        -0.413 -0.277       
X18E -0.160        -0.381 -0.239 -0.125
X19A -0.540        -0.443 -0.135 -0.254
X19B  0.633  0.119                     
X19C  0.694         0.158              
X19D -0.519        -0.424 -0.163 -0.256
X19E  0.687  0.112                     
X19F -0.415        -0.462 -0.124 -0.361
X19G -0.313        -0.491 -0.220 -0.359
X19H -0.383        -0.500 -0.214 -0.361
X19I  0.690                            
X19J -0.238        -0.507 -0.158       
X19K               -0.620         0.142
X19L               -0.714         0.124
X19M               -0.749         0.101
X19N -0.184        -0.654              
X21A               -0.120 -0.106       

                [,1]  [,2]  [,3]  [,4]  [,5]
SS loadings    5.576 1.851 4.965 1.987 1.076
Proportion Var 0.130 0.043 0.115 0.046 0.025
Cumulative Var 0.130 0.173 0.288 0.334 0.359

• In summary, packages provide a flexible environment.
• Employing multiple methods and algorithms in the same time
• Programming and using packages are two core elements in R.

Working Directory

• In order to work in R, you should specify a active working directory. In brief this is the location where R will get and save the files.
• You can call the active working directory with the command getwd()
• Note: If you don't understand the concept of working directory, you will probably get errors during importing dataset and locating files. So be careful.

# returns path for the current working directory
getwd()

'C:/Users/erhan/Documents/FEF1002'

• You can change your active working directory either by using setwd() function or by using Session $\rightarrow$ Set Working Directory $\rightarrow$ To Source File Location after saving a script.

# set the working directory to a specified directory
setwd("C:/Users/erhan/Desktop")
getwd()

'C:/Users/erhan/Desktop'

setwd("C:/Users/erhan/Documents/FEF1002")
getwd()

'C:/Users/erhan/Documents/FEF1002'

Data Types in R Programming {#data-types}

Vectors

• Vectors are one dimensional arrays that keeps only one type of variables.
• All the elements in a vector should be the same type. (Numeric, string, logical etc.)

x <- c(10.4, 5.6, 3.1, 6.4, 21.7) # Numeric Vector
x

1. 10.4
2. 5.6
3. 3.1
4. 6.4
5. 21.7

x <- c("boy","girl","boy","girl","boy","boy") # character vector
x

1. 'boy'
2. 'girl'
3. 'boy'
4. 'girl'
5. 'boy'
6. 'boy'

x <- c(TRUE,TRUE,FALSE,TRUE,TRUE,FALSE) # logical vector
x

1. TRUE
2. TRUE
3. FALSE
4. TRUE
5. TRUE
6. FALSE

# or you can use
x <- c(T,T,F,T,T,F) # logical vector
x

1. TRUE
2. TRUE
3. FALSE
4. TRUE
5. TRUE
6. FALSE

• What if I put different kind of values in a vector.

c(10, 20, 26, T) # numeric and logical values

1. 10
2. 20
3. 26
4. 1

c(10, 20, 26, "apple") # numeric and string

1. '10'
2. '20'
3. '26'
4. 'apple'

c(T, F, "apple", "banana") # logical and string

1. 'TRUE'
2. 'FALSE'
3. 'apple'
4. 'banana'

c(T, "apple", 10) # logical, string, numeric

1. 'TRUE'
2. 'apple'
3. '10'

• Accessing elements in a vector is easy,

x <- c(10.4, 5.6, 3.1, 6.4, 21.7) # Numeric Vector

• Indexing in R starts with 1 opposing to some programming languages like Python which starts indexing with 0.

• Select fifth element of the vector.

x[5]

21.7

• Select first, third and fifth element of the vector.

ind <- c(1,3,5)
x[ind]

1. 10.4
2. 3.1
3. 21.7

• Select second and fourth element of the vector.

ind <- c(F,T,F,T,F)
x[ind]

1. 5.6
2. 6.4

• A logical operation over a vector would create a logical vector (important!!)

x <- c(10.4, 5.6, 3.1, 6.4, 21.7) # Numeric Vector

• Find whether an element is higher than 7.

ind <- (x > 7)
ind

1. TRUE
2. FALSE
3. FALSE
4. FALSE
5. TRUE

• Find elements that is higher than 7.

x[ind]

