Peer Review

There are studies showing that peer review is subject specifically to political bias. Moreover, peer review is not just (or, often, even mostly) about technical mistakes but, rather, about how "interesting", "new", "appropriate", etc., a paper is for a given journal according to some anonymous reviewer. The reasons behind the rejections of papers are often completely absurd.

Oh, to the politics thing? I don't have a link on hand. Basically, there are studies in which the same study is submitted to peer review and the study is about how political ideology correlates with some life outcome and the results are altered and the research shows that studies are more likely to be accepted when they show liberalism being associated with good things.

That is, the exact same study is more likely to be accepted if it has a pro liberal finding

It is not unheard of for reviewers to explicitly reject papers for political reasons by stating that they think a paper is "socially irresponsible" or that the information is "socially dangerous" and such.

Oh, it is sometimes provable. Occasionally a researcher will get pissed at a peer review and blog about the reviewer comments.

More fundamentally though, peer review battles are asinine because if you understand the subject you don't need a peer reviewer, you can evaluate the study yourself, and if you don't understand the subject then you should not be debating it to begin with.

https://topseudoscience.wordpress.com/2016/01/10/the-forbidden-paper-on-the-population-genetics-of-iq/
http://www.sciencedirect.com/science/journal/01602896/46?sdc=1


Peer review is best thought of as consumer testing. A journal is a business that has a consumer base and they sent products, papers, out to a sample of those consumers to see what they think. If the test consumers, the peer reviewers, don't like the paper they won't run it. And there could be 100 different reasons why they might not like the paper. Not only politics wise, but, for instance, peer reviewers might reject a paper because they think that it is poorly written. A methodological flaw is only one of many possible reasons. A consumer testing model makes the best sense of the practice as it is done.
Yes, Piffer's rejection was pretty absurd

The reviewer was citing data against Piffer that hadn't even been released

Here is something I was reading a week or so ago. It is about replication research where they go back and redo a study and, usually, find that the studies findings do not replicate. Then they go back to the journal where the original study was published. The peer reviewers (and editor) or these journals then tell them that they won't publish their replication literally because it is boring.

https://thenewstatistics.com/itns/2017/09/04/adventures-in-replication-scientific-journals-are-not-scientific/
https://thenewstatistics.com/itns/2017/09/04/adventures-in-replication-scientific-journals-are-not-scientific/

http://nypost.com/2014/10/12/liberal-bias-in-academia-is-destroying-the-integrity-of-research/

http://sitn.hms.harvard.edu/flash/2015/as-good-as-it-gets-peer-review-and-its-discontents/

https://menghublog.wordpress.com/2015/06/18/bad-experience-at-mdpi-j-intell-journal/#more-2788
https://menghublog.wordpress.com/2015/06/18/bad-experience-at-mdpi-j-intell-journal/#more-2788
http://emilkirkegaard.dk/en/?p=6910

"There are many flaws and omissions in this work that attempt to lead the reader to the notion that Europeans are genetically smarter than people of African descent – a notion that has been largely rejected by the mainstream scientific community. Given this fact, it is odd that the authors do not make a case at all for the importance of this work in the introduction. It remains unclear why this work would be important and how the research world could or should use the findings."
That is peer review

"According to the authors, “psych 1.6.9 (Revelle, 2017), factor analysis was used to extract a g factor from the cognitive data (Jensen, 1998). Using this method, g scores for 1,369 individuals were derived.” The seven cognitive tests used to compute g were: Dimensional Change Card Sort Test (Card Sort); Flanker Inhibitory Control and Attention Test (Flanker); Picture Sequence Memory Test (Picture Sequence); Pattern Comparison Processing Speed Test (Pattern Recognition); Oral Reading Recognition Test (Pattern Recognition); List Sorting Working Memory Test (List Sort); and Picture Vocabulary Test (Vocabulary).

However, the authors need to provide some legitimate precedents (peer-reviewed papers) to show that those 7 tests are a legitimate means of computing “g”. Most scholars recognize that intelligence is multidimensional, so there is a problem with assigning it a single numerical value without any justification for this approach. "

This is hilarious, because it proves that the reviewer doesn't actually know anything about intelligence research
Its just some random social scientist

Perhaps they study how a mother's interaction with new borns impacts play behavior
Or anything other irrelevant thing
And they are the scientific reviewer of this paper

It is important to realize that peer review can go on and on for many months in which the paper authors have to basically slowly explain their field to the random clueless review who in the end may still reject the paper because it bores them, they don't like how it is written, or they think it has bad social implications. Given this, sometimes research is published w/o peer review just because a researcher doesn't want to go through all this horse shit.

It is really unfortunate that today freshman classes are taught to worship peer review. I think the world would be better if those outside of science basically didn't know what peer review is.
Even worse, sometimes people will be such perfect embodiments of shit that they actually complain about the impact factor of the journal in which a paper was published and on these grounds refuse to consider its findings.

--------------------
"Peer review works, except when the journals refuse to publish something because it goes against their agenda"
http://journals.sagepub.com/doi/abs/10.1177/1075547098019003002
Ideological bias is far from the only thing that contributes to admission of shoddy academic work. Only 25% of a sample of biomedical papers were replicated. To attribute this only to ideological blinders by researchers or journal referees is quite a stretch. There are systematic flaws with peer review that contribute the journals to be flooded with worthless publications.

.Nyborg's 2005 paper Sex-related differences in general intelligence g, brain size, and social status [2] was published in Personality and Individual Differences. Nyborg was suspended following criticism of his research.[12]Aarhus University assembled a committee to investigate, and found him innocent of fraud, but guilty of "grossly negligent behavior" and gave him a severe reprimand.[12] He was eventually acquitted by the Danish Committees on Scientific Dishonesty of the charges of scientific misconduct and the university was forced to reinstall him to his chair.[13] On 21 September 2006, the university gave Nyborg a "severe reprimand", revoked his suspension, and declared the case closed.[14] According to a 2006 news story in Science: "Colleagues from around the world have rallied to his defense, accusing the university of having political motives, and claiming that the errors in his research were trivial.
https://en.wikipedia.org/wiki/Helmuth_Nyborg

When a paper is peer reviewed, it is reviewed by professors in that field. Remember the fake gender studies paper about the conceptual penis that had made up citations? It really shows how that field of "science" isn't really science or reliable. But that doesn't mean the whole peer review process should be thrown out for actual science fields like physics.

I don't know if there is a better process than peer review. If there is I would support it. But if peer review is the best system we have right now and there is nothing to replace it if it gets thrown out, we shouldn't get rid of it.

http://footnote.co/why-biomedical-research-has-a-reproducibility-problem/

https://www.ncbi.nlm.nih.gov/pubmed/21098355