The ideal praise-to-criticism ratio (2013)
hbr.org
hbr.org
At the bottom of the article:
> *Authors’ Note: The journal that published this study has since expressed concern about the data. We first became aware of this research in Kim Cameron’s book, Positive Leadership. Like many others, we were distressed to learn of the incorrect data in the Heaphy and Losada research and we immediately ceased our citations upon learning that the study wasn’t correct. But we do believe the basic assumption and premise that leaders should provide more positive than negative feedback is correct.
Typical social science research, come to the conclusion first and then find data to back it up. There needs to be a reproducibility challenge of the top 100 papers, preferably by scientists on the opposite side of the political spectrum of the authors. We solved this in the hard sciences, I don't get why it's so hard to do it in the social sciences.
What does political affiliation have to do with the scientific method? Are you under the impression that this is done in other sciences?
I’m sorry but as a scientist, your assertion that we must hear from someone with the opposite political view before we can determine the veracity of some work is just so foreign to me. Must we also hear from creationists on evolution research? Why can’t that stand alone? If some experiment is reproducible it shouldn’t matter the political bias of the person running the study.
Moreover how would this even be implemented? The idea of a “political spectrum” is just a rule of thumb at best, and a dangerous misleading heuristic at worst. According to you, how would we identify an author’s political bias? Force them to self identify or take some sort of certified test? Will we divide all of our departments between liberals and conservatives? Right now I don’t know the political affiliation of any of my colleagues, and they don’t know mine. Are you suggesting we should bring our politics explicitly into our research?
I mean… have you even thought any of this through?
Why not? If indeed there is a bias ib social science then we could expect some conclusions to remain unpublished if they contradict the bias. But I don't see why the results that do get published wouldn't be reproducible.
The very article we are reading illustrates why published results can be irreproducible and indeed completely made up. The Losada research cited was invalid [1] yet nobody questioned it because it fit their views of how humans should interact (positivity ratio implies some form of constructive criticism, which is often taught in Western schools as the ideal way to give feedback to someone). Any attempt by a sceptic to reproduce it would have realized that the "math" used to model the positivity ratio was wrong. The authors themselves state in the article they still believe the positivity ratio idea even though the Losada research was shown to be egregiously wrong.
[1] https://retractionwatch.com/2014/07/29/positivity-ratio-rese...
I mean, if you’re talking about completely made up research then that’s a matter of fraud, not political biases.
Should we also only accept reproducibility studies from women on papers by men? Should we need to hear from Chinese or European researchers before American research is considered valid? Must we also take into account the race, ethnicity, and religion of all authors as well? What if a conservative and liberal publish research together, is their research bias free by definition? What mixture of political biases get to reproduce such research in a way you consider free from bias?
I’m sorry the entire idea just seems untenable.
What was that quote by Feynman about how you can tell that someone is full of it if they give a number that's absurdly precise?
[1] https://www.theguardian.com/science/2014/jan/19/mathematics-...
[2] https://physics.nyu.edu/sokal/complex_dynamics_final_clean.p...
* Only you are responsible for how you react to events in your life
* This implies that I'm not responsible for how you react to events in your life
* Therefore, I'm not responsible for how I interact in events in your life
* Consequently, I can speak to you unkindly or abusively and it's your responsibility to deal with your reactions.
It's my personal belief that this stems from a deranged form of stoicism because the same people exhibit toxic emotional detachment of their own subjective experience, but that's just a wild headcannon in order to rationalize what is a real collection of disturbing behaviors.
With regards to evidence, in my lifetime experience to date (limited and what-not), there seems to be minimal correlation in this regard. Some people are mean even if I am nice to them and some people are nice even if I am mean to them. The correlation seems to be no better than a coin toss. Is that just my experience? Possibly. Anecdotal then? Absolutely. Is it equivalent to a social science experiment that assumes the conclusion? Certainly. I would say it’s better if I look objectively at data collected over the decades.