1. 10.4
2. 21.7

• Find elements that is equal or lower than 7.

x[!ind]

1. 5.6
2. 3.1
3. 6.4

• We will use indices to manipulate data sets later. But a shorter version of the code is

x <- c(10.4, 5.6, 3.1, 6.4, 21.7) # Numeric Vector

x[x > 7]

1. 10.4
2. 21.7

• A logical operator checks whether the both sides have equal length or one side has length 1.

x <- c(10.4, 5.6, 3.1, 6.4, 21.7)
y <- c(4, 7, 8, 2, 35)

ind <- (x > y)
ind

1. TRUE
2. FALSE
3. FALSE
4. TRUE
5. FALSE

• If the number of elements are not equal:

x <- c(10.4, 5.6, 3.1, 6.4, 21.7)
y <- c(4,7,8,2)

• will give an output but with a warning.

ind <- (x > y)

Warning message in x > y:
"longer object length is not a multiple of shorter object length"

ind

1. TRUE
2. FALSE
3. FALSE
4. TRUE
5. TRUE

Modifying Vectors

• Any element of the vector can be modified easily:

x <- c(10.4, 5.6, 3.1, 6.4, 21.7)

x[4] <- 7.3
x

1. 10.4
2. 5.6
3. 3.1
4. 7.3
5. 21.7

• A group of elements can be modified too

x <- c(10.4, 5.6, 3.1, 6.4, 21.7)

x[x > 7] <- 100
x

1. 100
2. 5.6
3. 3.1
4. 6.4
5. 100

• Some advance stuff: (data imputation)

x <- c(10.4, NA, 3.1, 6.4, NA)

is.na(x)

1. FALSE
2. TRUE
3. FALSE
4. FALSE
5. TRUE

x[is.na(x)] <- mean(x, na.rm = TRUE)
x

1. 10.4
2. 6.63333333333333
3. 3.1
4. 6.4
5. 6.63333333333333

Manipulating vectors

• Merging vectors with c():

x <- c(10.4, 5.6, 3.1, 6.4, 21.7)
y <- c(4, 7, 8, 2, 35)
z <- c(x,y)

z

1. 10.4
2. 5.6
3. 3.1
4. 6.4
5. 21.7
6. 4
7. 7
8. 8
9. 2
10. 35

• Summation or multiplication over vectors. Note: Again both vectors either have be of same size or one has to be of length one.

x
y

1. 10.4
2. 5.6
3. 3.1
4. 6.4
5. 21.7

1. 4
2. 7
3. 8
4. 2
5. 35

z <- x + y
z

1. 14.4
2. 12.6
3. 11.1
4. 8.4
5. 56.7

z <- x * y
z

1. 41.6
2. 39.2
3. 24.8
4. 12.8
5. 759.5

Generating Sequences

• the colon :,

x <- 1:10
x

1. 1
2. 2
3. 3
4. 4
5. 5
6. 6
7. 7
8. 8
9. 9
10. 10

x <- 2*(1:10)
x

1. 2
2. 4
3. 6
4. 8
5. 10
6. 12
7. 14
8. 16
9. 18
10. 20

• the seq() function.

x <- seq(1,10)
x

1. 1
2. 2
3. 3
4. 4
5. 5
6. 6
7. 7
8. 8
9. 9
10. 10

x <- seq(1,10,by=0.5)
x

1. 1
2. 1.5
3. 2
4. 2.5
5. 3
6. 3.5
7. 4
8. 4.5
9. 5
10. 5.5
11. 6
12. 6.5
13. 7
14. 7.5
15. 8
16. 8.5
17. 9
18. 9.5
19. 10

• the rep() function.

x <- rep(3, 10)
x

1. 3
2. 3
3. 3
4. 3
5. 3
6. 3
7. 3
8. 3
9. 3
10. 3

y <- rep(c(F,T,F,T,T,T),3)
y

1. FALSE
2. TRUE
3. FALSE
4. TRUE
5. TRUE
6. TRUE
7. FALSE
8. TRUE
9. FALSE
10. TRUE
11. TRUE
12. TRUE
13. FALSE
14. TRUE
15. FALSE
16. TRUE
17. TRUE
18. TRUE

z1 <- rep(c(4,7,8,2,35),each=3)
z1

1. 4
2. 4
3. 4
4. 7
5. 7
6. 7
7. 8
8. 8
9. 8
10. 2
11. 2
12. 2
13. 35
14. 35
15. 35

z2 <- rep(c(4,7,8,2,35), times = 3)
z2

1. 4
2. 7
3. 8
4. 2
5. 35
6. 4
7. 7
8. 8
9. 2
10. 35
11. 4
12. 7
13. 8
14. 2
15. 35

