4 ms·

I believe it's the "aggressively tried to identify the research subjects" that probably warrants a more severe punishment. This is a HUGE no-no in research, as

I believe it's the "aggressively tried to identify the research subjects" that probably warrants a more severe punishment. This is a HUGE no-no in research, as research MUST be, for accurate information, anonymous. His attempt to identify them undermines the very most basic level of good research.

tokenadult 14y ago
This is a HUGE no-no in research, as research MUST be, for accurate information, anonymous. Are you sure that this is an invariant characteristic of good research? What is the authority for conducting a research project in a manner that doesn't allow another researcher to check the raw data? To the contrary, these days there are several very responsible researchers who call for much more open access to research data sets (yes, those could protect subject identities at the individual level while still being traceable to a specific site for follow-up) to promote better science. http://wicherts.socsci.uva.nl/ http://wicherts.socsci.uva.nl/ http://blogs.nature.com/news/2012/07/data-detective-makes-his-fraud-busting-algorithm-public.html http://blogs.nature.com/news/2012/07/data-detective-makes-hi...
- 001sky 14y ago
  I don't have a bone in this fight, but when sample sizes are approaching N=1, you run the risk of evidence tampering, absent anonymity. You also enable political retaliation (ie, principla X supported Y, pull his funding...or oh look at him slip on a sidewalk...woops). Its credible, given the obvious anonymosity (all around), and the politicization of teachers unions (viz, wisconson, chicago, ect) that this consideration be taken into account. Again, this is not a "political point" (partisan sense) as it is a practical one about Naivete, and what desperate people do when they feel threatened.
brg 14y ago
If they believed that the research had been misrepresented, then the fastest way to confirmation is to evaluate the actual data. Unmasking the source could be one way to do this. As they were not a party to the agreement of anonymity, I think their actions are fine.

tokenadult 14y ago

This is a HUGE no-no in research, as research MUST be, for accurate information, anonymous. Are you sure that this is an invariant characteristic of good research? What is the authority for conducting a research project in a manner that doesn't allow another researcher to check the raw data? To the contrary, these days there are several very responsible researchers who call for much more open access to research data sets (yes, those could protect subject identities at the individual level while still being traceable to a specific site for follow-up) to promote better science. http://wicherts.socsci.uva.nl/ http://wicherts.socsci.uva.nl/ http://blogs.nature.com/news/2012/07/data-detective-makes-his-fraud-busting-algorithm-public.html http://blogs.nature.com/news/2012/07/data-detective-makes-hi...

001sky 14y ago
I don't have a bone in this fight, but when sample sizes are approaching N=1, you run the risk of evidence tampering, absent anonymity. You also enable political retaliation (ie, principla X supported Y, pull his funding...or oh look at him slip on a sidewalk...woops). Its credible, given the obvious anonymosity (all around), and the politicization of teachers unions (viz, wisconson, chicago, ect) that this consideration be taken into account. Again, this is not a "political point" (partisan sense) as it is a practical one about Naivete, and what desperate people do when they feel threatened.

brg 14y ago

If they believed that the research had been misrepresented, then the fastest way to confirmation is to evaluate the actual data. Unmasking the source could be one way to do this. As they were not a party to the agreement of anonymity, I think their actions are fine.