Some Thoughts on Gradualism / Stepping Stone Theory

A place to discuss topics affecting quizbowlers as a community rather than quizbowl as a game.
Post Reply
User avatar
username_crisis_averted
Lulu
Posts: 51
Joined: Mon Feb 06, 2017 11:12 pm

Some Thoughts on Gradualism / Stepping Stone Theory

Post by username_crisis_averted »

This post is a sub-post of my previous post: Why College Recruiting Fails. It adds a lot of context to this post, so you might want to read it first if you haven't already.

I got a noticeable amount of (somewhat expected) pushback on my last post regarding gradualism: using non-quizbowl or high school novice* questions to hook people into the activity. I will contrast this with essentialism: making an effort to preserve the essential quizbowl-like attributes of the activity when introducing people to it. In this post, I hope to answer three questions:
  1. Why might we prefer essentialism over gradualism?
  2. What should we consider when selecting "demo question(s)" in an essentialist approach?
  3. What have my personal experiences been with using essentialism in the wild?
* = I only take issue with high school novice questions; standard high school and collegiate novice questions are fine.


In Defense of Essentialism

Before I talk about my issues with gradualism, I want to make the case that essentialism is not as bad as it seems. The conventional view is that essentialism is just too hard and/or overwhelming for newcomers. Unedited quizbowl questions are just too intense! In the same way that you wouldn't start someone off on the middle level of a video game, you also wouldn't start someone off on conventional college quizbowl questions.

I have two main objections to this. The first is that (at least in my experience) the depth of quizbowl questions (which I'll define as a measure of how long and substantive they are) is not usually the problem for newcomers. I've seen plenty of quizbowl newcomers power the first question they were ever read. Most people who would thrive in a quizbowl club (whether it takes them a long time to warm up to it or not) will be able to answer some relatively difficult quizbowl question if given proper direction (explaining pyramidality, putting extra emphasis on the pronoun, clarifying that you should buzz ASAP, etc.). If you read someone a tossup on something they’re super into, they’ll probably do very well on it. At the very least, they’ll have the information needed to do well.

The real hard part of quizbowl is the breadth. We aren’t surprised that tossups on Avatar: the Last Airbender get powered in most rooms, even ones full of novices. But how many people have deep knowledge about Benedict Arnold? How many people know a similar amount about SN2 reactions, action potentials, and the Standard Model? Even for so-called specialists, quizbowl is a generalist’s game. The real test is having the persistence to sit through things you don’t know over and over until you know them (or get used to it). That is what sets a quizbowl player apart, and that is also what makes recruitment so difficult. In order to maximize recruitment chances, you have to figure out what question(s) would be in a newcomer's wheelhouse.

This brings me to my next point: quizbowl difficulty does not always translate to accessibility for a non-quizbowl audience. A common misconception about quizbowl is that it tries to test the things in a subject that people are the most likely to know about. This is a goal of quizbowl, but it is a secondary one. The primary goal of quizbowl is to test the things that are foundational to human knowledge and culture. High school novice sets have tossed up The Call of the Wild, Gustav Holst's The Planets, and Alfred Hitchcock. These are unquestionably foundational topics, but they'd probably play worse than a lot of topics that never show up in academic quizbowl, like Gone Girl, Taylor Swift's 1989, and (to quote from the Discord) Knives Out. This is not a criticism of quizbowl; the focus on "canonicity" is a feature, not a bug! I just want to point out that accessibility can't be boiled down to difficulty alone.

Question depth can be adjusted for accessibility purposes, and you should keep that in mind when demoing questions. (Don't start people out on an ACF Regionals eight-liner.) However, it is just one of the many levers you can pull to make quizbowl more appealing to a new audience. There are other levers that can be much more effective. For example, I would argue that question topic affects the appeal of demo questions significantly more.

And depth is also the riskiest lever to mess with. You don't want to attract a bunch of people who are just there for the trivia*, only for them to leave once they realize that trivia bears very little resemblance to the activities of the club. Sure, maybe you could convince them to stick around. But wouldn't you rather have a starting cohort that already knows what they're getting into?
* = This principle goes for using excessively easy questions, too. The default should be to faithfully showcase the activity in question, and any deviation should be treated with caution.