An Example on Vectors

x <- c(2,4,6,8,10)
y <- c("apple", "banana", "peach", "walnut", "apple")

sum(x)

30

sum(x < 6)

2

mean(x < 6)

0.4

x[x < 6]

1. 2
2. 4

x
y

1. 2
2. 4
3. 6
4. 8
5. 10

1. 'apple'
2. 'banana'
3. 'peach'
4. 'walnut'
5. 'apple'

mean(y=="apple")

0.4

mean(x > 6 & y=="apple")

0.2

Factors

A factor is a special type of vector used to represent categorical data, e.g. gender, social class, etc.

• Stored internally as a numeric vector with values $1, 2, ..., k$, where $k$ is the number of levels.
• Can have either ordered and unordered factors.
• A factor with $k$ levels is stored internally consisting of 2 items.
  • a vector of $k$ integers
  • a character vector containing strings describing what the $k$ levels are.

Factor Example

Five people are asked to rate the performance of a product on a scale of 1-5, with 1 representing very poor performance and 5 representing very good performance. The following data were collected.

• We have a numeric vector containing the satisfaction levels.

satisfaction <- c(1, 3, 4, 2, 2, 3, 4, 2, 1, 2, 1, 1, 4, 3)

• Want to treat this as a categorical variable and so the second line creates a factor. The levels=1:5 argument indicates that there are 5 levels of the factor. We also set the labels for each factor.

fsatisfaction <- factor(satisfaction,
                        levels=1:5,
                        labels = c("very poor", "poor", "average","good", "very good"))

fsatisfaction

1. very poor
2. average
3. good
4. poor
5. poor
6. average
7. good
8. poor
9. very poor
10. poor
11. very poor
12. very poor
13. good
14. average

Levels:

1. 'very poor'
2. 'poor'
3. 'average'
4. 'good'
5. 'very good'

Matrices

• Matrices are used for many purposes in R.
• First let's create a vector from a normal distribution that we will convert to matrix.

set.seed(100) # to ensure the numbers are same for each of you
m <- rnorm(12,0,1)
m

1. -0.502192350531457
2. 0.131531165327303
3. -0.07891708981887
4. 0.886784809417845
5. 0.116971270510841
6. 0.318630087617032
7. -0.58179068471591
8. 0.714532710891568
9. -0.825259425862769
10. -0.359862131395465
11. 0.0898861437775305
12. 0.0962744602851301

dim(m) <- c(3,4)
m

-0.50219235	0.8867848	-0.5817907	-0.35986213
0.13153117	0.1169713	0.7145327	0.08988614
-0.07891709	0.3186301	-0.8252594	0.09627446

• or you can specify the dimensions with the matrix() function

set.seed(100) # to ensure the numbers are same for each of you
m <- rnorm(12)
m

1. -0.502192350531457
2. 0.131531165327303
3. -0.07891708981887
4. 0.886784809417845
5. 0.116971270510841
6. 0.318630087617032
7. -0.58179068471591
8. 0.714532710891568
9. -0.825259425862769
10. -0.359862131395465
11. 0.0898861437775305
12. 0.0962744602851301

m  <- matrix(m, nrow = 3, ncol = 4, byrow = F)
m

-0.50219235	0.8867848	-0.5817907	-0.35986213
0.13153117	0.1169713	0.7145327	0.08988614
-0.07891709	0.3186301	-0.8252594	0.09627446

• Basic functions on matrices.
  • nrow() and ncol() calls numbers of rows and columns.
  • t() calls the transpose of the matrix.
  • rownames() and colnames() are the names of columns and rows.

set.seed(100) # to ensure the numbers are same for each of you
m <- rnorm(12)
m  <- matrix(m, nrow = 3, ncol = 4, byrow = F)
m

-0.50219235	0.8867848	-0.5817907	-0.35986213
0.13153117	0.1169713	0.7145327	0.08988614
-0.07891709	0.3186301	-0.8252594	0.09627446

nrow(m)

