A few days ago I announced a small quality control project where I would measure twenty “1 kg” Albert Heijn (AH) potato bags to assess whether or not AH is systematically underfilling them, as some social media posts had implied. I have now completed the measurements and I know what the data show. Before revealing the results, however, I would like to ask you a simple question: what do you expect the results to look like? I believe this is a vital question that people often skip.
Let’s say you are in a meeting and somebody is about to present some data relevant to your organization. Just before presenting the key results, the presenter does well to stop and ask the audience “what do you expect the results to look like?” This question:
- …forces the audience to really think about the problem, instead of zoning out or not-so-secretly checking their email;
- …attends the audience to the difficulty of the problem, and makes them aware of how little they actually know;
- …prevents hindsight bias — the distortion that once you are told about a result, it will tend to appear trivial or natural (e.g., “of course Trump was going to win the election; everybody with a bit of common sense could have seen that coming”);
- …sets the stage for a useful learning experience;
- …increases attention and makes the big reveal more exciting;
- …makes the data more memorable.
When my former advisor presented his academic work during conferences, he became increasingly frustrated with the fact that the audience believed the results were obvious. So instead of bombarding the audience with data, which was his usual presentation style, he started to pause before revealing the data and ask “what do you expect the results to look like?” This simple question made the audience aware that they in fact had no clue. And when the advisor then presented the data, the audience had a much better appreciation of the fact that they had learned something new.
So if you really want to learn something about data, don’t just passively absorb the results, but step it up and become mentally involved. When you fail to take the time to think about your expectations, you are robbing yourself of a valuable learning opportunity.
So do the 1kg Albert Heijn Potato Bags Really Contain 1kg of Potatoes?
What do you expect the results to look like? On average, do you think that AH overfills the bags, underfills the bags, or fills them according to the specification? How much variability do you think there is from one bag to the next? And do you expect 20 bags to be enough to draw any meaningful conclusions? I promise that all will be revealed in the next post! (in the mean time, please send me your potato recipes)
Authors
Eric-Jan Wagenmakers (CEO) and the JASP Services BV Team