Gradualism Has Costs, Too

And this is my first problem with gradualism: it assumes that using trivia for recruiting is cost-free. The reasoning goes something like: even if you attract some people who only like trivia, you should still be successful in attracting everyone that you want to recruit. I would argue that attracting people who don't know what they're getting into is already bad. If a club thins out dramatically after the first couple meetings, that's not going to make the other people want to stick around. Better to start small and grow than to start big and shrink.

But even if you don't agree with that, there's another thing to consider: the quizbowl audience is not a proper subset of the trivia audience.

At minimum, I can conclude this because of myself. I don't dislike trivia (especially after having played quizbowl), but I would not consider myself a "trivia person". I'm much worse at trivia than quizbowl, and I find a lot of trivia frustrating and not that fun. If I hadn't done quizbowl in high school, the pitch of "like trivia, sort of" followed by trivia questions would not have worked on me in college. With so many other things to do, it just wouldn't have seemed compelling.

I first saw quizbowl my freshman year of high school. At the first practice, the veteran players played a demo match and encouraged us to hop in and participate. The questions used were normal high school quizbowl questions: not novice and not trivia. Seeing these full questions didn't turn me off from the activity, nor did it scare off any of my peers. In fact, it made me significantly more likely to join. I was inspired, not intimidated, by the depth of the questions, and I could immediately see why people might want to spend a lot of time doing this. I don't know how universal my experience is, but I would at least want my recruiting strategy to be able to capture people like me.

Even though the trivia audience is a promising base from which to recruit players, we shouldn't overlook other places to get potential geeks. One advantage of a more specialized activity like History Bowl or Science Bowl is that it has a more natural recruitment base: the people who are already "geeks" in the relevant subject. Intuitively, this feels like a more promising recruitment base than trivia fans. I'd be interested to see what would happen if a quizbowl club devised a "Literature Bowl" or "Philosophy Bowl" event for recruitment purposes.

Finally, using trivia or HS novice questions for recruitment comes with the risk of scaring away "slow geeks". Halle Friedman captures that idea pretty well in this forum comment. TL;DR if your questions are excessively short and/or "easy", it increases the risk of a small number of fast people dominating the questions. Allowing players to dominate increases the risk of players feeling like quizbowl is "not for people like me". Remember, non-quizbowl players can dominate, too! While you obviously want to separate out experienced players from newer ones, it's important to recognize that you can still have naturally skillful players in your audience.


Selecting the Right Demo Questions

Despite being pro-essentialism, I still think gradualism tries to address important concerns. We can and should adapt normal quizbowl packets to make them more fun and accessible for recruiting purposes. This is easiest when you're showing quizbowl to someone you know well; for example, if they're really into TV, you'd probably get the most positive reception if you read them TV tossups. You probably also know what kinds of TV they like and don't like, allowing you to further focus in on their knowledge base. But it's impossible to select similarly targeted questions when you are demoing quizbowl for a larger audience. Here is a list of things that I consider important for "demo questions" that may or may not be desirable in a normal tournament packet:
  • High Power Probability (HPP)
  • The Marie Kondo Test
  • Wide Coverage
  • Sentence Clarity
This list is primarily motivated by the geeks / MOPs analogy. Once your potential geeks are in the room*, you want to impress them! You don't just want your geeks getting questions. You want your geeks powering questions. The feeling that we all love as quizbowl players is the high of a sick buzz. If a demo can't give that to our potential recruits, then it will be much less compelling. Because of this, good demo questions should prioritize the chance of a power (HPP) over the chance of a conversion. You'd rather have one tossup go dead and one get powered than have two unmemorable last-line buzzes.
* = This should go without saying, but please separate out the players who have already played quizbowl before in any demo. While I don't think deep questions are inherently alienating to potential geeks, this changes when you allow former high school players to dominate.

The first thing you can do for HPP is make sure that the clues in power aren't overly hard. In a competition, you want hard first lines to differentiate competitors. But in a demo question, more "middle clues" is better: you want to preserve the norm of buzzing before the question is over but avoid overwhelming your audience with too many wholly unfamiliar clues. You don't want your power clues to be so easy that fast players dominate, but hovering around the same gettable difficulty for a while is ideal.