3

ncol(m)

4

colnames(m) <- c("A", "B", "C", "D")
m

A	B	C	D
-0.50219235	0.8867848	-0.5817907	-0.35986213
0.13153117	0.1169713	0.7145327	0.08988614
-0.07891709	0.3186301	-0.8252594	0.09627446

Merging Vectors

• rbind() and cbind() functions merges vectors or matrices into matrices.

set.seed(100)
X1 <- rnorm(12)
X2 <- 1:12

m <- cbind(X1,X2)
m

X1	X2
-0.50219235	1
0.13153117	2
-0.07891709	3
0.88678481	4
0.11697127	5
0.31863009	6
-0.58179068	7
0.71453271	8
-0.82525943	9
-0.35986213	10
0.08988614	11
0.09627446	12

• Number of columns should be equal for rbind.
• Likewise, number of rows should be equal for cbind.

• Create two matrices

set.seed(100)
data_1 <- matrix(rnorm(12),nrow=3,ncol=4,byrow=T)
data_2 <- matrix(rnorm(16),nrow=4,ncol=4,byrow=F)

• and combine them.

data_new <- rbind(data_1,data_2)
data_new

-0.50219235	0.1315312	-0.07891709	0.88678481
0.11697127	0.3186301	-0.58179068	0.71453271
-0.82525943	-0.3598621	0.08988614	0.09627446
-0.20163395	-0.3888542	-0.43808998	-0.81437912
0.73984050	0.5108563	0.76406062	-0.43845057
0.12337950	-0.9138142	0.26196129	-0.72022155
-0.02931671	2.3102968	0.77340460	0.23094453

Indexing Matrices

set.seed(100) # to ensure the numbers are same for each of you
m  <- matrix(rnorm(12), nrow = 3, ncol = 4, byrow = F)
m

-0.50219235	0.8867848	-0.5817907	-0.35986213
0.13153117	0.1169713	0.7145327	0.08988614
-0.07891709	0.3186301	-0.8252594	0.09627446

• Extract the first row.

m[1,]

1. -0.502192350531457
2. 0.886784809417845
3. -0.58179068471591
4. -0.359862131395465

• Extract the second column.

m[,2]

1. 0.886784809417845
2. 0.116971270510841
3. 0.318630087617032

• Extract all the rows except the first row.

m[-1,]

0.13153117	0.1169713	0.7145327	0.08988614
-0.07891709	0.3186301	-0.8252594	0.09627446

• Extract all the columns except the first and the third one.

m[, -c(1,3)]

0.8867848	-0.35986213
0.1169713	0.08988614
0.3186301	0.09627446

index_row <- 1:3
index_col <- c(1,3,4)

• Extract the first, second and third row and first, third and fourth column.

m[index_row,index_col]

-0.50219235	-0.5817907	-0.35986213
0.13153117	0.7145327	0.08988614
-0.07891709	-0.8252594	0.09627446

• or alternatively you can use

m[1:3, c(1,3,4)]

-0.50219235	-0.5817907	-0.35986213
0.13153117	0.7145327	0.08988614
-0.07891709	-0.8252594	0.09627446

Data Frames

• A data frame

  • can be thought of as a data matrix or data set;
  • is a list of vectors and/or factors of the same length;
  • has a unique set of row names.

• Data in the same position across columns come from the same experimental unit.

• Can create data frames from pre-existing variables.

• The main spec of data frame is the ability to keep variables with different forms.

• Both numeric, string and logical variables can be reserved in a single dataframe unlike vectors and matrices.

• Creata a vector called mean_weight.

mean_weight <- c(179.3, 179.9, 180.5, 180.1, 180.3, 180.4)
mean_weight

1. 179.3
2. 179.9
3. 180.5
4. 180.1
5. 180.3
6. 180.4

• Creata a vector called Gender.

Gender <- c("M", "M", "F", "F", "M", "M")
Gender

1. 'M'
2. 'M'
3. 'F'
4. 'F'
5. 'M'
6. 'M'

• Convert Gender to a factor variable.

Gender <- factor(Gender,levels=c("M","F"))
Gender

1. M
2. M
3. F
4. F
5. M
6. M

Levels:

1. 'M'
2. 'F'

• Combine both vectors into a dataframe.

d <- data.frame(mean_weight, Gender)
d

mean_weight	Gender
179.3	M
179.9	M
180.5	F
180.1	F
180.3	M
180.4	M

• Note that the resulting variables have different data types.
  • mean_weight is numeric.
  • Gender is factor.
• This wouldn't be the case if we try to store them in a matrix as they can only store one type of variable.

