# Data Types When we were making some vectors in the last lesson, you might have noticed a word appear in the environment panel between the name of the vector, and the values it contained**:** ![You can see the data type of your vectors in the environment panel!](https://2245987779-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M38c0s4s0YCM6dGwrz1%2F-M5o5FZ2-gUE2tmQpQWK%2F-M5pT0pgWGSTIreiG2AW%2FScreen%20Shot%202020-04-26%20at%205.49.37%20pm.png?alt=media\&token=baad7670-0477-4fc4-b7fd-510b59c781a9) These are the **types** of data that make up the vector, and determine the properties of the data, and what kind of manipulations you can perform on it. \ In R, *all* data objects have a data type. In this lesson we are going to have look at 5 important data types, which from most simple to most complex are: * Logicals `logi` * Integers `int` * Doubles `num` * Complex Numbers `cplx` * and Strings `chr` Let's go into a little detail about these different data types. ## Logicals The simplest of data types you can encounter in R are the **logicals**. Logical type data can take one of two values, either `TRUE` or `FALSE.`\ We have already encountered logicals in the last chapter, when performing comparisons. ```r > 5 >= 3 [1] TRUE > c(1, 2, 3, 4, 5) >= c(5, 4, 3, 2, 1) [1] FALSE FALSE TRUE TRUE TRUE ``` If you're feeling lazy, logicals can also be defined just by using `T` and `F` ```r > T [1] TRUE > F [1] FALSE ``` While it's a little unusual to assign logicals as variables, it is quite common to receive them as outputs from tests or functions you might run and use them in conditional `if else` type statements (more on this in a later lesson). ## Integers Moving one step up in complexity from logicals, we get to **integers**, which are used to represent, well, integers?\ To differentiate themselves from the more common double type, integers in R are written as a number followed by a capital `L`. Let's explore this difference using the `typeof()` function, which will return the **data type** of an object. ```r # not an integer > typeof(1) [1] "double" # is an integer > typeof(1L) [1] "integer" ``` For the most part, R will automatically handle and swap between integers and doubles as necessary, and the difference is generally not something you generally need to worry about. ```r > 1L + 1.5 # although we started with an integer [2.5] # R knows we want a double at the other end ``` That said, integers do serve an important purpose when R passes code to software written in`C`or `FORTRAN,` but this is well beyond the scope of this course. At the very least, now you know integers exist! ## Doubles In most cases, your numeric R data is going to be stored using the **double** data type. Doubles are used to represent any the real numbers you might want to use. ```r > typeof(1) [1] "double" > typeof(3.4) [1] "double" > typeof(-3e15) [1] "double" > typeof(1/3) [1] "double" ``` #### Why the name? The term *double* is short for *double-precision number*, meaning they can store twice the information as our integers. As well as storing decimal values, this means you can store much bigger numbers with doubles than integers. ```r # the largest possible integer value... > .Machine$integer.max [1] 2147483647 # is much smaller than the largest possible double > .Machine$double.xmax [1] 1.797693e+308 # remember, e+308 means x 10 to the power of 308 ``` However, they also take up more space in storage than integers. ```r > object.size(as.integer(1:1000)) 4048 bytes > object.size(as.numeric(1:1000)) 8048 bytes ``` ## Complex Numbers Moving up in complexity from our doubles, we get to imaginary and **complex** numbers.\ In R, a complex number are written as ` + i.` ```r # R will try to stay in the real world > sqrt(-1) [1] NaN Warning message: In sqrt(-1) : NaNs produced # unless you explicitly tell it not to > sqrt(-1 + 0i) [1] 0 + 1i ``` For some of us it would be pretty rare to need to delve into the world of complex numbers, but R has full support for complex algebra if that's what takes your fancy. ## Strings/Characters Characters are how R deals with data that comes in the shape of words, rather than numbers. \ Characters can consist of any characters you want, as long as they're wrapped up inside a pair of double `""` or single `''`inverted commas. ```r # This is a string of letters > "abcdefg" [1] "abcdefg" # And a string of numbers > '12345678' [1] '12345678' # And a string of words > "string of words" [1] "string of words" ``` A common slipping place for new R users is dealing with strings of numbers. While it's perfectly reasonable to do something like ```r > 10 + 22.5 [1] 32.5 ``` Trying to do the same thing with numbers in characters will land you in some strife! ```r > "10" + "22.5" Error in "10" + "22.5" : non-numeric argument to binary operator ``` ## Type Coercion Just because your data starts as one type, that doesn't mean it needs to stay that way! To change data from one type to another, we can use the inbuilt **type coercion** functions. To coerce your data into: * A logical, use `as.logical()` * An integer, use `as.integer()` * A double, use `as.numeric()` * A complex number, use `as.complex()` * Or a string, use `as.character()` While you can always coerce a less complex data type to a more complex one, the reverse isn't always true. ```r > as.numeric(TRUE) [1] 1 # no problemo > as.character(10) [1] "10" # all good here too > as.numeric("scooby snacks") [1] NA Warning message: NAs introduced by coercion # ruh roh! ``` ## Challenges Now that we've explored the different **types** of data in R, here are a few challenges to test what you've learned. #### Challenge 1 {% tabs %} {% tab title="Challenge" %} ```r # What is the data type of each of the following objects: > 1 > 5L > sqrt(-1 +0i) > "25" ``` {% endtab %} {% tab title="Solution" %} ```r > typeof(1) [1] "double" > typeof(5L) [1] "integer" > typeof(sqrt(-1 +0i)) [1] "complex" > typeof("25") [1] "character" ``` {% endtab %} {% endtabs %} #### Challenge 2 {% tabs %} {% tab title="Challenge" %} ```r # Which of these coercions work? What do they produce? > as.numeric(FALSE) > as.logical("Hello!") > as.integer(4.4) > as.character(TRUE) > as.numeric("scooby snacks") ``` {% endtab %} {% tab title="Solution" %} ```r > as.numeric(FALSE) [1] 0 > as.logical("Hello!") # Coercion from string to logical is not defined [1] NA > as.integer(4.4) [1] 4 > as.character(TRUE) [1] "TRUE" > as.numeric("scooby snacks") # Coercion from string to double is not defined [1] NA Warning message: NAs introduced by coercion ``` {% endtab %} {% endtabs %} #### Challenge 3 {% tabs %} {% tab title="Challenge" %} ```r # What is the output of each of these lines of code? Why? > sqrt(-1) > "1" + "5" > 1 + 7 > TRUE + FALSE ``` {% endtab %} {% tab title="Solution" %} ```r > sqrt(-1) [1] NaN # R will try to stay in the real world try sqrt(-1 +0i) instead Warning message: In sqrt(-1) : NaNs produced > "1" + "5" # You can not add 2 characters Error in "1" + "5" : non-numeric argument to binary operator > 1 + 7 [1] 8 > TRUE + FALSE # They are coerced as numbers as 1 and 0 respectively [1] 1 ``` {% endtab %} {% endtabs %} --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://r-unimelb.gitbook.io/rbook/demystifying-data-types/untitled-1.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.