In my last post, I asked what UpGoer Five (along with its constrained-vocabulary ilk) is for. Survey says: UpGoer Five is a toy.
At least, that’s what my readership poll (data above) says. The sample size is small, and it’s an utterly unscientific poll (more about that later). But 63% of respondents believe UpGoer Five is a kind of stunt writing, something we might do for fun, but having no connection to actual science communication/outreach (“SciComm”). This is a bit of a relief, given my misgivings about it if seen as SciComm. But it’s also a bit of a puzzle, because if UpGoer Five is just a toy for fun, I’m not quite sure why it’s gotten so much prominent attention. I’m not trying to argue we shouldn’t have fun as scientists – we absolutely should! I’m just a bit worried that our “fun” might look like condescension to people outside science.
Another common response: 24% find UpGoer Five a useful practice exercise for SciComm. I guess I can see this, as long as we’re clear that it’s just one kind of exercise and not directly a model for actual SciComm. Actually, I bet the discussion about why it’s not a good model is just as useful as the exercise itself.
10 people (12%) voted that UpGoer Five is something else I hadn’t thought of, but only one (I think) explained what that something is. I hadn’t considered the possibility that UpGoer Five style SciComm might be useful for reaching second-language communities. I’m honestly not sure what I think about that.
Finally, one person – one person – voted for UpGoer Five being actual SciComm. This half-way surprised me, as it’s pretty clear for a variety of reasons that this view is not really that rare. But it only half-way surprised me, for reasons having to do with polling, and the fact that scientists, as a rule, do it wrong.
OK, so maybe that was a bit of a sweeping statement. But poll data are hard. My UpGoer Five poll, like UpGoer Five itself, is mostly a toy, and you shouldn’t look beyond its very broadest brush. For one thing, I asked the poll question after arguing against the proposition that UpGoer Five is good SciComm. If my argument was effective at all, it should have coloured the responses. (For that matter, if was ineffective, it probably had the same effect, with those fervently believing UpGoer Five is SciComm quitting in disgust before making it to the poll). I conducted what political operatives would call a “push poll”, priming audience response. My sin was flagrant; but it’s very, very easy to commit a similar but more subtle sin, either deliberately* or not.
Because SurveyMonkey and PollDaddy (among others) have made it free and very easy to gather poll data, I’m starting to see it crop up in papers (and I’m even involved in one such research project). It is incredibly easy to do this poorly, and scientists as a rule lack any training in how to avoid the pitfalls. In my experience, most scientists lack even the awareness of how their lack of training might matter. Please, if you’re considering such a thing, talk to a social scientist. A lot is known about how to do polls properly, but it’s known by them, not by us**.
And having wandered from UpGoer Five to cautions about polling methodology, I now return you to your regularly scheduled day.
© Stephen Heard (firstname.lastname@example.org) September 22, 2016
**^And not to wander off-topic or anything (squirrel!), but that sentence is a pretty good example of where the passive voice actually works better than active-voice alternatives. I among others shout myself hoarse advocating for active-voice writing, but the passive does have its place.