Story Data Story Three: Airrelle Cast & Fauna

Sorry for the delay, Imagined Audience!

The Blood Moon Mother is still wrapping up her stay at mines, I had my first post-editorial letter meeting with my editor, I’ve been overwhelmed by preventative doctor’s appointments regarding my chronic health issues (four), and the Just Like Heaven music festival was a nostalgic delight.

I also completed courses in Tableau and Power BI, started looking into my desired SQL platform (tending towards SQLite), solved a SQL Murder Mystery, and took a couple other steps forward in my data analysis journey.

I am putting the finale of the Black Names Analysis trilogy on hold for a second, though. Given that my original plan was to post one of these a week, I want this entry to be a quick little catchup! This Story of the Data of my Stories will cover the construction of a dashboard for my thesis, a middle-grade novel that necessitated a lot of research into California flora & fauna. Let the data storytelling begin!

Ask

Today’s question for analysis is: Who lives in the setting of my thesis novel?

You might say that sounds like a job for a moodboard, but I’ve never been femme enough for moodboards. No, instead I get my inspiration from the Pew Research Center or City Data or (my new favorite since returning home) CalMatters or so on. I’m also a megafan of the NYTimes Graphics section, as frustrated as I am with the NYT.

I like to poke around sites like Pew when I’m trying to figure out how normal my political views are, whether I truly count as middle class, or what kind of middle class I am.

Apparently, I am:

a progressive leftist,

who is low-income in the context of San Diego,

or for a highly-educated Black American women as a whole,

(oof, but this is why I’m working on a career change. I think I successfully fooled some of my middle-class colleagues into thinking I grew up like them.)

but cultured beyond my income level.

(Ah yes, my point of pride.)

All the same, I believe data like this is good for describing the real world, and therefore good for describing a fictional world.

So let’s see if we can make a nice little dashboard that gathers similar demographic details about our fictional witch village setting.

Prepare

My thesis novel takes place partially in the current-day real-world but partially among hidden witch societies that don’t use money. There is no real social class.

But, there are power disparities in terms of magic.

My advisors asked me some good and deep questions about how those disparities affected the setting, and that’s what influenced the shape of this spreadsheet. (Well, other than needing to track characters.)

Thinking through the nature of inequality in your fictional world is always an exercise worth exploring, however. What does hierarchy look like in your setting?
Is hierarchy even necessary? Is it (viewed as) a good thing?

Personally, I decided that a little hierarchy was fine so long as differences in societal power didn’t result in differences in societal treatment.

I was inspired by a study that asked Americans to imagine their ideal wealth distribution (“Building a Better America—One Wealth Quintile at a Time” by Michael Norton and Dan Ariely).

I remember reading vernacular breakdown of the article in undergrad and one of its figures stuck in my mind:

The ideal wealth distribution suggested by most Americans is close to a 30/25/20/15/10 split between income quintiles, so that’s what I aimed for. I then tweaked the numbers to tweak the setting.

The data I collected/created for this included biographical name, age, and gender but also witchhood status—ie, whether they are a full witch, in training, or without magic—and the number of animal patrons that they have attained.

Given that you need to obtain all of your patron animals to become a fully-confirmed witch, patrons are something like a measure of power. The four types of patrons needed to become a full witch are a Bird, a Bug, a Beast, and a Bloom (terminology still tentative). One of the major areas of inequality is therefore that some characters only form partnerships with one or two or three patrons.

The main character sheet tracking all of that looks a little like this:

This baby covers all 800+ members of the protagonist witch’s hidden village as well as a few key non-witch characters.

You’ll see that I have a primary key column titled “#” but also a random number column in case I want to play around with chance and be surprised by who pops up.

I did not sit down and create individual profiles for 800 characters, however. Only the first 43 names on the list are fully-fledged characters. I came up with some general traits and ideal demographic details, then combined all that with Black names from the databases I’ve confessed to being enamored with. I then semirandomly generated a population large enough to fill a village.

As you can see from the tabs at the bottom of the spreadsheet, I also have 3 sheets of data that I used to select patrons from: BIRDS, MAMMALS, and HERPS & FISH:

As you can see, these sheets list Order, Family, Common name, and so on—although I don’t like when animals are named after random guys and decided to lean on indigenous Kumeyaay wisdom or come up with names similar to the new bird names at times.

The sources for these sheets include:

All and all, this stage of research data collection was a lot of fun!

I learned about foraging and went on at least one foraging trip, as well as a couple camping trips and a research trip to the San Diego Botanic Garden that you may recognize from The Ultimatum: Queer Love.

You may notice that I don’t have a sheet for plants, but I did use Kumeyaay Ethnobotany: Shared Heritage of the Californias by Michael Wilken-Robertson a lot for deciding what plants fit the setting.

(I ended up barely using the herps and fish sheet. Yes, some witches have lizards and snakes and other herpetons as patrons, but I thought it was too cruel to give a witch the power to only speak to fish. That’s definitely happened to someone in this setting, but nobody that I chose to feature in the story).

Process

After preparing the data, the goal is to clean and organize it, changing it into a format that's easier to work with. In this case, I turned the main All Characters sheet into a bunch of calculations tables on a PIVOT TABLES sheet.

It looks like this:

As I worked on this last week, I was researching other data analytics dashboards and dashboard templates. I learned how to make all four kinds of sparklines and cool things, but it would be hard to demonstrate that skill on data like this that doesn’t track money or changes over time.

