A lot of research is wrong for one reason or another. If a scientist finds out that a paper of someone wrote contains errors the usual process is to write a new paper. Science progresses through new papers and everyone accepts that old papers often contain errors and are overtaken by new research.
The only thing that might be research misconduct is to to forge your research results by not changing the data you publish. Given how hard it is to persue people for massive fraud it’s however unlikely that anybody will argue in present day academia that leaving out one sample for understandable reasons will constitute fraud.
An erratum might say:
“We initually omitted one sample from our results because we believed it to be due to measurement error (or whatever rationalization you had at the time). That sample was X. Given our new research reported in paper Z we now believe that we made an error in omitting that sample and think our here proposed method has systematic flaws that lead to results like the sample.”
A lot of research is wrong for one reason or another. If a scientist finds out that a paper of someone wrote contains errors the usual process is to write a new paper. Science progresses through new papers and everyone accepts that old papers often contain errors and are overtaken by new research.
The only thing that might be research misconduct is to to forge your research results by not changing the data you publish. Given how hard it is to persue people for massive fraud it’s however unlikely that anybody will argue in present day academia that leaving out one sample for understandable reasons will constitute fraud.
An erratum might say:
“We initually omitted one sample from our results because we believed it to be due to measurement error (or whatever rationalization you had at the time). That sample was X. Given our new research reported in paper Z we now believe that we made an error in omitting that sample and think our here proposed method has systematic flaws that lead to results like the sample.”