Good Constructs
Representing Reality: What Makes Good Data
•
6m 36s
Choosing a good data construct is the most important part of getting good data. The construct is the thing we choose to measure. Hopefully, a construct is a good match for the thing we want to know. But sometimes it isn't. What makes a good construct?
Up Next in Representing Reality: What Makes Good Data
-
Accuracy
What's the difference between accuracy and precision? They are not the same, and the difference is important. "Accurate" data gives an impression that closely matches the real thing it is measuring. "Precise" data is measured in a way that gives highly specific measures. For example, if you measu...
-
Representativeness
How well does your data match the reality of the the thing you're trying to measure? "Representative" data can be incredibly powerful, because it means your sample is a good match to the reality you are trying to understand. That allows you to analyze the data with a lot more confidence. Ask ques...
-
A Source for Comparison
We often prefer to have consistency in our data, but data without variation in crucial measures doesn't tell us anything. For example, if everyone in your office gets the same performance score (4 out of 5), the score will have no value in terms of how it explains actual work performance. In orde...
