Discussion about this post

User's avatar
Zephyr Bonilla's avatar

Why stop at n=1000?

More sample size is always better and in particular n=1000 is probably good enough for getting the sample mean for all your questions but for subgroup analysis (eg, how do NSR vs OSR players feel about x, y, and z) it probably isn't enough.

Expand full comment
xaosseed's avatar

This is a great idea - I suggest sharing to socials - twitter, reddit and facebook are deep pools to try and tap. If you can put some of those out there, happy to boost it.

Expand full comment
5 more comments...

No posts