[Events] getting people to fill out exit surveys

[Greg Wilson]

On 2014-02-18 10:01 PM, Shauna Gordon-McKeon wrote:
> Now that you mention it, I do remember the "what went right/what went 
> wrong" part of the day!  And I've seen the blog posts summarizing that 
> feedback for individual events.  Do you find that feedback adequate 
> when trying to get a sense of how changes to curriculum design, event 
> structure, etc are impacting the event?
>
It works really well for us - the trick is that no one is allowed to 
repeat a previous point, which means people really have to dig and divulge.
> Your post-survey response rates make me pretty pessimistic.  If such a 
> competent, effective group as Software Carpentry can only get a 10% 
> response rate, what hope is there for us?  ;)
>
*shrug* We gotta play the cards we're dealt - if 10% is what we get, 
that's still better than 1% (or 0%).  My real question is whether the 
people who respond are representative of the general population...
> I think probably we need to think about what, precisely, we want to 
> know and if there are better ways to figure that out than a survey. 
>  Our needs have changed a bit as we move to organizing events that 
> neither Asheesh nor myself is physically present for. I'm not sure how 
> to measure how an event is going when I can't see attendees' faces or 
> see for myself how engaged they are.
Post-event interviews by Skype are the best way to find things out, esp. 
if you do them 60-90 days later: anything people are still using then, 
they'll almost certainly continue to use long-term. It's pretty 
labor-intensive, though, so the trick is to find a grad student who 
"wants" to do it to get a paper out - after a couple of false starts, we 
have someone at MSU who's doing the shovel work for us.  Small N, but 
lots of insight from each.

The other thing I'd (strongly) suggest is that you run training events 
for your instructors, so that they have a more informed idea of what to 
look for and why.  That'll give you much more confidence that their 
impressions are accurate and on the same scale.  We're doing a live 
three-day version of our instructor training in Toronto April 28-30 if 
you know anyone who'd like to take part...

Cheers,
Greg