3 ms·
There are more reasons to doubt our results, so we lower our standards of evidence?
by yummyfajitas 16y ago
There are more reasons to doubt our results, so we lower our standards of evidence?
- sesqu 16y agoSounds reasonable. When some people find that their model doesn't quite fit, they make a more accurate model. Others make a less specific model. It's the difference between model parametrization and model selection. So when we get a dubious result, we can either say "no result" or "possible result". The choice tends to depend on how the finding affects future research. Biology is more exploratory than confirmatory, so they go that way.