greglnelson
commented
6 years ago For Ballmer's Peak: What's the decision context? Should I drink during a test or not? When writing or not? I think this would be much more interesting for drinking habits in general rather than during a task (you will have the correlation/causation issue there). Why alcohol? You could perhaps include other substances, like food. Doing a 2 person pilot study was interesting. What is the prior research in this area (alcohol is very well studied, thousands of studies at least).
For Mac vs. PC, the assumption that the way the computer is used doesn't affect is clearly false.
Overall great job in thinking about data gathering issues (validity, survey bias).
The diversity of your pilot studies is incredible. A few things:
Ballmer's Peak: There is a PhD student collecting data on people's task completion time on mobile phones and their level of intoxication. It may be better to conduct analysis on a preexisting data source. Let me know if you want me to reach out.
Mac vs PC: A causes B is a hard claim to make, especially for something like mac vs PC (which could very well be a proxy variable for socioeconomic status/investment in education). You'll want to make sure the decision context and analysis is very clear if you want to move forward on this.
Some ways to further focus your thinking:
(no further action is required for this "issue" and you can close it after you've reviewed it)