Hello, I think that you you can do the anova as usual, but you have to use the part of the missing values if you are going to use the spss for analysis. otherwise you can consider the location of the second year as replicate to the same location of the first year
Because your plan isn't complete (you have missing values for the 2nd year 2nd location), I would advise you to combine your 2 variables into one : you will have one variable with 3 modalities (location1_year1, location1_year2, location2_year1). Then you apply a one-way anova with this new variable
I would use the data for the first year to first test for any differences between the two locations. If there are no significant differences present, then only would I ignore any information on location, and then use the data for the two years. Then I would use "year" as a blocking factor to make sure there is no time effect.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Modeling
- Analyses of Variance