The median sample size of the studies subjected to replication was n = 5 specimens (https://osf.io/atkd7). Probably because only protocols with an estimated cost less than BRL 5,000 (around USD 1,300 at the time) per replication were included. So it's not surprising that only ~ 60% of the original biomechemical assays' point estimates were in the replicates' 95% prediction interval. The mouse maze anxiety test (~ 10%) seems to be dragging down the average. n = 5 just doesn't give reliable estimates, especially in rodent psychology.
This should be the top comment on HN where most users claim to have some grasp of statistics. N=5 implies a statistical uncertainty of about 45%, so they measured what one would expect, which is essentially nothing. Also this is specifically about Brazilian biomedical studies, and contains no evidence to support people's various personal vendettas against other fields in other countries. At least read the article people.
I can confirm that some (30%?) mechanical and biomedical engineers follow the described problem "solving" strategy exactly. It's not just software engineers.
> Well, obviously the part-time thing will bring a reduction in my institutional teaching and admin duties. I have to say there is uncertainty about how much relief will arise in practice
As someone who has tried something similar, institutional bureaucracy expands to fill all available time. People engaging in bureaucratic empire-building will still happily consume your personal unpaid time. And splitting attention between multiple lines of work creates some legitimate additional overhead, which doesn't help.
I'm not sure what the winning strategy is. I think it is necessary to either get out entirely or play the bureaucrats' "system-game" to some degree, but not on their terms and not fairly. When the bureaucracy demands useless work, maximize their costs and minimize yours. Many academics constitutively cannot make themselves do a lazy, poor job, but is a useful skill to deploy defensively so that you can fulfill your education and research responsibilities. Often you'll find that the bureaucracy only cares about the superficial appearance of compliance; the actual actions performed are irrelevant to them. Shift responsibility to some other part of the bureaucracy and use LLMs to generate boilerplate. If the bureaucracy never attempts to punish you in any way, that may indicate you're being more compliant than necessary. This approach is safest if your retirement plan is fully funded and you don't truly need to keep the job. It is also helpful if at some part of what you do is visibly important to someone who does have power; this helps deflect consequences when you accidentally step a over a line. Everything depends on context and execution though; I hope the part-time approach works out for you.
Prioritize work by ROI and alignment with the institution's mission, communicate the prioritization to relevant decision-makers in your management chain, and actually follow through on it unless coerced otherwise.
Edit: There's no guarantee this will work out positively—nothing is guaranteed—but it's worth considering if the alternative is giving up / changing careers.
Securing the funding often means writing a research plan that is interesting and convincing enough to be selected for funding in a competitive review process (80–97% rejection rate). The plan usually represents a substantial intellectual contribution that serves as the foundation for derived papers. As long as the people who join the project later don't freeze the planner out of the paper writing process, they'll usually meet all authorship criteria.
The upvote and downvote buttons are a couple millimeters apart on mobile and there's no feedback (that I've noticed, anyway) regarding which one you've pressed. There's probably a lot of accidental voting.
Well, the authors' byline says they're from Northeastern University, Shenyang 110169, China and Faculty of Management and Economics, Dalian University of Technology, Dalian 116024, China. I've heard that Chinese universities sometimes have explicit publication quotas or offer cash bonuses for publications. Having never worked at a university in China, I cannot personally verify that though.
MDPI also spams anyone it thinks might be willing to submit a paper or serve as an editor, and does not seem to care much about the quality of the submissions it receives.
The concept of indirect fitness must be more complicated than explained here. The article explains it as a worker bee sharing 75% of her genes with her sisters, but only 50% with a child, so there is selection pressure for workers to be sterile and self-sacrificing. But few genes actually differ between individuals, so the percentages are much higher. E.g., I share ~ 99% of my genes with each one of you reading this. Assuming honey bees' genetic variation is not much more extreme than human variation, we're talking about 99.5% vs. 99.75% sharing, which sounds more like an explanation of why altruism would be preferred in general rather than uniquely affecting bees.
