I have some values in a bunch of values stem from more similar environments with respect to the other values in that bunch; they are nested.
Nevertheless, I measure them all at the same time and throw them together and plot them in a boxplot. So the values are all treated equally.
I don't know if this is statistically appropropriate.
Is there a way to account for nesting, i.e. weighing the nested values by something? I'm thinking about using values from a nested anova R model.
Is this necessary? Can I just compare the values since the values are descriptive of the status quo (or something like that)?
Thanks for an answer!
Hi Vera,
One method is the barNest function in the plotrix package in the R system. This allows you to display hierarchic nesting. If you would like to try it, see the "Titantic" examples on the help page. If you don't use R, I think there may be a similar function in Excel, but I rarely use that.
Jim
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA