Data and its impact on behaviour
In his keynote at the Festival of Work, the Financial Times’ Undercover Economist Tim Hartford took a look at how data and behaviour can transform work.
He used running as the metaphor for his talk, recounting his journey from Parkrun to the London Marathon and the role data from his Garmin watch had to play in it. Ultimately, the data helped in many ways – the goals, reminders and social accountability. But it was the human side of the story that was more important – why he started running (because of his friends) and how he managed to finish the marathon despite being in a lot of pain, helping and being helped by another runner.
In his talk Harford picked up on three concepts/biases that shape behaviour based on data.
- Value capture
- Quantification fixation
- Goodhart’s Law
Value capture
Value capture happens when the value in something is captured by a simpler proxy. For example, the value in learning might be curiosity, critical thinking and understanding but the capture is test scores. Over time, people start treating the test scores as if they are learning.
The problem is that the real value in something is in danger of being captured by something simpler and easier to measure. And then that becomes the measure of success.
Quantification fixation
A cognitive bias where people focus too much on the metric even though it might not fully capture what matters. An example of this might be the amount of content published vs the quality or impact.
This feeds into the idea of what gets measured gets managed but the problem with this is that not everything can be easily measured.
Goodhart’s Law
This describes the idea that once a measure becomes a target, it ceases to be a good measure.
A metric can work well until people know they are being judged by it (it becomes a target). At this point people change their behaviour to improve the metric. That metric then becomes distorted and no longer reflects the thing it was intended to measure.
An example would be call centre call times. If a target time is set for all calls, call centre workers adjust their behaviours and approach to hit the target. However, that might negatively affect customer satisfaction so the behaviours might not be the ones your organisation is ultimately looking for. Hartford provided numerous examples of this, from the Vietnam war to ambulance waiting times.
So what are organisations to do? The L&D sector continues to focus its attention on data and impact but talks less about value. However, data can drive the wrong behaviours and impact. Value can be an antidote to value capture as it is a way to look more broadly at what matters. Measuring qualitative and quantitative data can help here (if you act on it, that is).
When it comes to data, Harford says:
- Use data – but with transparency, purpose and protections
- Use fewer, simpler metrics
- Acknowledge what can’t be measured
- Remember people matter more than numbers.