The article does eventually circle around to acknowledge this, but it's easy to miss and very underdeveloped compared to the discussion of kin selection: "So why do bees die when they sting you? Perhaps because they're disposable parts of a larger super-organism which has evolved by multi-level selection."
It doesn't matter how much bees have in common. The idea is that in bees, altruistic traits, that is those that produce more sisters by helping the queen have a 75% chance of being passed, because sisters share 75% of a worker bee genes. Most of the genes are the same, of course, bees won't become dogs or anything like that, but a few of them differ, and these are the one that matter.
Could worker bees be fertile and have a selfish traits that let them have more children, they would only have a 50% chance of passing these, because children share 50% of genes.
So: 75% of altruistic genes pass vs 50% of selfish genes. Altruistic genes win. Humans can't pass 75% of their genome this way, so that altruistic genes have no intrinsic advantage over selfish genes.
Right, "sharing" here must mean DNA that was cloned from the same ancestral DNA strand, not merely that it shares the same informational content. I got lost in the analogies that frame things in terms of what's "better" for the organism and lost sight of this.
The most important thing from the perspective of replication of a DNA strand is the number of copies of DNA passed to the next generation, and future generations, right? Which would be 0.75 * (mean marginal increase in next-generation sisters) + 0.5 * (mean # offspring). The probability that these next-generation individuals actually get to reproduce in turn would also factor in somewhere.
What's also interesting is that if we take the point of view of the queen (through whom the altruistic genes must pass), the queen's reproductive strategy is relatively few children + hordes of sterile helpers + killing her own sisters. So are we talking about a fitness advantage of altruistic traits (maximizing # sisters), or a fitness advantage from selfish traits [maximizing P(fertile child survival) I guess, since # children is small] that produce hordes of sterile helpers?
Edit: Circling back to the organism perspective, in the sense of "I would gladly give up my life for two brothers or eight cousins.", how many bees is it worth giving up one's own life for in that specific sense? We do have a common ancestor after all and thus a non-zero R-factor.
Hmm, I understand this difference in genes differently.
You and I probably share 99% of effective genes, but still the difference in genes is much greater because there we are comparing the entire DNA. There is a lot of non-affecting DNA. And that is what they analyze when comparing DNA of two individuals in forensics.
Based on the information I found, the % difference between two randoms humans in terms of base pairs (including non-coding DNA) is even less than the difference in terms of genes, so the distinction does not materially alter the discussion. Also the article framed its explanation in terms of genes, not base pair sequence.
Forensic comparisons are mostly about comparing the number of short tandem repeats at handful of loci, a very small part of the the whole genome.
If you have any information that indicates the DNA similarity between people is less than 98–99% I would love to hear it. I have not personally analyzed the sequences from the 1000 genome project to check, and am relying on summaries written by other people.
Shifting the topic from research misconduct to good laboratory practices, I don't really understand how someone would forget to take pictures of their gels often enough that they would feel it necessary to fake data. (I think you're recounting something you saw someone else do, so this isn't criticizing you.) The only reason to run the experiment to collect data. If there's no data in hand, why would they think the experiment was done? Also, they should be working from a written protocol or a short-form checklist so each item can be ticked off as it is completed. And they should record where they put their data and other research materials in their lab notebook, and copy any work (data or otherwise) to a file server or other redundant storage, before leaving for the day. So much has to go wrong to get to research misconduct and fraud from the starting point of a little forgetfulness.
I mean, I've seen people deliberately choose to discard their data and keep no notes, even when I offered to give them a flash drive with their data on it, so I understand that this sort of thing happens. It's still senseless.
Recipes like these aren't useless, but yes, they really need to be prefixed with whether they expect to start from a clean work tree and empty staging area. Or describe what they'll do to uncommitted changes, both staged & unstaged. Otherwise they pose a substantial risk of making the problem worse.
> Or, better yet, the gay satanic-panic currently gripping half the country, and the insane culture war being waged around it. You can't actually believe that all those people who have strong opinions about it have been somehow personally wronged by homosexuals.
Or the satanic panic over Dungeons & Dragons in the 1980s. One of the cops ("school resource officers") in the middle school I went to still believed in that nonsense and it was the early 2000s by that point.
reply