operators-data-types.qmd

# Operators & Data Types {#sec-operators-data-types}

```{r}
#| echo: false
#| message: false

source("_common.R")
```

> ### Learning Objectives {.unnumbered}
>
> * Be able to use R as a calculator.
> * Be able to compare values in R.
> * Know the distinctions between how R handles different types of data types (numbers, strings, and logicals).
>
> ### Suggested Readings {.unnumbered}
>
> * Chapter 3 of Danielle Navarro's book ["Learning Statistics With R"](https://learningstatisticswithr.com/book/introR.html)

## R as a calculator

You can do a ton of things with R, but at its core it's basically a fancy calculator. Let's get started with some basic arithmetic!

### Doing basic math

R handles simple arithmetic using the following **arithmetic** operators:

<div style="width:500px">

```{r}
#| label: 'arithmetic'
#| echo: false

knitr::kable(rbind(
    c("addition", "`+`", "`10 + 2`", "`12`"),
    c("subtraction", "`-`", "`9 - 3`", "`6`"),
    c("multiplication", "`*`", "`5 * 5`", "`25`"),
    c("division", "`/`", "`9 / 3`", "`3`"),
    c("power", "`^`", "`5 ^ 2`", "`25`")
), 
    col.names = c(
        "operation", "operator", "example input" , "example output"), 
    align = "lccc"
)
```

</div>

The first four basic operators (`+`, `-`, `*`, `/`) are pretty straightforward and behave as expected:

```{r}
7 + 5 # Addition
7 - 5 # Subtraction
7 * 5 # Multiplication
7 / 5 # Division
```

Not a lot of surprises (you can ignore the `[1]` you see in the returned values...that's just R saying there's only one value to return).

Powers (i.e. $x^n$) are represented using the `^` symbol. For example, to calculate $5^4$ in R, we would type:

```{r}
5^4
```

### Slightly more tricky math

There are two other operators that are not typically as well-known as the first five but are quite common in programming:

<div style="width:500px">

```{r}
#| label: 'mod'
#| echo: false

knitr::kable(
    rbind(
    c("integer division", "`%/%`", "`4 %/% 3`", "`1`"),
    c("modulus", "`%%`", "`8 %% 3`", "`2`")), 
    col.names = c(
        "operation", "operator", "example input" , 
        "example output"), 
    align="lccc")
```

</div>

#### Integer division

Integer division is division in which the remainder is discarded. Note the difference between regular (`/`) and integer (`%/%`) division:

```{r}
4 / 3 # Regular division
```

```{r}
4 %/% 3 # Integer division
```

With integer division, 3 can only go into 4 once, so `4 %/% 3` returns `1`.

With integer division, dividing a number by a larger number will always produce `0` (because the larger number cannot go into the smaller number):

```{r}
4 %/% 5 # Will return 0
```

#### The Modulus operator

The **modulus** (aka "mod" operator) returns the remainder after doing integer division. For example:

```{r}
17 %% 3
```

This returns `2` because because 17 / 3 is equal to 5 with a remainder of 2. The modulus returns any remainder, including decimals:

```{r}
3.1415 %% 3
```

If you mod a number by itself, you'll get `0` (because there's no remainder):

```{r}
17 %% 17 # Will return 0
```

Finally, if you mod a number by a larger number, you'll get the smaller number back since it's the remainder:

```{r}
17 %% 20 # Will return 17
```

### Tricks with `%%` and `%/%`

The `%%` and `%/%` operators can be really handy. Here are a few tricks.

#### Odds and evens with `n %% 2`

You can tell if an integer `n` is even or odd by using `m %% 2`. If the result is `0`, `n` must be even (because 2 goes in _evenly_ to even numbers with no remainder). If `n` is odd, you'll get a remainder of `1`. Here's an example:

```{r}
10 %% 2 # Even
11 %% 2 # Odd
```

This trick also works with negative numbers!

```{r}
-42 %% 2 # Even
-43 %% 2 # Odd
```

### Number "chopping" with 10s

When you use the mod operator `%%` on a **positive** number with factors of 10, it "chops" the number and returns everything to the _right_ of the "chop" point:

```{r}
123456 %% 1 # Chops to the right of the *ones* digit
123456 %% 10 # Chops to the right of the *tens* digit
123456 %% 100 # Chops to the right of the *hundreds* digit
```

Integer division `%/%` works the same way, except it returns everything to the _left_ of the "chop" point:

```{r}
123456 %/% 1 # "Chops to the right of the ones digit
123456 %/% 10 # "Chops to the right of the tens digit
123456 %/% 100 # "Chops to the right of the hundreds digit
```

This trick works with non-integers too!

```{r}
3.1415 %% 1
3.1415 %/% 1
```

But **be careful** - this "trick" only works with _positive_ numbers:

```{r}
-123.456 %% 10
-123.456 %/% 10
```