Converting other Structures to Dataframes

You can also convert other data types to dataframes

• Converting a matrix to a data frame:

d <- cbind(mean_weight,Gender)
d

mean_weight	Gender
179.3	1
179.9	1
180.5	2
180.1	2
180.3	1
180.4	1

• We created a matrix from mean_weight and Gender. Gender is automatically converted to a numerical variable as variables in the matrices should be in the same data type.

d <- as.data.frame(d)
d

mean_weight	Gender
179.3	1
179.9	1
180.5	2
180.1	2
180.3	1
180.4	1

• Even if we convert the matrix to a dataframe the categorical names of the Gender is gone.

Accesssing elements in a dataframe

• There are a lot of different way to access rows and columns in a dataframe.
• You can either use single bracket [ ], double bracket [[ ]] or $ sign.
• Investigate the following code snippets to understand how R behaves.

d$mean_weight # output in vector format

1. 179.3
2. 179.9
3. 180.5
4. 180.1
5. 180.3
6. 180.4

d[["mean_weight"]] # output in vector format

1. 179.3
2. 179.9
3. 180.5
4. 180.1
5. 180.3
6. 180.4

d[,1] # output in vector format

1. 179.3
2. 179.9
3. 180.5
4. 180.1
5. 180.3
6. 180.4

d[,"mean_weight"] # output in vector format

1. 179.3
2. 179.9
3. 180.5
4. 180.1
5. 180.3
6. 180.4

d["mean_weight"] # output in dataframe format

mean_weight
179.3
179.9
180.5
180.1
180.3
180.4

d[1] # output in dataframe format

mean_weight
179.3
179.9
180.5
180.1
180.3
180.4

• You can access a subset of rows by indexing the data frame

d[c(1,4,5),] # Shows 1., 4. and 5. rows of the dataframe

	mean_weight	Gender
1	179.3	1
4	180.1	2
5	180.3	1

• It is suggested that you use drop=FALSE when indexing (to sustain the data frame type).

d[1:3,"mean_weight"]

1. 179.3
2. 179.9
3. 180.5

d[1:3,"mean_weight",drop=FALSE]

mean_weight
179.3
179.9
180.5

Creating a new variable in a dataframe

d

mean_weight	Gender
179.3	1
179.9	1
180.5	2
180.1	2
180.3	1
180.4	1

d$color <- NA
d

mean_weight	Gender	color
179.3	1	NA
179.9	1	NA
180.5	2	NA
180.1	2	NA
180.3	1	NA
180.4	1	NA

d$weight_two_times <- d$mean_weight*2
d

mean_weight	Gender	color	weight_two_times
179.3	1	NA	358.6
179.9	1	NA	359.8
180.5	2	NA	361.0
180.1	2	NA	360.2
180.3	1	NA	360.6
180.4	1	NA	360.8

Importing Data

• The most popular functions for reading data sets
  • read.table() function is used mainly for reading data from formatted text files.
  • read.csv() function is used mainly for reading data from files with csv format ("Comma Separated Values"format)
  • read_excel() function is used to read data directly from an excel file. It requires the external package readxl.

Datasets

You can download the datasets used in this lecture from the lecturers AVESIS page.

Pima Data Set

• Indian females of Pima heritage (Native americans living in an area consisting of what is now central and southern Arizona)

• Columns (or Variables) of the Pima data set:

  • NTP: number of times pregnant
  • PGC: Plasma glucose concentration a 2 hours in an oral glucose tolerance test
  • DBP: Diastolic blood pressure (mm Hg)
  • TSFT: Triceps skin fold thickness (mm)
  • SI: 2-Hour serum insulin (mu U/ml)
  • BMI: Body mass index (weight in kg/(height in meter square))
  • Diabetes pedigree function:
  • Age: Age (years)
  • Diabetes: f0,1g value 1 is interpreted as "tested positive for diabetes"

• First, you have to check whether the working directory of R and the location of file matches.

getwd()

'C:/Users/erhan/Documents/FEF1002'

• The location of the working directory can be changed with setwd() function.
• Alternatively you can change the working directory to the where the Rcode is located with RStudio by using:

Session -> Set Working Directory -> To Source File Location

Working on the Posit Cloud

• If you are working on the Posit Cloud you don't have to change your working directory since you are on the Cloud.
• But you have to upload your datasets to the cloud so that RStudio can locate and use it.
• On the left below pane, click the button Upload in Posit Cloud then a sub menu named Upload Files will emerge.
• Click Select Files from the menu

• From the file upload menu select the data files you wish to upload. If you intend to upload multiple files create a zip file from the datasets.

