Lies, lies and statistics in the arts sector

Occasionally I turn into a data nerd: researching away, looking through published data, cross checking tables and graphs, that kind of fascinating stuff. Sometimes I stumble across something that knocks me sideways in its audacity. Now is one of those times and I just had to share it, whether you like pie graphs or not.

Today I’ve been finishing our debrief for the May 2017 Anywhere Theatre Festival. Presenting the data is important, even for a little organisation like us. We want the figures to reflect where we are, not where we think we should be. Part of the debrief is comparing our figures to other Queensland based organisations to see if we doing well and badly.

I went to the wrap document for one festival – who don’t traditionally have to publish figures due to their company status – and found this set of pie graphs from their 2015 festival. I would have used the 2016 document but it lacks any data beyond headline numbers and a more image heavy layout.

Have a look.

Have you noticed anything? No? Look again. The pie graph for VISITOR ORIGINS looks a bit over half. Hang on – why does the percentage say either 71% for free events or 78% for ticketed and yet the graph look closer to 49-55%?

Now, look at the AGE graphs: 35-44 slice is bigger than the percentage and there is a whole load of weird stuff going on.

Moving to the right and the NEW VS RETURNING graph: 82% returning for free events yet the graph would indicate about 74%.

So, what do we have here? Graphing incompetence from an organisation that has enough money to get this kind of stuff right? Surely not. Excel and Numbers create graphs automatically based on the figures you put in, so I can’t see how this can be put down to a mistake. This was done intentionally, and that’s a worry.

I think what we have here is an acknowledgement of the figures this organisation wants to be bigger than they are. Wish fulfilment made public as it were.

They want it to seem like more people come from outside Brisbane, they desperately want their audience to be 35-44 and they want more new audience members coming to their festival.

Unfortunately, their statistics don’t match so they have edged the wedges of the graphs so that at a quick glance you think the figures are skewed in the direction they would prefer. Maybe they have agreed to KPIs that are not relevant or they have little power to control. I don’t know.

Arts organisations are meant to be about integrity, honesty and shining a light or at least a mirror back at society. If we feel a need to resort to this kind of thing of skewing of results to make ourselves look more like what the powers to be expect, then we as a sector will lose the integrity we need to remain relevant.

 

4 Replies to “Lies, lies and statistics in the arts sector”

  1. Good eyes Paul. Some of the pieces-of-pie bear no resemblance to the figures. However, which is correct, the number, or the piece of pie?? Are you able to confirm the validity of either?

    1. Good point Stuart. I can’t be sure whether it is the pie graphs or the numbers inside them as there are no further detailed tables in the reference document… which is AT THIS LINK if you want to have a gander at the complete document.

  2. Knowing some of the people who were responsible for generating these specific graphs, I’d suggest that the likelihood of it being numbers overlaid on draft, pre-data graphs is far greater than some big conspiracy to influence the tiny audience of this document. The charts are all consistently and wildly inaccurate in their proportions.

    Also knowing how under-resourced their marketing department was in 2015, simple oversight becomes all the more likely.

    Good spot, but perhaps simple human error is the cause here.

    Devil’s advocate point: perhaps the pie charts are accurate, and the percentages are not!

    1. Absolutely on the last point, and I think as a couple of people pointed out on Facebook, this guide also should have been sighted and signed off by people above and they didn’t notice either… and I am picking it up now a year after publication. 😉

Leave a Reply

Your email address will not be published. Required fields are marked *