Changing bad data to other bad data is bad. Is there a case against changing bad data to accurate data? Obviously you need to start over on the experiment, which I assume to be obvious.
You changed the argument. If you can't tell whether a tweak improves or worsens the data, obviously there is no reason to tweak. The case in question is when experts agree on a specific tweak improving the data.
-
-
Problem here is collection of historic data is in past. One can make guesses at systemic inaccuracies in the past data but most of those changes are untestable- since the data is in the past. Historic instruments can be retested in the present & prove they have bias now.
-
Data in question collected by different humans w/differing behaviors. Pretending these humans made consistent errors is unprovable. Changing data other than recalculating Fahrenheit to Celsius is unscientific. Using data beyond significant digits is scientifically unacceptable
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.