R is a programming language that was originally thought for statistical and data analysis, but it has grown to be able to do much more. It can handle and store data; perform all sort of calculations (also on arrays, such as matrices or vectors); visualise and plot data. It is free software and open source! Anyone can collaborate, create something new and integrate it in R. R allows users to add additional functionalities (defining new functions) and can be extended (easily) via packages. You can most likely find a package designed to suit whatever new needs you came up with! Try and google it! The core functionalities of R are typically referred to as “base R”. In this course, we will deviate from base R syntax in many regards, because we will be mostly relying on the tidyverse, i.e. a collection of packages with a coherent syntax that focuses on the idea of tidy data [link to tidy data when we have it]. However, the basics of R syntax is common to both programming styles, so let’s briefly discuss them. Because we are linguists and we are discussing a (programming) language, we can use some metaphors to easily grasp what we are talking about.
Natural languages have different word classes (nouns, verbs, adjectives…) and R has different data types. The three most important types that you need to know are character, numerical and boolean. Character elements must be introduced by quotes. Everything that is within a pair of quotes is one single character element, which means that in the following example there are three character elements, even if there are five words.
"Hello"## [1] "Hello""there"## [1] "there""this is Carlota"## [1] "this is Carlota"Numeric elements can be either real or decimal (and R calls them different, so your numeric elements can actually be of the integer type, the numeric type or the double type, but this is not too relevant for us!). These must not be introduced by quotes.
1## [1] 12.13## [1] 2.13865473846823## [1] 865473846823You can also treat numbers as character, as long as they are within quotes. Compare
'1' + '1' # will give us no result but warn us that we have used a non-numeric argument with a binary operator, that is the '+'with
1 + 1## [1] 2R cannot add character elements, but for sure it can add numerical elements!
Logical elements (or boolean elements in the statistical literature) are basically binary data types: one value is TRUE and the other one is FALSE. So they are basically answers to questions. I know this sounds a bit weird right now, but you’ll understand it better when we start using them. They can be abbreviated to T and F.
TRUE## [1] TRUET## [1] TRUEFALSE## [1] FALSEF## [1] FALSEData structures are more complex structures that combine different types of data types, so in a way they are similar to phrase-level elements in natural languages. We have unidimensional data structures, such as vectors and lists, and bidimensional data structures, such as data frames and matrices.
Vectors are a collection of elements of the same data type and we build them using the function c() (more on functions later). Each element must be separated by commas.
c("Hello", "there", "this is Carlota")## [1] "Hello" "there" "this is Carlota"c(1, 2, 3, 3.141596)## [1] 1.000000 2.000000 3.000000 3.141596Lists are a collection of elements of different data types (or even a collection of vectors). We build them through the function list(). They are more complicated to handle and we will be using mostly vectors, but lists will come up eventually.
list("Hello", 2, 3, "there", T, "this is Carlota", 1, 3.141596, FALSE)Matrices and data frames have dimensions: the number of rows and columns. In everyday language we would normally call them tables, but incidentally “table” is also an R data type… (Don’t worry about it.) While matrices have elements of the same data type, data frames have elements of different types. We will be using almost exclusively the latter. Every column of a data frame is a… vector! (Or a list.)
Missing data are handled in a special way in R. They are represented as NA. They can appear in all the data structures we covered. We will deal with them in the course :)
Variables are like proper nouns in a way. Whenever we want to store an element in R we need to give it a name. Variables are sequences of alphanumerical characters that are not within quotes! Variable names cannot: 1) start by a number, 2) contain - + / *. Variables name, however, can contain . _ Also, it’s probably for the best if you do not use special characters in your variable names (á ñ ü).
We assign names through the arrow operator:
two <- 2Now you can use two to call 2 (which is not very useful, but that’s because the variable we created is not very complex!) I promise this will be tremendously useful.
two + 3## [1] 5Functions are pretty much like verbs. They even have arguments! A function is an instruction to perform a specific task. Functions have arguments, which specify on which element you need to perform the instructions (the subject?) and some other requirements (objects and adjuncts!). This metaphor is great. Arguments have a default order, but they also have names, that we use when we don’t want to follow the default order (or when we are not sure about it).
mean(c(two, 3)) #The function mean() takes a numeric vector## [1] 2.5log(3, 2) #The function log() takes the numeric vector for which the logarithm is computed and as its first argument and a positive number which defines the base of the logarithm as the second one. (If you don't remember what a logarithm is, don't worry! This is just an example)## [1] 1.584963You can check the arguments of a function by calling the help file (which will appear in the the bottom right window in RStudio).
?mean"Carlota" == "carlota"## [1] FALSEmean( c(2, 3 ) )## [1] 2.5mean( c(2, 3))## [1] 2.5mean(c(2,3 ) )## [1] 2.5mean(c(2, 3))## [1] 2.5…unless they are inside a name or an element):
"Carlota" == "Ca rlota"
t wo <- 2You might have found this a lot already. Trust me: you will get the grasp of it much faster than you think! I recommend that you do a few of the starting swirl lessons from the R Programming course in order to practice these concepts a bit. You might not see the point of all this at first, because this is like learning inflectional morphology of a new language. A bit boring and useless unless you can build sentences! But you can also try and start building sentences right away by starting with our course first. I still think is a good idea that you do the basic courses at some point, but you’ll might find more motivation if you shuffle the order :)
When you learn a language, you make mistakes all the time. In natural languages, mistakes do not necessarily prevent communication, because the hearer is collaborative (and a nice person). In programming languages, mistakes make communication fail, because computers are… quite dumb, actually. They have no idea about Gricean rules or anything similar.
So what to do if our code is not working?
Read the error message R gives you. You might not understand most of what it says, but you’ll get better at it. It will give you a clue about where to look.
Most mistakes in your code (especially at the beginning) will be typos: is there a comma missing? Are all brackets and quotation signs closed (and where they should be)? Did you use a capital letter where you shouldn’t? Is everything spelled correctly? Did you run your previous code or simply write it on the script?
Sometimes it’s hard to find the typos – it helps if someone else looks at it.
Grammatical mistakes: maybe your syntax is wrong, are you sure you’re using that function right?
Run ?function() to get a help file. These files are not always easy to read, but you’ll get better at it too!
If you can’t still figure it out: just ask! We have a forum here! :)
Google it: copy your error message or write what you were trying to do – there are lots of helpful forums! You’ll also get used to understanding their answers.
Coding is a lot of fun, but it can also be a source of frustration… Having someone going through the same process really helps a lot. Try to convince a pal to do this course with you. Your motivation will be better and you will find each other extremely helpful! And remember that you can always ask me!