This is the fourth episode in a new series on Data Fluency. You can also watch the first episode, second episode, or third episode.
So, now we know that a good HR tool will have three things:
- Reliability: consistency of measurement.
- Variation: it produces data with range.
- Validity: the range validly predicts something else in our world.
Let’s look at a few common HR tools and see how they stack up.
Competencies ask employers to rate their employees’ abilities in certain areas, like leadership skills or strategic thinking. If you look back on what we’ve said about the Idiosyncratic Rater Effect and how humans are unreliable raters of other people’s abilities, you see that they are clearly unreliable data. Humans are unreliable raters of other people, therefore competency models are completely unreliable and give us bad data.
Your manager’s rating of your performance is going to be unreliable, for exactly the same reason as competency models.
Employee Opinion Surveys
Employee Opinion Surveys ask you to rate your own experience and intentions – that means that many are reliable. That’s great!
Some departments will score higher than other departments, meaning that these surveys often do produce variation. That’s even better!
But do these scores matter? Do they have Criterion Related Validity? That’s where Employee Opinion Surveys often fall apart. On most employee engagement surveys, the teams that score higher on these questions don’t produce better outcomes in the future. So frankly, these surveys don’t matter and don’t produce good data.
Next week we’ll be talking about a tool that does work.
Check back next week for more information on how to become data fluent, and the impact of bad data on our businesses.