Here's some mental notes to remember how this works:

- `%%` returns everything to the _right_ (`<chop> ->`)
- `%/%` returns everything to the _left_ (`<- <chop>`)
- The "chop" point is always just to the _right_ of the chopping digit:

<div style="width:500px">

Example         | "Chop" point | "Chop" point description
----------------|--------------|----------------------------
`1234 %% 1`     | `1234 | `    | Right of the `1`'s digit
`1234 %% 10`    | `123 | 4`    | Right of the `10`'s digit
`1234 %% 100`   | `12 | 34`    | Right of the `100`'s digit
`1234 %% 1000`  | `1 | 234`    | Right of the `1,000`'s digit
`1234 %% 10000` | ` | 1234`    | Right of the `10,000`'s digit

</div>

## Comparing things in R

Other than simple arithmetic, another common programming task is to compare different values to see if one is greater than, less than, or equal to the other. R handles comparisons with **relational** and **logical** operators.

## Comparing two things

To compare two things, use the following **relational** operators:

- Less than: `<`
- Less than or equal to : `<=`
- Greater than or equal to: `>=`
- Greater than: `>`
- Equal: `==`
- Not equal: `!=`

The *less than* operator `<` can be used to test whether one number is smaller than another number:

```{r}
2 < 5
```

If the two values are equal, the `<` operator will return `FALSE`, while the `<=` operator will return `TRUE`: :

```{r}
2 < 2
2 <= 2
```

The "greater than" (`>`) and "greater than or equal to" (`>=`) operators work the same way but in reverse:

```{r}
2 > 5
2 > 2
2 >= 2
```

To assess whether two values are equal, we have to use a double equal sign (`==`):

```{r}
(2 + 2) == 4
(2 + 2) == 5
```

To assess whether two values are _not_ equal, we have to use an exclamation point sign with an equal sign (`!=`):

```{r}
(2 + 2) != 4
(2 + 2) != 5
```

It's worth noting that you can also apply equality operations to "strings," which is the general word to describe character values (i.e. not numbers). For example, R understands that a `"penguin"` is a `"penguin"` so you get this:

```{r}
"penguin" == "penguin"
```

However, R is very particular about what counts as equality. For two pieces of text to be equal, they must be _precisely_ the same:

```{r}
"penguin" == "PENGUIN"        # FALSE because the case is different
"penguin" == "p e n g u i n"  # FALSE because the spacing is different
"penguin" == "penguin "       # FALSE because there's an extra space on the second string
```

## Making multiple comparisons

To make a more complex comparison of more than just two things, use the following **logical** operators:

- And: `&`
- Or: `|`
- Not: `!`

**And**:

A logical expression `x & y` is `TRUE` only if *both* `x` and `y` are `TRUE`.

```{r}
(2 == 2) & (2 == 3) # FALSE because the second comparison if not TRUE
```

```{r}
(2 == 2) & (3 == 3) # TRUE because both comparisons are TRUE
```

**Or**:

A logical expression `x | y` is `TRUE` if *either* `x` or `y` are `TRUE`.

```{r}
(2 == 2) | (2 == 3) # TRUE because the first comparison is TRUE
```

**Not**:

The `!` operator behaves like the word *"not"* in everyday language. If a statement is "not true", then it must be "false". Perhaps the simplest example is

```{r}
!TRUE
```

It is good practice to include parentheses to clarify the statement or comparison being made. Consider the following example:

```{r}
!3 == 5
```

This returns `TRUE`, but it's a bit confusing. Reading from left to right, you start by saying "not 3"...what does that mean?

What is really going on here is R first evaluates whether 3 is equal to 5 (`3 == 5`), and then returns the "not" (`!`) of that. A better version of the same thing would be:

```{r}
!(3 == 5)
```

### Order of operations

R follows the typical BEDMAS order of operations. That is, R evaluates statements in this order^[For a more precise statement, see the [operator precedence](http://stat.ethz.ch/R-manual/R-devel/library/base/html/Syntax.html) for R.]:

1. **B**rackets
2. **E**xponents
3. **D**ivision
4. **M**ultiplication
5. **A**ddition
6. **S**ubtraction

For example, if I type:

```{r}
1 + 2 * 4
```

R first computes `2 * 4` and then adds `1`. If what you actually wanted was for R to first add `2` to `1`, then you should have added parentheses around `1` and `2`:

```{r}
(1 + 2) * 4
```

A helpful rule of thumb to remember is that **brackets always come first**. So, if you're ever unsure about what order R will do things in, an easy solution is to enclose the thing you want it to do first in brackets.

