Question

So I get a bit confused by the lm()-command. I tried it with lm(x~y, mydata) and lm(y~x, mydata) and I got different output.
So is that just which variable to use as x and which one to use as y? I’m sorry to ask such a noob question but I am not sure and I coulnd’t find anything explaining the parameters of that command!

Answer 1

The answers can be found on the help page for the function. In the Details section we have:

A typical model has the form response ~ terms where response is the (numeric) response vector and terms is a series of terms which specifies a linear predictor for response.

There are more details (also linked to from the lm help page to formula. In the details sections for formula, we have:

The ~ operator is basic in the formation of such models. An expression of the form y ~ model is interpreted as a specification that the response y is modelled by a linear predictor specified symbolically by model.

So to summarize, you define your model in symbolic terms where the LHS is your response variable, and the RHS are your predictor variable(s). You get different answers because in one model, y is your response variable and the other is x.

If you weren’t aware, you can access the help page for nearly all functions with ? at the command line, i.e. ?lm or ?formula.