• Posit Cloud will automatically extract it once it uploads the zip file to the cloud.
• For example, week1_dataset.zip contains five different data formats for the pima dataset. When it is uploaded all five data files will be extracted under the folder week1_dataset.

• Now you can use the following importing functions inside the Posit Cloud to call dataset into R.

Importing Data from Text Files

• Now let's check our working directory once more.

getwd()

'C:/Users/erhan/Documents/FEF1002'

• Now if I want to import any data by using a function with R, I have to either:

  • Provide the full location of the dataset inside the function such as: C:/Users/erhan/Documents/FEF1002/pima.dat
  • or since my working directory is the folder FEF1002, if I put my pima.dat data inside the FEF1002 folder, it would be sufficient for me to provide pima.dat as the location.

• So in practice both

pima_data <- read.table("pima.dat", header = TRUE, sep = " ")

• and

• You can use the head() function to see if everything is imported okay.

head(pima_data)

NTP	PGC	DBP	TSFT	SI	BMI	DPF	Age	Diabetes
6	148	72	35	0	33.6	0.627	50	positive
1	85	66	29	0	26.6	0.351	31	negative
8	183	64	0	0	23.3	0.672	32	positive
1	89	66	23	94	28.1	0.167	21	negative
0	137	40	35	168	43.1	2.288	33	positive
5	116	74	0	0	25.6	0.201	30	negative

• You can use the str() function to see the structure of the dataset.

str(pima_data)

pima_data <- read.table("C:/Users/erhan/Documents/FEF1002/pima.dat", 
                        header = TRUE, sep = " ")

will work and import the data.

• Remember to change C:/Users/erhan/Documents/FEF1002/pima.dat to where the pima.dat is actually located.
• You can use both approach for the following data importing processes.

head(pima_data)

NTP	PGC	DBP	TSFT	SI	BMI	DPF	Age	Diabetes
6	148	72	35	0	33.6	0.627	50	positive
1	85	66	29	0	26.6	0.351	31	negative
8	183	64	0	0	23.3	0.672	32	positive
1	89	66	23	94	28.1	0.167	21	negative
0	137	40	35	168	43.1	2.288	33	positive
5	116	74	0	0	25.6	0.201	30	negative

str(pima_data)

'data.frame':	768 obs. of  9 variables:
 $ NTP     : int  6 1 8 1 0 5 3 10 2 8 ...
 $ PGC     : int  148 85 183 89 137 116 78 115 197 125 ...
 $ DBP     : int  72 66 64 66 40 74 50 0 70 96 ...
 $ TSFT    : int  35 29 0 23 35 0 32 0 45 0 ...
 $ SI      : int  0 0 0 94 168 0 88 0 543 0 ...
 $ BMI     : num  33.6 26.6 23.3 28.1 43.1 25.6 31 35.3 30.5 0 ...
 $ DPF     : num  0.627 0.351 0.672 0.167 2.288 ...
 $ Age     : int  50 31 32 21 33 30 26 29 53 54 ...
 $ Diabetes: Factor w/ 2 levels "negative","positive": 2 1 2 1 2 1 2 1 2 2 ...

• Here, the argument header = TRUE is used to denote that the variable names are given at the first line of the data.
• The argument sep = " " is used to denote how the variables are separated from each other. For this dataset space is used to separate variables.