I also found out the hard way that Google Sheets’ map charts only go down to the nation level, so I can’t map locations like I planned.

All the same, here is my first attempt at a witch demographics dashboard:

I’m quite proud of it!

Everything pictured is either a label or a calculation—and some of the labels are even calculations in that they pull their titles from other sheets.

I was most excited about learning to layer pie charts and build population pyramids for the sake of this project, both of which were intuitive enough but things I’d never done before.

The funniest thing about this dashboard is that you can see the point where I ran out of energy in collecting data. I randomly assigned all of the witches Birds and Beasts, hence their very diverse population pie charts. But the Bug and Bloom populations are a bit more indefinitely, being 97% and 94% placeholders.

I was using so many placeholders that “idk” comes up as a calculated result for the most commonly listed family on the HERPS & FISH sheet:

See? I’ve gone and made the functions visible, but the result is going to remain “idk” because I’d lost steam before tagging all the families for that chart. “idk” is indeed the most common entry.

I instead gained steam in finishing the novel, which I did.

Analyze

So, what’s the answer to the question “Who lives in the setting of my thesis novel?”

Well, a lot of nonbinary witches, for one. So many that it was hard to do a normal male/female age pyramid. I assigned genders based on names and wound up with a population 40.62% female, 34.87% male, and 24.51% nonbinary or other. I did want the witch population to be more queer and femme than the real world, but did I perhaps overdo it? I can continue to think through how this affects the dynamics.

Another semi-intentional result is that the witch population is roughly stationary, not constrictive or expansive. I knew I wanted a pillar in terms of a population pyramid, and the other shapes get me worried about under- and overpopulation (tho overconsumption is more dangerous than overpopulation, and underpopulation is often just a mismatch between a society’s structure and its people). I read up on population pyramids to double-check this and National Geographic describes the three types of populations trends this way:

“The first is when there are both high fertility and high mortality rates among younger members. This type of population, known as “expansive,” creates a sharp triangle shape in the graph. “
“The second trend, known as “constrictive,” is when there is a lower mortality rate with the fertility rate remaining constant.”
The third trend is “stationary” which is a population with low mortality and low fertility rates. These graphs have a square or “pillar” shape rather than a pyramid one. These population pyramids represent a stable population that will not change significantly barring any sudden changes to fertility or mortality rates.”

Still, there is a bit of a pinch in the data that makes it look like a catastrophe happened about fifty years ago—What was it? Should I write about that?

There’s also not enough babies being born recently… Now that’s ominous.

Judging by the pie charts, the patron animals are a good and diverse blend with cats, owls, crow, and ravens as the intentionally-set most common familiars. Still, I wonder what patterns I would find if I looked by common name and not species name. There would definitely be more duplicates, but in a good way?

I mentioned that I ran out of steam in data collecting by the time I got to assigning bugs and blooms, so their pie charts are filled with placeholders and are less useful for visualizing how the village looks. I guess we will need to rely on our imaginations to know if there are ladybug swarms because of that one kid who befriended all the Coccinellidae.

This data raises more fun questions, like what would we find if we tracked witch jobs and plotted those against their power(s)? I do have spreadsheets based on the Bureau of Labor Statistics info, but I would need to decide/create witch-specific roles that match the setting. I didn’t really think of the characters as having jobs, however. I was thinking they could hunt in the morning, fish in the afternoon, herd cattle in the evening, and philosophize after dinner without needing to commit to being only a hunter, fisherman, cowhand, or philosopher. But there are witches who specialize in healing, cooking, gardening, performing, or so on. So hmmm…

The embedded result is here:

In case SquareSpace lets me down and won’t render it, I will share screenshots of the result:

Because this is my data, I can edit the demographics based on the needs of the story—making the population more or less of a pillar, naming all the bugs, decreasing the amount of enbies—and it’ll autopopulate the pivot tables and update the outcomes on the dashboard.

Of course, any good dashboard has gotta be able to auto-update. (And you’re a good dashboard, aren’t you, girl? Yes, you are!)

I actually changed a few things during the drafting of this post, even adding more figures: the Adult Witch Patron Attainment graph, a pie chart of village age range distribution, and a stacked bar chart showing how many witches are too young to have magic (PRECHANTED), are undergoing witchhood training (IN TRAINING), have some patrons (SEMICHANTED), have no patrons, (UNCHANTED), or are full witches (ENCHANTED).

With all that, I will call it a day. This wasn’t a quick little catchup at all, though!

Act

So the way I have acted on this analysis is to better avoid what I call Main Character Syndrome even in my main character. I want a lived-in world where anyone has the potential to be the main character, not a world that spotlights the protagonist like a tiny stage.

I’m a big fan of sonder, so I hate settings where things are only happening to the main character that should logically be happening to multiple characters. I don’t like when the main character is so much the center of attention that it creates logical plot holes. After all, every other character in a piece of fiction is presumably also experiencing their own training arcs, rags-to-riches stories, redemption arcs, and love stories.

I like to use a bit of randomness to help beef up the world, working in a few more geckos or neighbors or cats or owls. I think I need more scenes with other people’s patron animals, for one. I want the protagonist to exist in the context of her community and all that came before her.

That said, there’s not as much data here as I thought!
I guess the majority of my research is the qualitative stuff in my 50-page planning doc:

I’m not going to share much about this document here though—you would have to ask me directly.

If you’ve read this far, feel free to let me know!

Blog

Ask