Currently the mean and median of precip (range probably 0 to something like 1000) are rounded to the nearest 10 mm and growing degree days to the nearest 100 (range in the thousands) . The SD seems to be rounded to the nearest integer. Variables with a smaller range lke PMDI and Z score standardized data show 3 decimals,. Variable means very close to zero show values like 6.7e-17.
The larger range variables should never be rounded more than to the nearest integer. If how this is treated is a function of the range or something like that, the nearest integer is probably best. I can also live with up to 3 decimals.
For Z score standardized data and PMDI or things with a similar range I think 2 decimals is good and 3, as it is now, is fine.
For the essentially 0 values it would be best if they were shown like the others for that variable display (e.g. 2 or 3 decimals). If something is zero to three decimals it is fine to say 0.000 even if it is not exactly zero.
The UI seems to be applying some complicated rule that I can't figure out offhand. I think is is very important that we stop the extreme rounding of #1. The other two things are not terribly important. It would be perfectly good to display 2 or 3 decimals for everything.
Currently the mean and median of precip (range probably 0 to something like 1000) are rounded to the nearest 10 mm and growing degree days to the nearest 100 (range in the thousands) . The SD seems to be rounded to the nearest integer. Variables with a smaller range lke PMDI and Z score standardized data show 3 decimals,. Variable means very close to zero show values like 6.7e-17.
The UI seems to be applying some complicated rule that I can't figure out offhand. I think is is very important that we stop the extreme rounding of #1. The other two things are not terribly important. It would be perfectly good to display 2 or 3 decimals for everything.