On island biology I am currently doing a research on the difference in seed fall the between island. For this research I collected plants on the islands and dropping 5 seeds for eacht plant. This means that seeds are nested in plant and plant is nested in island. My data looks as follows:
| Drop time | Seed ID | PlantID | Distance to coast(cm) | plant height(cm) | Island |
|---|---|---|---|---|---|
| 5 | 1 | 1 | 20 | 100 | Vis |
| 8 | 2 | 1 | 100 | 20 | Bishevo |
I used the following code: lm10 <- lmer(Drop.time~IslandDistance.to.coastplant.height+(1+plant.height+Distance.to.coast|Seed.ID:PlantID))+((1+plant.height+Distance.to.coast.|PlantID:Island))
Unfortenatly I get the following error messages: ixed-effect model matrix is rank deficient so dropping 4 columns / coefficients boundary (singular) fit: see help('isSingular') Error in lmer(Gemiddelde.valtijd ~ Eiland * Distance.to.the.Coast..meter. * : non-numeric argument to binary operator In addition: Warning messages: 1: Some predictor variables are on very different scales: consider rescaling 2: In Plant:Eiland : numerical expression has 120 elements: only the first used 3: In Plant:Eiland : numerical expression has 120 elements: only the first used
Now I am scared that the statistical analysis is not usable. Can I ignore this error.
I already wrote some discusion on this data but I do not know how correct using this data is.
factorvariables. Then fix the parentheses in your code. $\endgroup$