Data classes & data frames

1 Data classes

There are different data classes we will work with in R. Definitions of them and how to determine what data type you are working with is a key foundation of R. See Table 1.

Let’s start elaborating on what we’ve already used. Use class() and str() to explore data types.

1.1 Definitions

Use class() and str() to see how x is defined in this table.

Table 1: Definitions of data types in R.
Term	Example	Code (x)	Description
Logical	TRUE, FALSE	`x <- TRUE`	Two data values, also called boolean
Numeric	243, 92.1, or 3.459	`x <- 3.14`	Numbers, including those with decimals. Standard input for most statistical tools.
Integer	10L, 65L, or 0L	`x <- 314L`	Numbers without decimals. Needs to be specified as an integer.
Character	“seven”, “hello world”, “FALSE”, ‘63.354’, “600”	`x <- "150"`	Represents a string of values as a variable. Use `''` for character variables and `""` for string variables.
Complex	6 + 5i	`x <- 6 + 8i`	Data type with an imaginary part, in this case i.

The data classes above represent the core basic “data types”. Vectors are comprised of elements of different classes.

1.2 Exploring data types and vectors

Use the below examples to understand how R interprets different data classes and how they are elements within a vector.

What is “juice”?

juice <- 4 + 4
class(juice)
class(4)
class(8.3)
class(4L)

What about x?

x <- c(1, 7, 9)
class(x)

apples <- 5
oranges <- 6
fruit <- apples + oranges

How is the below vector different from above?

x <- c("1", "7", "9")
class(x)

Let’s write over x, and set it equal to types of fruit.

x <- c("banana", "grapefruit")
class(x)

Instead of a numeric value, a character is a string type. Let’s experiment with the numeric vs. character data type. Lets combine our character list of fruit with the earlier defined “apples”.

x
apples

new_x <- c(x, apples)

What happened to the data class of new_x when we combined character and numeric?

Create another vector

numeric_vector <- c(1, 10, 49, 5)
character_vector <- c("a", "b", "c", "d")

names(numeric_vector) <- character_vector

What does names() do above?

numeric_vector

# Isolate a numeric within the vector
q <- numeric_vector[4]

# Modify elements within a vector
q + numeric_vector

# Evaluate vector based on value
ans <- q > numeric_vector
ans

2 Lists

A list can be a list of anything. But everything in the list does NOT need to be the same data type (in contrast to the vector, above). Because lists can include different data types and data sets, they can be used to store information, like a dictionary.

list_example <- list(2^4, "cabbage", TRUE, 1+5i)

list_example_2 <- list(2E2, "lettuce", VALUES = 1:25, VALUES_BINARY = FALSE)

list_example_2$VALUES

typeof(list_example_2$VALUES)

typeof(list_example_2$VALUES_BINARY)

3 Matrices

When do you use one over the other? It depends on what kind of data you’re working with. If you’re working with different types of information in the same table, you likely need a data frame. But if you’re working with a single data type, a matrix may be better for you. Additionally, matrices are more efficient with respect to memory. Therefore, a lot of statistical tools/methods require a matrix as input.

Create a matrix.

matrix(1:9, byrow = FALSE, nrow = 3)

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

q <- c(460, 314)
r <- c(290, 247)
w <- c(309, 165)

c(q, r, w)

[1] 460 314 290 247 309 165

mydataset <- matrix(c(q, r, w), byrow = TRUE, nrow = 3)
mydataset

     [,1] [,2]
[1,]  460  314
[2,]  290  247
[3,]  309  165

region <- c("one", "two")
category <- c("A","B","C")

rownames(mydataset) <- category
colnames(mydataset) <- region

rowSums(mydataset)

  A   B   C 
774 537 474

totals <- rowSums(mydataset)

cbind(mydataset,totals)

  one two totals
A 460 314    774
B 290 247    537
C 309 165    474

4 Data frames

Create a data frame and explore the structure. Data frames are the most common way to view datasets that you eventually want to modify and/or make plots from. They can also include different types of data.

greetings <- c("Hey", "Hi", "Howdy", "Hello", "Morning")

n <- c(99, 15, 324, 54, 23)

df <- data.frame(greetings, n)

View the data frame you put together and what types of data are in your data frame?

# print (df)
# df
# View(df)

Data are representative of a type of greeting in the first column, and the number of times the greeting was observed (n).

4.1 Isolating elements of a data frame

We need to use specific R syntax to pull out individual rows and columns of a data frame. Isolate a single column.

We will use this [row, column]

# df$

# df[]

4.2 Modify the data frame

Let’s use the command rbind() to add on a row. We forgot to add in the greeting “Afternoon”, let’s say this was observed 18 times.

addition <- c("Afternoon", 18)
# ?rbind()

df <- rbind(df, addition) # writing over the original df

And let’s change the headers so they are more meaningful.

colnames(df)
names(df)

colnames(df)[1:2]

colnames(df)[1:2] <- c("Greeting", "Observed")

4.3 Activity

Isolate the number of times “howdy” was observed. Set this equal to an R object.

4.4 Activity

Set the number of times a greeting occured to a numeric.