Working with data types of R

I have discussed about package management in the previous post. This post concentrates on data type, data structure and data coercion.

Data types & Data Structure

We have different Object types in R as given below.

• Vectors
• Lists
• Matrices
• Arrays
• Factors
• Data Frames

We have the following data types in R.

• Logical
  [code language=”r”]
  myvar <- TRUE
  [/code]
• Numeric
  [code language=”r”]
  myvar <- 23.5
  [/code]
• Integer
  [code language=”r”]
  myvar <- 2L
  [/code]
• Complex
  [code language=”r”]
  myvar <- 2+5i
  [/code]
• Character
  [code language=”r”]
  myvar <- "GOOD"
  [/code]
• Raw
  [code language=”r”]
  myvar <- charToRaw("GOOD")
  [/code]

We could see them with examples below.

I have already written about c command in Basic functions in R language

[code language=”r”]
> id1 <- c(1,2,3,4)
> id1+10
[1] 11 12 13 14
[/code]

rep command helps us to repeat the values.

[code language=”r”]
> gender <- rep(c("male", "female"), c(6,4))
> gender
[1] "male" "male" "male" "male" "male" "male" "female" "female"
[9] "female" "female"
[/code]

male is repeated 6 times where as female repeated 4 times.

Serial number is given by the command sn.

[code language=”r”]
> sn <- seq(1, 10, 1)
> sn
[1] 1 2 3 4 5 6 7 8 9 10 > sn <-seq(1, 10, 2)
> sn
[1] 1 3 5 7 9
[/code]

we passed start value, end value and interval.

cbind meant for column bind.

[code language=”r”]
> #vector to matrix – cbind
> slno_gender <- cbind(sn, gender)
> slno_gender
sn gender
[1,] "1" "male"
[2,] "2" "male"
[3,] "3" "male"
[4,] "4" "male"
[5,] "5" "male"
[6,] "6" "male"
[7,] "7" "female"
[8,] "8" "female"
[9,] "9" "female"
[10,] "10" "female"
[/code]

This gives column wise data, in matrix. Matrix takes single data type only. Hence serial number is changed to character data type.

Let’s look at similar example using DataFrame. We shall overcome this data type problem.

[code language=”r”]
> #data frame
> df <- data.frame(sn, gender)
> df
sn gender
1 1 male
2 2 male
3 3 male
4 4 male
5 5 male
6 6 male
7 7 female
8 8 female
9 9 female
10 10 female
[/code]

Let’s create another object called names

[code language=”r”]
> name <- c("Sophia","Jackson","Isabella","Lucas","Charlotte","Oliver","Amelia","Benjamin","Sarah","Julian")
> name
[1] "Sophia" "Jackson" "Isabella" "Lucas" "Charlotte" "Oliver"
[7] "Amelia" "Benjamin" "Sarah" "Julian"
[/code]

Let’s create another object called salary.  when we specify number of samples (10), mean (1000) and standard deviation (20), this command would generate random numbers to satisfy this.

[code language=”r”]
> sal <- rnorm(10, 1000, 200);
> sal
[1] 898.5731 1105.1638 757.7067 934.9025 1006.0053 1014.4837 1188.6763 611.0265
[9] 643.5498 1121.6005
> mean(sal)
[1] 928.1688
> sd(sal)
[1] 200.4666
[/code]

Let’s combine 3 objects together now to form a new object, using data.frame

[code language=”r”]
> mydataset <- data.frame(sn, gender, sal)
> mydataset
sn gender sal
1 1 male 898.5731
2 2 male 1105.1638
3 3 male 757.7067
4 4 male 934.9025
5 5 male 1006.0053
6 6 male 1014.4837
7 7 female 1188.6763
8 8 female 611.0265
9 9 female 643.5498
10 10 female 1121.6005
[/code]

After showing the above example for data frame, lets switch to list.

[code language=”r”]
> mydatalist <- list(v1=id1, v2=sn, v3=slno_gender, v4=mydataset)
> mydatalist
$v1
[1] 1 2 3 4

$v2
[1] 1 2 3 4 5 6 7 8 9 10

$v3
sn gender
[1,] "1" "male"
[2,] "2" "male"
[3,] "3" "male"
[4,] "4" "male"
[5,] "5" "male"
[6,] "6" "male"
[7,] "7" "female"
[8,] "8" "female"
[9,] "9" "female"
[10,] "10" "female"

$v4
sn gender sal
1 1 male 898.5731
2 2 male 1105.1638
3 3 male 757.7067
4 4 male 934.9025
5 5 male 1006.0053
6 6 male 1014.4837
7 7 female 1188.6763
8 8 female 611.0265
9 9 female 643.5498
10 10 female 1121.6005
[/code]

I put different object types together in a single collection and stored it in one object. We shall refer to individual data as variables as shown below.

[code language=”r”]
> mydatalist$v1
[1] 1 2 3 4
[/code]

Cool isn’t it!

Data Coercion

Let’s look at data coercion now. mode command helps is to find out the data type of an object

[code language=”r”]
> mode(mydatalist$v1)
[1] "numeric"
> mode(mydatalist$v3)
[1] "character"
> mode(gender)
[1] "character"
[/code]

Let’s convert this character type to factor type now.

[code language=”r”]

gender1<-as.factor(gender)

[/code]

How is this possible to represent the text  as numeric? Let’s look at the class, which shows us the data structure.

[code language=”r”]
> class(gender1)
[1] "factor
[/code]

Okay, let’s unclass gender1 now to know how is it stored.

[code language=”r”]
> unclass(gender1)
[1] 2 2 2 2 2 2 1 1 1 1
attr(,"levels")
[1] "female" "male"
[/code]

So 1 represents female, 2 represents male.

See you in another interest post.