Intro to Jamovi (Lab 1)

Today’s focus is on learning to do some basics in jamovi. You may already know how to do these things!

jamovi is a statistical package that will allow you to conduct most of the analyses you will learn in PSY 202 without calculating every stastisic by hand. jamovi is an open source graphical user interface (GUI) that runs on the R statistics platform. (R is a coding based language–by using a GUI you will be able to conduct your analyses without needing to code.) The goal for today’s lab is to get familiar with jamovi and set you up for success when doing your own analyses this semester, and later in your 300-level classes and during your Senior Project.

There are a lot of resources about jamovi on the Internet and you should make use of them while you do your analyses. Even seasoned statisticians need to figure things out from time to time, so some of the exercises will go beyond what is strictly demonstrated in class to get you used to problem-solving these kinds of issues. You should give a good faith attempt at solving these problems on your own (by looking in your textbook, searching the Internet, watching some of the many jamovi videos on the web, and working with your classmates). Of course, if you are really stuck, you should call the instructor or the study assistant over for a hint or to help troubleshoot.

If you are using a lab computer: It’s already installed!

If you are using your own laptop: Begin by being sure that you have jamovi installed. Instructions can be found here or you may just install the software from the jamovi website. (The “solid” download is fine.)

Work through the following questions—and wave me down if you have questions or get bogged down, although figuring it out yourself is ideal in this scenario.

One last note: I’ve added extra info in these collapsed tips. Click on them if you would like extra info, but feel free to skim over the collapsed ones if you know the answers. Later this semester, these will be places where I will answer questions, etc.

Learn about the data window (expand for more)

1. The first screen that you see in Jamovi is where you enter your data (or, where your data will appear when you import it). Once you do item 1 below, you should be seeing the penguins data here.
2. On the left-hand side is a spreadsheet-like array. In general, each variable will be a column, and each observation will be a single row.
3. On the right-hand side is a currently blank window where your analyses will appear.
4. There are four different menus in blue at the top: Variables, Data, Analyses, and Edit. We will discuss them more soon. All the way at the left—in front of “Variables”—is the hamburger menu (). Clicking on that gives you options like New [data], Open [data or analyses], Special Import, Save, Save As, and Export.
5. Within the Data menu are different options such as paste, setup, compute, etc. We will learn what some of those do later today, or later in the semester.

1. Import data. Find the dataset on Brightspace or from github. Download it. Then, figure out how to open/load/import the data into Jamovi. You can refer to the textbook chapter, section 3.4, for this, or just play around on your own until the data has been imported.

2. Using the Data menu in Jamovi, double click on one of the variable names (e.g., “species”). What kind of variable is it? Go through the remainder of the variables and determine whether they are the correct “type” of variables based on what we discussed in class? You’ll probably want to read about the dataset, which you can do here.

   This is a very important thing to remember about jamovi: the program only knows what we tell it about each variable and we want to give it as much information as we can about each variable. This process will be important for you; when you return to a data set and don’t quite remember what each variable was, it’s helpful to have all this info. This is why sometimes you want to add a description, especially if something is confusing. You could, as you read about penguins, explain what some of these variables are… but most of them are self-explanatory.

   In the future, you may find data that includes a codebook. A codebook tells you what each variable/column is, and what each value means. This is important for sharing data broadly.

   Within this variable menu, you can see the following pieces:

   1. Variable name (species)
   2. Variable description (currently blank; could add one—this is not used in analysis)
   3. Measure type (this is a nominal or categorical variable—it’s a grouping variable—but others include Continuous, Ordinal, and ID [identification])
   4. Data type (text, integer, or decimal; species is text)
   5. Missing values (see below)
   6. To the right, Levels (all possible values for nominal or ordinal values)

   Missing values: Within this variable menu, it also asks how to treat missing values. (For example, sometimes a research participant skips a question, or data is missing.) jamovi defaults to leaving these values blank. For example, you should see that row 4 has an Adelie penguin who doesn’t have that information. In some cases, you can handle them differently, but we will not do that in this class.

3. Click the up arrow to the right of the variable name, which will close the variable info menu. Play around with other settings in the Data main menu.

What is a variable? (expand for more)

A variable is a symbol, often a letter, that represents an unknown or changeable value. For example, in the basic math equation y=x+2, y and x are variables. But so is species in the penguins data. Variables can take on different values depending on the situation or context.

