Julius has incorporated a new feature that can help you perfect your graphs! The graph editor is a great solution to all your graphing needs! Let’s take a look at how to effectively use this new feature.
Scenario: you’re a researcher who is interested in studying the travel patterns of the Eastern Bluebird (Sialis sialis). Using a mist net, you capture three individuals, band them, place a glue on GPS tracker, and then release them. You monitor their movement over a period of time and receive the following data:
| band_num | heading |
|---|---|
| 1166 | 29 |
| 1166 | 167 |
| 1166 | 14 |
| 1166 | 233 |
| 1166 | 261 |
| 1166 | 32 |
| 1166 | 254 |
| 1166 | 236 |
| 1166 | 173 |
| 1166 | 280 |
| 1166 | 147 |
| 1166 | 132 |
| 1166 | 63 |
| 1166 | 300 |
| 1167 | 307 |
| 1167 | 41 |
| 1167 | 258 |
| 1167 | 325 |
| 1167 | 111 |
This is a snippet of the entire dataset.
You are interested in creating a polar plot to see what directions each individual travels.
Polar Plot Data Visualization
This is by far one of my favourite data visualizations because it looks so clean when you plot it. It also gives you a ton of information in a small visualization. Let’s plot this data!
Question 1: what are the heading directions for each individual?
Prompt 1: can you create a polar plot showing me the direction each bird travels? band_number corresponds to 3 birds (1166: bird 1; 1167: bird 2; 1168: bird 3) and heading is the direction the bird is going (out of 360 degrees).
Quick aside: notice how I describe how I want my data plotted. This includes defining which graph to use, explaining what the band_number means and what value corresponds to what bird, and explaining the parameters of heading and what they mean.
Doesn’t this just look lovely (I’m such a nerd, I know)? Okay, so this is nice, but I want to move the legend around because it is kind of intersecting with the polar plot. Instead of prompting again, we can do this handy little trick where we can edit the plot directly!
I’m going to use this tool to move the legend to a more appropriate place.
After clicking on the edit graph button, Julius will bring up the following options: Legend, Size, Title and Custom. Since I want to change the legend position first, I’ll click on that. Once you click on it, it gives you the option to show the legend. Since we have three different individuals in this study, it would be beneficial to have this legend.
Julius automatically defaults to the best legend position, which from my experience is usually in the top right corner. However, it looks like our plot is a little large and in charge, so we need to pick a different location for this. The lower right seems like a safe bet for my legend, so let’s choose that option:
Hmmm… it seems like the legend is still obscuring the polar plot. Don’t worry, I got another trick that can come in handy for issues like this.
Click on the size section. This should drop down a width and height box where you can manually modify the size of your graph. The default for me was 8 width and 8 height. Let’s play around with this and see if we can move make some room for the legend…
Success (for any of you Super Smash Bro fans, I hope you read this in the announcers voice). Okay step one complete, moving onto addressing the title. Julius gives you a very simplistic title for your figures, which is fine. But, I was always told when making graphs and figures for publications that, “your graphs/tables should NOT have a title”. Needless to say, I’m forever scarred so I’m going to get rid of this title. Click on the title section to pull up your current title.
Quick sidenote: Please make sure you SAVE your changes. Julius will not automatically save every one, so treat this process like you’re writing a 5000 word essay and your computer is on low battery life: save every change.
This is what comes up when you click on the Title section in the graph editor. As you can see I have the option to change my title, and label my x-axis and y-axis. Polar plots typically do not have an x- and y-axis label because most of the information on it is self-explanatory. Therefore, I’m going to leave that section blank. However, I will remove the title that says, “bird travel directions” (again, scarred for life).
Crisp and clean! This is almost ready to be placed in my research paper! There are a couple of things I still want to change in the legend. Bird 1, 2 and 3 sound a little “unprofessional”, so let’s rename this to something a little more suitable. When working with individuals, we tend to give them names to identify each bird from one another. So, for simplicity reason, bird 1 will be named Temi, bird 2 as Willis, and bird 3 as Lumpy (still unprofessional probably, but it’s cute). So let’s move to the Custom section in the editor.
This is the coolest part of the graph editor in my opinion. You can simply avoid coding and just tell Julius what you want to change on the graph using plain language. Let’s ask it to change the legend names to what I suggested:
It did it! But my title came back… so I’ll just have to readjust that. Now let’s change the colour of the heading points for fun:
This is important to note: when you want to change the colour of your dependent variable (mine is heading), it is recommended that you tell Julius the variable name in the prompt. You can see I specified Temi heading colour, which let’s Julius know that I’m specifically talking about Temi’s data points in the heading column.
I highly recommend trying out multiple different prompts for the customization section in the graph editor. This was only a small snippet of what it can do, but the possibilities are endless. Just remember when you do write a customization down that you direct Julius to what variable you are targeting and what you want to do with it. It likes short and simple directions, so try and describe it that way.
Happy graphing!
Keywords: AI statistical analysis, GPT , graph editor, polar plot, graph customization, data visualization
