Any calculation that involves lesser degree of dilution will be always be more accurate . For example , net plot yield converted into per hectare yield will be more accurate compared to per hectare yield , based on yield parameters of few plants in a given plot (since you dont consider the yield translation to all plants in a given plot ) , then converted into yield per hectare ... For sure , two different results will come up ..
Any calculation that involves lesser degree of dilution will be always be more accurate . For example , net plot yield converted into per hectare yield will be more accurate compared to per hectare yield , based on yield parameters of few plants in a given plot (since you dont consider the yield translation to all plants in a given plot ) , then converted into yield per hectare ... For sure , two different results will come up ..
If you have taken unbiased measurement of yield parameters from representative plant samples, use of these data for yield estimation will be equally good. But this practices is common in pot experiment. In field experiment net plot area should be considered for yield estimation.
If any one is unbiased, result will remain same in both cases.