For the most part, I will use this kind of type in our labs to designate variables: var. (I’ll also use this for something closer to code, as you’ll see below.)

jamovi variables are case sensitive, so I would recommend that you generally enter your variable names all in lowercase. Sometimes using underscores can be helpful like, e.g., naming a variable age_in_years.

4. Under the Data main menu, click Filters. A new part of the window will slide out. You may want to read the Jamovi help page on filters.

   Imagine that we were only interested in Chinstrap penguins (not Adelie or Gentoo). Well, we can filter those out with this function. \(f_x\) is where you will type your filter. We can use simple ones or more advanced ones.

   If you set your filter = species == "Chinstrap" (with a double equal sign, which refers to testing for equivalence—a single equal sign will not work), and then click elsewhere, what happens? Some rows should be grayed out. (Note that Chinstrap penguins are at the bottom!) Try clicking the eye symbol () which should turn into an eye with a slash through it (like ). What happens? Make all rows show up. Then play around with the “active” toggle switch (), etc. Try to make sure you understand what is happening here.

You have to be careful about quotation marks!

Do you see a difference between these marks: ” and "? (Besides the gray background!) jamovi expects you to use straight quotation marks, not curly ones.

And numbers don’t need quotation marks at all! We only need quotation marks for something like "Chinstrap", not for something like 40.

Variable names also do not need quotation marks, ever. So species is on its own, as is bill_length_mm.

5. Try another filter by replacing the one you have, or by turning it off. For the new filter, you want to select only rows from this penguins data where the bill length in millimeters (column name: bill_length_mm) is more than 40, and where the species is Adelie (not Gentoo or Chinstrap). Can you make that happen?

What if you want to use more than one criterion? How do I do the above question? (expand for more)

Sometimes, you may want multiple filters. There are two plus signs that let you do that. Hitting the plus sign () to the left (directly above the eye symbol ) will add a totally new filter. This can be turned off/on independently of the first filter.

Hitting the plus sign to the right (within your \(f_x\) window) will add another condition to the current filter. It uses the word and because it’s combining them automatically. (We’ll tackle or at a later point.)

So, to do the above:

1. Set species == "Adelie"
2. Click the plus sign at the right () to add a new filter
3. Then, write code to test if bill_length_mm is more than (what symbol should you use?) 40.

6. Select the body_mass_g column by clicking on the name, and then use the Transform button in the Data main menu to convert body_mass_g into body_mass_kg. That means you’re dividing the $source (i.e., the original data) by 1000, right? Can you make that happen? This is a simple transform, but we could do more complicated ones… The first row, e.g., should go from 3750 in body_mass_g to 3.75 in body_mass_kg.

What is Transform and how do I do this? (expand for more)

Transform allows you to use a function to change a variable. When you open the menu by clicking Transform, it’ll first let you define your new variable. I would call this one body_mass_kg but otherwise you can leave this blank.

Then, you will click on the dropdown under “Using transform” and select “Create new transform”. You can name the transform if you like (“divide by 1000”) but the important thing is the “recode condition”. It uses the variable $source. Don’t touch that part. Instead, just write /1000—the whole box should read $source/1000. Then you can click the down arrow .

7. Now click on the Analyses tab of the main menu ribbon. It’s okay if you still have filters on, or if you turn them off, so long as you have some data to use. Click on “Exploration”, with the small image of a histogram, all the way to the left. Then click “Descriptives”.

   Can you figure out how to make a frequency table of island? (You will need to use the Plots menu, below.) What about histograms of body_mass_kg and flipper_length_mm? What do you learn from these?

8. What happens when you move a categorical (nominal) variable into the “Split by” box? Play around with this. What does “violin plot” do? Feel free to discuss with your classmates.

9. Save your data (if you want) using the hamburger menu () and then clicking Save As. Saving as a Jamovi file type (.omv) will keep any edits or filters; a CSV (comma separated values) file will be able to be opened in other programs like Excel/Google Sheets/R.

Finish before class is over? Wonderful! Play around with this data. Read the sections of the textbook more closely. See what you can do with creating figures from the data.

You should leave today’s class familiar with using Filters. If you’re not entirely sure, you should read this page again. Often students who struggle early on in this lab do so because they don’t put in the effort to understand how to use the filters. We will have time to practice! You can also discuss in the stats study rooms.

There is nothing to turn in today.