from my escort survey, self-reported age and estimated monthly earnings.
sample size kinda small, each bin is n=~40
9
5
134
Conversation
feel free to doublecheck all my work, as I'm learning as I go
raw data here: docs.google.com/spreadsheets/d
survey I used here:
5
3
40
Replying to
Column charts are not showing the legend values but the horizontal bar charts do. Pie charts can be difficult to interpret but once I oriented I could figure it out. But here’s a cool thing I the chart you first posted. The younger escorts have a higher variance in hourly rate/1
1
Oops my bad, it was the monthly income by BMI graph, but the point is the same — differences in variances across groups can tell a story that might be worth following up, but most analysis methods don’t focus on understanding differences in variability /2
1
What might cause the variances for low BMI to be higher than for high BMI? What else is going on? When there is high variance then there is more opportunity for other variables to “covary” which is where follow up multivariate analysis would generate additional hypotheses/fini
1
Replying to
I'm not sure I understand what you're saying, it's possible I haven't learned about the concept yet.
my intuition is that it's hard to draw strong conclusions from the differences in CI due to low sample size, but I'm open to learning if that's wrong
Replying to
You intuition is directionally correct my guess is the sample size in this case is adequate, and you often have large Ns which is why I bring up the opportunity. Also because you are interested on complex analysis which is where thinking about how to explain large variances /1
1