I’ve spent a good portion of my day doing something, that while arguably productive, has nothing to do with my to-do list. And that’s at the expense of my to-do list. And now I’m writing about it, which is setting me back even more. And yet, I’m not stopping.
It all started yesterday when I sent out a one-question survey to my email list. Among other things, I run communications for a non-profit here in Portland, and our board had a question for our readership.
Side note, I’ve found a way to get a really great survey response rate with our email list. It’s a two-step process that involves sending one question first, and then following up with a few more. If there’s interest, I’ll write more about that some other time.
Anyway, back to the task at hand… procrastinating.
In a nutshell, our non-profit is looking to host a late-night fundraising event for tennis players, and we’ve been wondering whether people would even be interested in playing tennis at midnight. For those who responded yes, I followed up with questions about whether it should be kid-friendly, whether they would want to play singles, doubles, or both, whether it should be a short event from 10pm to 2am or an all-night affair, and whether Friday or Saturday would be the better option.
More emails started coming in, and there went my morning — not because I was busy corresponding, but because I’d put my math hat on and started coding the responses. 1 for yes, 0 for no.
Looking at the simplest results, it was becoming clear that an adult-only event was by far the preference, with 96% of respondents happy with that format. Doubles was outweighing singles, though quite a few would be willing to play both. Most people preferred a four-hour event over pulling an all nighter, and Friday seemed to be the winning evening.
I could’ve stopped there. After all, I had enough info to go to the board and make suggestions on how to plan this thing. But no. There was something nagging at me to keep going. The procrastination monster had gotten the best of me.
So I dusted off my old statistics program from back in the day and started plugging in those ones and zeros. Would an affinity for singles correlate with a preferred day of play? Would those who wanted to play both singles and doubles also be willing to stay all night long? Or would they prefer to play the shorter duration that, paradoxically, wouldn’t leave enough time for both categories?
Basically, what I was interested in was whether a little number crunching could prepare my board even better, and prevent conflicts by anticipating their probability ahead of time. Was there a story in the numbers that I could tap into?
Looking at the correlations (the dots here represent statistically significant findings, p < .05), there are some expected and some unexpected findings. First, it seems kind of obvious that someone that votes for a kid-friendly event will correlate negatively with a non-kid-friendly option. Same with Friday vs Saturday, though several people had no preference, which is why that dot isn’t a gleaming red hot fireball. Similarly, those who preferred to stay for the shorter duration correlated negatively with the longer duration. Makes sense. Where it got a bit more interesting was in the blue zones (positive correlations). First, people who wanted to play singles were also quite likely to play both events. This was not necessarily true for doubles. And then there’s the kids/day correlation, hmm. People who would prefer having no kids at the event also prefer to have it on a Friday night vs a Saturday. Now that one I can’t quite figure out. Maybe it’s because without kids, you can drink more, and when you drink more, you need TWO days to recover before work on Monday? Maybe it’s that it’s easier to find a babysitter for Friday nights than Saturday nights? Dunno… I’m curious though. Maybe the answer to that question will inform something important about our event planning. I’ll have to go digging. Maybe I’ll find that it’s just a coincidence though. After all, correlation isn’t the same as causation. Probably the most helpful correlation so far, though, was the one showing that those who wanted to play both singles and doubles were more likely to vote for an all-night event, vs a shorter one. Whew. I was afraid that if it turned out the other way, I’d have people happy they could play both categories, but upset that they had to stay all night to fit the matches in… or that they’d be happy not to have to stay all night, but frustrated that they didn’t get enough matches for their money. But generally speaking, this correlation suggests things should be pretty smooth. Well, smooth-ish. It’s not the strongest number, so we’ll want to be ready with a proactive solution. Maybe we’ll offer two types of tickets, the full all-nighter experience as well as a “lite” version. Responses are still coming in so I’m sure things will shift a bit between now and the event date. But what I love about this exercise is that it’s giving us reason to plan facets of the event based on something measured. Even if the data we're getting is based on self-report (which is known to have reliability issues), our team will have a means of planning with purpose—and iterate as new information comes in. We won’t just be planning an event the way we want it to go down. We’ll be organizing something that has been crafted in partnership with our audience. And what better way to bring a community together than to make everyone feel like they played a part in its creation? Ok, back to work.