[Auto-generated transcript. Edits may have been applied for clarity.]
So in this video, I'll walk you through how to create a new variable using the mutate function.

Now we'll use an example of combining the responses of several questions into a combined score.

But you can use mutate for many other things as well.

Any kind of new variable that you want to create. First things first we need to load the libraries.

So we've done that before. So we'll we need the library here.

And we need the library tidy verse. Then we need to read in the data again.

We've done that before so I won't dwell on it here. But we used to read underscore CSV function.

We tell our where to look and what the file is called.

And we assign that to a data frame called data under the underscore.

MW. Okay.

So so far when we've used that data we've looked at the responses to individual items.

So question one for instance. However, single questionnaire items are noisy,

so a person's answer on to one question could be shaped by how they read the wording of that particular question,

or how they applied the response options to that particular question.

Or maybe they were distracted while they answered that question.

So if you had, it is much better to have several items that tap into the same underlying construct.

And then if that's the case, you can combine the responses to those items into one score.

So that would usually be a mean or a kind of you would add them up.

So some. And that lets these item specific quirks average out.

So the score is more reliable. And it is a better reflection of the thing that we care about.

Um, so the, uh, the brief mind wandering three factor skill that we've been working with, the BMW three is a good example of that.

So it has 12. Um. It has. It's a questionnaire with 12 items.

Um, so it treats mind wandering as kind of task, unrelated thought, but it measures three sub dimensions.

So unintentional mind wandering, um, intentional mind wandering and matter awareness of mind wandering.

So rather than asking one question about mind wandering and hoping, it captures everything, uh,

the skill uses four items for each dimension and combines them into a subscale score by averaging.

So we're going to calculate a score for unintentional mind wandering first.

So and the authors of this questionnaire, they define this as task unrelated thoughts um that occur spontaneously.

So that is in contrast to intentional mind wandering which they define as task unrelated thoughts,

where attention is deliberately redirected away from the current activity.

Okay, so let's compute a single score for unintentional mind wandering.

Now, to do that, we use the mutate function. Um, and this is what it requires.

So if you want to create a new variable you can use mutate.

Um, and then here on line two we tell it which data to use.

So that is our data frame. And then we tell it what the new variable should be called.

So that is unintentional underscore MW in this case then an equal sign.

And then we specify how that variable is computed.

So in this case we're going to calculate the average of the responses to these four items.

So this is item unintentional mind wandering two and 123 and four.

So we add those up. We divide it by four. And that gives us the average.

So unintentional mind wandering will be the average of the responses across those four items.

So I've just listed the those different items here at the bottom.

Um, to kind of make it a little, a little bit more concrete.

So the first item was while listening to a presentation, my thoughts started to trail off unintentionally.

Um, the second is when watching T.V., other things inadvertently cross my mind.

Um, I am often absent minded is the third item, and the fourth one is when I am engaged in an activity.

My thoughts wandered to other things all by themselves. You can kind of see how each of those kind of taps into that same, um,

construct of unintentional mind wandering that is thought, a task, unrelated thoughts, that happens spontaneously.

Now we can do the same thing for the second dimension of this questionnaire meta awareness of mind wandering.

Um, but if you look at the questions that make up the meta awareness of mind wandering skill,

you'll notice that the black ones and the red ones as listed on the slide are phrased kind of in opposite directions.

So, um, we need to do so.

So let's let's read them quickly. So, um, the black ones say I quickly catch myself when I am not listening attentively,

uh, or I immediately notice when my thoughts are not in here and now.

So if you score higher on that, um, you would be more show more awareness of your mind wandering.

In contrast, the items that are red on the slide are kind of read in the other direction.

So it takes a very long time for me to notice that my thoughts have wandered off, and it takes me a while before I realise that I zoned out.

So if we look at the scoring. So these were the response options that are available to the participants.

Right. So, um, and they are translated into a score of zero, one, two, 3 or 4.

Um, so if you look at item two, for instance, I quickly catch myself on.

I am not listening attentively. Um, if you would say fully agree, that would give you four, uh, points.

So, um, that would increase your score on that subscale of matter awareness of mind wandering.

So you would it would indicate that you're more aware of when you mind wander.

But if we take one of the read items, so it takes a very long time for me to notice that my thoughts have wandered off.

If you fully agree with that, that is kind of shows the opposite, right?

That you are not very aware that you that your your thoughts have wandered off.

Um, and we do we include items that are phrased in these opposite directions in a questionnaire to make sure that, um,

People actually read the questions and don't just, you know, circle,

select all the same response options on the same side of the, um, of the questionnaire.

Um, so yeah, that but but when we then come to calculating a combined score,

we need to pay attention to that because we need to reverse the scoring for that.

So if somebody's selected for, for this item in red and said yes, I fully agree with that, they would get four points.

So it would mean that it take a very long time to notice that their faults have wandered off.

We want to show that, um, we want to kind of make sure that that contributes to that overall score as showing,

um, little matter awareness of mind wandering. So we then need to reverse score that.

And we do that by calculating it in this way.

So if you have an item like that, you would create a new variable and say, okay, um, I'm going to take the maximum of the of the response options.

So for in this case and I'm going to um, subtract whatever their score on that item is.

So if they select it for, um, it would be four minus four.

So that is zero. So that translate is as a low score on metta awareness of mind wandering.

And that is what we need right for that particular red item.

Okay. So if I just quickly go back um, once we've done that we can do exactly the same thing as we did for unintentional mind wandering, right?

So then we would use mutate, tell it which data to use, tell it what the new variable is called,

and then explain a kind of write out how that variable is computed.

So we add the responses on all of these items. Divide by four and there you go.

You will see that. So the variable names here don't reflect that.

It is the reverse score but it is the reverse score. So that had already been done on that or in that particular data set.

Now finally, we have the dimension of intentional mind wandering.

And we do exactly the same thing there. So if we want to compute a single score for intentional mind wandering,

we again use mutate talent which data to use and tell it what the new variable name is, and explain how that is computed.

And there is no reverse scoring for any of those items. Um, yeah.

So that would give us a score that reflects the the kind of average response across these four items that you can see here at the bottom of the slide.

Now, once we've done that, we often want to save that file so that we don't have to do all of that again.

And this is how we do that. So we've used read underscore CSV a number of times already.

Now we're going to use write underscore CSV and that does the equivalent.

It basically writes a CSV file for you. So what we then need to do is that we specify, um, which data we want to use.

Um, we tell it where we wanted to store that file.

So here we use the here function again that we've encountered previously.

We specify the file path as which folder we want this file to go separated by a comma.

And then we specify the file name including the extension.

And just the same way as with read underscore CSV.

These um these the file path and the file name need to be in double quotation marks.

So we need to tell our which data frame to include. That is this first bit and then where to save it and what to call it.

And that's it for now.