All data in of itself is useless. A sample pool of 10,000 volunteers is pretty good in the realm of statistics.
All data in of itself is useless. A sample pool of 10,000 volunteers is pretty good in the realm of statistics.
If a study is self-reporting, it’s an observational study which can only establish correlation, meaning the study can only say, “there might be something here that warrants a further research”.
A clinical trial is needed to establish causation.
So while 10k sample size reduces the error bars, it only increases the confidence that there might be something here worth doing a more rigorous study later.
No it’s not. No sample size can make up for other biases.
Data’s usefulness stems first from its many defining factors, which validates it, and then opens doors to using that data to explore insights.
Now we are getting somewhere.
We spend so much of our life determining what is useful in one sense of the word, and rarely stop to think about the usefulness of what we do in the philosophical sense.
The point isn't the size. The point is there are too many confounding factors. It could be that people who work longer hours do so because they are high on trait conscientiousness, which also causes them to report low productivity
It could even be because they are conscientious that they actually work slightly slower, and/or it could be that people who are less conscientious just don't care as mcuh about producing what was expected.
To paraphrase the late Hitch: productivity metrics are not great. Neither is the word "connection" in science-adjacent reporting.