> Note that for logical operators, the order precedence is `! > & > |`

For example, consider the following statement:

```{r}
TRUE | FALSE & FALSE
```

This returns `TRUE` because the `&` statement (`FALSE & FALSE`) is evaluated first, so the whole statement simplifies to `TRUE | FALSE`, which returns `TRUE`. If you put parentheses around the `|` statement, it would evaluate first and the whole statement would return `FALSE`:

```{r}
(TRUE | FALSE) & FALSE
```

Similarly, consider the following statement:

```{r}
! TRUE | TRUE
```

This returns `TRUE` because the `!` statement is evaluated first (`! TRUE` is `FALSE`), and the simplified statement `FALSE | TRUE` returns `TRUE`. Again, if you put parentheses around the `|` statement the whole statement becomes `FALSE`:

```{r}
! (TRUE | TRUE)
```

## Data types

Every programming language has the ability to store data of different types. R recognizes several important basic data types (there are others, but these cover most cases):

Type      | Description                 | Example
----------|-----------------------------|----------
`double`   | Number with a decimal place (aka "float") | `3.14`, `1.61803398875`
`integer`   | Number without a decimal place    | `1`, `42`
`character` | Text in quotes (aka "string") | `"this is some text"`, `"3.14"`
`logical`   | True or False (for comparing things) | `TRUE`, `FALSE`

If you want to check with type a value is, you can use the function `typeof()`. For example:

```{r}
typeof("hello")
```

### Numeric types

Numbers in R have the `numeric` data type, which is also the default computational type. There are two types of numbers:

- **Integers**
- **Non-integers** (aka "double" or "float")

The difference is that integers don't have decimal values. A non-integer in R has the type "`double`":

```{r}
typeof(3.14)
```

By default, R assumes all numbers have a decimal place, even if it _looks_ like an integer:

```{r}
typeof(3)
```

In this case, R assumes that `3` is really `3.0`. To make sure R knows you really do mean to create an integer, you have to add an `L` to the end of the number^[Why `L`? Well, it's a bit [complicated](https://stackoverflow.com/questions/24350733/why-would-r-use-the-l-suffix-to-denote-an-integer), but R supports complex numbers which are denoted by `i`, so `i` was already taken. A quick answer is that R uses 32-bit _long_ integers, so `L` for "long".]:

```{r}
typeof(3L)
```

### Character types

A character value is used to represent string values in R. Anything put between single quotes (`''`) or double quotes (`""`) will be stored as a character. For example:

```{r}
typeof('3')
```

Notice that even though the value _looks_ like a number, because it is inside quotes R interprets it as a character. If you mistakenly thought it was a a number, R will gladly return an error when you try to do a numerical operation with it:

```{r}
#| error: true

'3' + 7
```

It doesn't mattef if you use single or double quotes to create a character. The only time is _does_ matter is if the character is a quote symbole itself. For example, if you wanted to type the word `"don't"`, you should use double quotes so that R knows the single quote is part of the character:

```{r}
typeof("don't")
```

If you used single quotes, you'll get an error because R reads `'don'` as a character:

```{r}
#| error: true

typeof('don't')
```

We will go into much more detail about working with character values later on in [Week 7](L7-strings.html).

### Logical types

Logical data only have two values: `TRUE` or `FALSE`. Note that these are not in quotes and are in all caps.

```{r}
typeof(TRUE)
typeof(FALSE)
```

R uses these two special values to help answer questions about logical statements. For example, let's compare whether `1` is greater than `2`:

```{r}
1 > 2
```

R returns the values `FALSE` because 1 is not greater than 2. If I flip the question to whether `1` is _less_ than `2`, I'll get `TRUE`:

```{r}
1 < 2
```

### Special values

In addition to the four main data types mentioned, there are a few additional "special" types: `Inf`, `NaN`, `NA` and `NULL`.

**Infinity**: `Inf` corresponds to a value that is infinitely large (or infinitely small with `-Inf`). The easiest way to get `Inf` is to divide a positive number by 0:

```{r}
1/0
```

**Not a Number**: `NaN` is short for "not a number", and it's basically a reserved keyword that means "there isn't a mathematically defined number for this." For example:

```{r}
0/0
```

**Not available**: `NA` indicates that the value that is "supposed" to be stored here is missing. We'll see these much more when we start getting into data structures like vectors and data frames.

**No value**: `NULL` asserts that the variable genuinely has no value whatsoever, or does not even exist.

## Page sources {.unnumbered}

Some content on this page has been modified from other courses, including:

- Danielle Navarro's book ["Learning Statistics With R"](https://learningstatisticswithr.com/book/introR.html)
- Jenny Bryan's [STAT 545 Course](http://stat545.com/)
- [RStudio primers](https://rstudio.cloud/learn/primers/1.2)
- Xiao Ping Song's [Intro2R crash course](https://github.com/xp-song/Intro2R)