Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It only takes a minute to sign up.
Sign up to join this community
I'm trying to work out what test to use to see whether a factor (presence vs absence) affects a frequency distribution with distance as a predictor.
Just wondering if anyone knows what the best test would be for this
My hypothesis is that the presence of a factor - as an example lets say the presence of a tree canopy will influence the frequency distribution - for example below
I would think about selecting a limited number of most interesting distances, randomly(!) selecting several measurement points and sources of the signal (with and without trees between them), and then running a two-way ANOVA (where one category is the presence of trees, and the other is the distance, which is a discrete variable).
If the relationship between the signal and distance had been linear in all cases, you'd have had an option of using ANCOVA as well, leaving the distance not discrete.
Seems that the problem could be reformulated as testing the equivalence of distance distributions in two populations (with and without canopy). In that case, I'd use two-sample goodness-of-fit test (e.g. two-sample version of Anderson–Darling).