Another thing you can do for HPP is ask about topics with more fans. Let's consider the books Song of Achilles and Cry, the Beloved Country. Song of Achilles has never been tossed up below CO difficulty while Cry, the Beloved Country is among the most asked novels at the high school level. Despite this, I would expect a Song of Achilles tossup to play significantly better in a demo. Song of Achilles is huge on BookTok and has lots of devoted fans around college age. In comparison, Cry, the Beloved Country is a relatively common book assigned in high school curricula, but it doesn't have the same enthusiastic fanbase that Song of Achilles has. While I would expect Song of Achilles to be converted slightly more than Cry, the Beloved Country, the real disparity I would expect to see is in the relative power rate. This is because of the fans component: there are so many people out there who really know Song of Achilles and remember enough of the book to power a question it. Chances are high that one of them is in your audience!

This is where the "Marie Kondo test" for quizbowl questions comes in: does the question "spark joy"? Quizbowl is at its best when it rewards people for knowing something that's meaningful to them. So make sure that you're asking about things that people have a genuine interest in, in addition to things that people kind of remember from high school. Like, my gut says that it's preferable to ask about something trendy like CRISPR or climate change rather than, say, organelles; the first two topics are (subjectively) more likely to have "fans". Pick topics that you think people will be passionate about!

Finally, you can promote HPP with wider question coverage. I'd sacrifice one of the four literature questions in a standard distribution if it means having one question on film rather than none at all. More pop culture in a packet is fine as a hook; the main essential ingredient is question diversity. If there is a topic that is particularly liked by students at your university (like computer science at UW), make sure to include it! There should be something for everyone amongst your potential geeks.

I don't have as much to say about question structure, but some questions do get tied up in the weeds; I'd do a quick test of any demo question to make sure that the intention is clear to a general audience. One type of question that I would avoid is the common link: demoing these has only led to confusion in my experience, and the reward just isn't really worth it. You also want to have simpler sentences— if there's a really long compound sentence, you might want to make a quick edit for the sake of clarity. Basically, just make sure your questions aren't a struggle to interpret!

Of course, none of this can replace a good explanation. And to be honest, most of this is probably overkill. With a good presentation, I'd argue that a maximally essentialist "just read ACF Fall" approach would capture most geeks. But if we can show people quizbowl in a way that's more fun and inclusive, we should aspire to do that. If so many people are concerned with this question, then it must have some impact at the margins. And I don't see much harm in striving for that goal. I've thought about whether it would be worth compiling a "recruitment set" with each packet designed for a common target audience (general first-years, bookworms, people who are into programming, etc.); it would be cool to have something like that on hand rather than having to manually sift through tossups. But that's just an idea; I'm interested in hearing other people's thoughts!


Demoing Quizbowl in the Wild

Anyway, all of this is just theory. You shouldn't just take my word for it! Like I said in my last post, I feel strongly that more people in quizbowl should be in the habit of introducing it to others. So I'll conclude this post with some instances of demoing quizbowl "in the wild", just to show that these ideas have been partially tested.

At the beginning of this year, we had a recruitment event at this university event for first-years called Dawg Daze. Dawg Daze is essentially a more interactive version of the traditional student activity fair: it gives student clubs an opportunity to advertise and put on an event for a first-year audience. We called our event "Quizbowl: A New Kind of Trivia" (emphasizing that the activity would be different from the trivia that people are accustomed to), and specified that we would be explaining the rules of quizbowl and then playing some mock games. The event was held in a mid-size lecture hall, and we had our buzzers out for the players in the front row to use. I believe around 60 people showed up.

To support this event, we compiled a quizbowl packet that was loosely tailored to our projected audience. For instance, I made sure to include a computer science question since UW has a large population of computer science students. (Incidentally, a CS first year with no quizbowl experience powered that tossup and ended up joining the club as a result!) We also included significantly more pop culture, with a focus on representing every subcategory. In the end, 8 out of the 20 tossups were powered and only 1 of them went to the last line. We had 11 different people answering tossups, and to my knowledge only 2 of them had played quizbowl before. I don't think that having real quizbowl questions harmed our recruitment, and a noticeable number of the people who buzzed ended up joining the club. We are hoping to continue this tradition in future years!

