AnchoringPosted: October 12, 2011
There’s this really interesting phenomenon called ‘anchoring’, where by introducing an irrelevant piece of information to a conversation you can influence people’s judgements of numbers, even where the person knows that the initial ‘anchor’ information is totally irrelevant. For example:
… an audience is first asked to write the last two digits of their social security number and consider whether they would pay this number of dollars for items whose value they did not know, such as wine, chocolate and computer equipment. They were then asked to bid for these items, with the result that the audience members with higher two-digit numbers would submit bids that were between 60 percent and 120 percent higher than those with the lower social security numbers, which had become their anchor. (source Wikipedia)
The same kind of effect occurs with the scales on graphs. No matter how smart and numerate the audience is, they will be influenced by the scale you use to present a picture, even though you know and they know that the data is the same stuff whatever scale it’s shown on.
If you show a big number with a zero-based scale, then it’s going to look pretty smooth and stable.
Nobody’s going to ask you for much explanation of those figures.
Show the same number with a cropped axis, and the picture looks a lot more eventful. Maybe eventful enough for an analytics fire drill.
If you start playing around with the time range …
… or constructed metrics …
… then the picture can change again, and start creating all sorts of different perceptions about whether numbers are high or low, noisy or smooth.
So what’s the moral of this story? There’s no simple lesson here, but these pictures make it very clear that there’s no such thing as a purely neutral presentation of data. Your presentation choices affect how the audience feels about the data, which affect their decisions based on the data. So tread with caution.