Importing Data From CSV Files

pima_csv <- read.csv("pima.csv", header = TRUE, sep = ",")

head(pima_csv)

NTP	PGC	DBP	TSFT	SI	BMI	DPF	Age	Diabetes
6	148	72	35	0	33.6	0.627	50	positive
1	85	66	29	0	26.6	0.351	31	negative
8	183	64	0	0	23.3	0.672	32	positive
1	89	66	23	94	28.1	0.167	21	negative
0	137	40	35	168	43.1	2.288	33	positive
5	116	74	0	0	25.6	0.201	30	negative

str(pima_csv)

'data.frame':	768 obs. of  9 variables:
 $ NTP     : int  6 1 8 1 0 5 3 10 2 8 ...
 $ PGC     : int  148 85 183 89 137 116 78 115 197 125 ...
 $ DBP     : int  72 66 64 66 40 74 50 0 70 96 ...
 $ TSFT    : int  35 29 0 23 35 0 32 0 45 0 ...
 $ SI      : int  0 0 0 94 168 0 88 0 543 0 ...
 $ BMI     : num  33.6 26.6 23.3 28.1 43.1 25.6 31 35.3 30.5 0 ...
 $ DPF     : num  0.627 0.351 0.672 0.167 2.288 ...
 $ Age     : int  50 31 32 21 33 30 26 29 53 54 ...
 $ Diabetes: Factor w/ 2 levels "negative","positive": 2 1 2 1 2 1 2 1 2 2 ...

• From the code we can understand that column names are included in the data (header = TRUE) and variables are separated with comma (sep = ",")

Importing Data From Excel Files

• Remember that you can always use an external package to complete a different task
• Suppose we want to import directly from an excel file with .xls or .xlsx format
• We will use the readxl package.

library(readxl) # Remember youj should use install.packages('readxl') if you didn't install it before

pima_xls <- read_excel("pima.xls", sheet = 'pima')

head(pima_xls)

NTP	PGC	DBP	TSFT	SI	BMI	DPF	Age	Diabetes
6	148	72	35	0	33.6	0.627	50	positive
1	85	66	29	0	26.6	0.351	31	negative
8	183	64	0	0	23.3	0.672	32	positive
1	89	66	23	94	28.1	0.167	21	negative
0	137	40	35	168	43.1	2.288	33	positive
5	116	74	0	0	25.6	0.201	30	negative

str(pima_xls)

Classes 'tbl_df', 'tbl' and 'data.frame':	768 obs. of  9 variables:
 $ NTP     : num  6 1 8 1 0 5 3 10 2 8 ...
 $ PGC     : num  148 85 183 89 137 116 78 115 197 125 ...
 $ DBP     : num  72 66 64 66 40 74 50 0 70 96 ...
 $ TSFT    : num  35 29 0 23 35 0 32 0 45 0 ...
 $ SI      : num  0 0 0 94 168 0 88 0 543 0 ...
 $ BMI     : num  33.6 26.6 23.3 28.1 43.1 25.6 31 35.3 30.5 0 ...
 $ DPF     : num  0.627 0.351 0.672 0.167 2.288 ...
 $ Age     : num  50 31 32 21 33 30 26 29 53 54 ...
 $ Diabetes: chr  "positive" "negative" "positive" "negative" ...

pima_xlsx <- read_excel("pima.xlsx", sheet = 'pima')

head(pima_xlsx)

NTP	PGC	DBP	TSFT	SI	BMI	DPF	Age	Diabetes
6	148	72	35	0	33.6	0.627	50	positive
1	85	66	29	0	26.6	0.351	31	negative
8	183	64	0	0	23.3	0.672	32	positive
1	89	66	23	94	28.1	0.167	21	negative
0	137	40	35	168	43.1	2.288	33	positive
5	116	74	0	0	25.6	0.201	30	negative

str(pima_xlsx)

Classes 'tbl_df', 'tbl' and 'data.frame':	768 obs. of  9 variables:
 $ NTP     : num  6 1 8 1 0 5 3 10 2 8 ...
 $ PGC     : num  148 85 183 89 137 116 78 115 197 125 ...
 $ DBP     : num  72 66 64 66 40 74 50 0 70 96 ...
 $ TSFT    : num  35 29 0 23 35 0 32 0 45 0 ...
 $ SI      : num  0 0 0 94 168 0 88 0 543 0 ...
 $ BMI     : num  33.6 26.6 23.3 28.1 43.1 25.6 31 35.3 30.5 0 ...
 $ DPF     : num  0.627 0.351 0.672 0.167 2.288 ...
 $ Age     : num  50 31 32 21 33 30 26 29 53 54 ...
 $ Diabetes: chr  "positive" "negative" "positive" "negative" ...

• We should define the sheet name of the data with the argument sheet inside the read_excel() function.