I have also shown quizbowl to a significant number of people in an informal context. Here is a quick recap of some of the more memorable instances:
  • Friday: I read someone in my algebra class the ACF Winter tossup on the Book of Judges. I chose this tossup because she was pretty involved with her local church, and she said she was good at Bible stuff. She got the question on the second line.
  • A Week Ago: I was talking to a barista and showed her the tossup on James Joyce from CALISTO 2. I chose this tossup because she said she was an English major, and Joyce seems like an author that would be disproportionately read by people who are into modern literature. There was a pretty famous quote in power, and I also figured that Joyce has a greater number of popular works than the average literary author. She got it on the second-to-last line. After learning more about the club, she said, "I would join if I was still enrolled (at UW)".
  • Two Weeks Ago: I was talking to a guy who sits next to me in math class and showed him the tossup on boundedness from 2021 ARCADIA. I chose this tossup because it had some good real analysis content in power, which I knew was a class we had both taken. He got it on the second-to-last line and was annoyed that he didn't buzz on the "Archimedian property" clue in power. Although the activity didn't resonate with him a whole lot, he said he understood the appeal and would be interested in showing up to a meeting some time.
  • A Year Ago: I decided to show my probability professor some quizbowl questions during office hours. The first question I read was a high school question on standard deviation (which I chose because I was in probability class). He was kind of confused and ended up getting the question last line. After he got the answer, he was annoyed by the triviality of the clues. I then read him the tossup on the Central Limit Theorem from 2016 Penn Bowl. He powered it on the second line, muttering that the question was silly because Donsker's theorem is so obscure, or something. In general, he did not seem amused.
  • In High School: I often came to my history teacher's classroom after school to discuss the American history questions I was writing. He was definitely a "slow geek": it took him longer than most to parse questions and arrive at an answer. (One of the first questions I read him was the 2018 ACF Regionals question on Elizabeth Cady Stanton, and if I remember correctly he spent around five minutes pondering the second line.) But he really did enjoy hearing about question writing, and he offered lots of helpful suggestions. In the end he thought the activity was pretty interesting, and he said he would be interesting in taking us to a competition if needed. I do think the writing side of quizbowl can be pitched independently of the playing side— while you can't turn a non-player into a writer straight away, you can still "hook" people with the art of it, if that makes sense. That's an area that our club is trying to experiment in going forward.
Now, none of these instances actually led to anyone joining quizbowl. But I really do believe that 1) if these people were potential geeks, a similar pitch would have been effective in getting them to join and 2) this was a good way to spread awareness of the activity. But I'd be interested to hear more about other people's experiences with showing other people quizbowl, especially if their experiences have been different! Have you tried using essentialism? Gradualism? Did it go especially well? Especially poorly? I'd be interested to see if there's anything that I've overlooked!


Key Takeaways
  • In most cases, the breadth of quizbowl (not the depth) is what makes it intimidating. In other words, question topic matters more than difficulty.
  • Novice questions will not necessarily be the best questions for your audience. It's more important to tailor the questions to the audience that you have in mind.
  • Using trivia or high school novice questions does carry a risk of alienating certain audience members. More importantly, it may fail to inspire.
  • You don't want to recruit too many people who don't know what they're getting into; early exits from the club can be bad for morale.
  • Make sure to consider other avenues of recruitment besides trivia fans and former quizbowl players, particularly "geeks" for a certain quizbowl subcategory. This goes double for colleges that don't have a preexisting high school quizbowl pool to draw from!
  • The ideal recruitment questions should be designed to get a diverse range of newcomers to experience the joy of a sick buzz. Because of this, you should prioritize coverage of the widest possible range of topics, as well as power count (as opposed to raw conversion).
  • Test your intuitions! Find out what people are receptive to in the wild; they will definitely surprise you.
Thanks for reading, and I should be back to my regular posting schedule next week!
Kevin Kodama
University of Washington '23
FrancesWalsh
Kimahri
Posts: 1
Joined: Fri Mar 17, 2023 7:14 am

Re: Some Thoughts on Gradualism / Stepping Stone Theory

Post by FrancesWalsh »

The preference for essentialism over gradualism in the context of quizbowl-like activities may have several reasons. First, essentialism preserves the uniqueness and value of the activity. If we use questions unrelated to quizzes or for freshmen in high school to draw people into the activity, we risk changing the essence of the activity and losing its value and uniqueness. Second, essentialism helps maintain quality standards. If we use simpler questions to attract new participants, we risk lowering the level of difficulty and quality of the questions, which may alienate more experienced participants.

User reminded to add signature -mgmt
Post Reply