Before you link that next study

Post Reply New Topic

Science may be going through a bit of a crisis at the moment.

http://www.laboratoryequipment.com/news ... ing-funded

Some highlights:

This is also an interesting site that tracks fraud and other abuses in science and academic research:

http://retractionwatch.com/

Well, I think it's nice that people are actually trying to replicate studies. That's the messy process of science/research.

The answer here is not to cut funding for research but to INCREASE funding for follow up studies.

As a matter of fact, every research project should have a follow up project as a matter of course. That's just good science.


Ultimately the best indication of good research is successful real world applications.

Which is why we have the following...

https://en.wikipedia.org/wiki/Randomize ... lled_trial
https://en.wikipedia.org/wiki/Placebo-controlled_study
https://en.wikipedia.org/wiki/Cochrane_Library
https://en.wikipedia.org/wiki/Meta-analysis
https://en.wikipedia.org/wiki/Funnel_plot

... and the entire concept of scientific peer review, of course...

It is refreshing that the OP has decided to focus on the reproducibility of scientific studies, however...

After all, one of the studies which could not be replicated was the 2012 study by Mark Regnerus - presented by the OP here - to "support the traditional family structure " - which claimed that children of homosexual parents had suboptimal life outcomes.

See here:
http://www.sciencedirect.com/science/ar ... 9X1500085X
https://www.sociologicalscience.com/articles-v2-23-478/
http://www.ncbi.nlm.nih.gov/pubmed/25018575

And I'm sure that the OP's sudden - and surprising - embrace of scientific rigour will extend not only to his evaluation of the factual accuracy of peer-reviewed scientific studies, but also to the factual accuracy of any and all YouTube videos he might consider relevant for sharing on this website...

Last edited by GGPViper on 20 Jul 2016, 3:45 pm, edited 1 time in total.

I'm a scientist. Let me explain something. I would LOVE to do replications. I would ADORE that possibility. Who is going to pay me to do it? Not the government. All government agencies want "novel" proposals. If you propose to replicate something, you won't even make it past the first level of review. Industry? No. They want something that nobody has any possible IP rights to, so replication is not what they will fund. Charity? No. Just TRY to explain to some charitable donor or institution the importance of replication. They will conclude that you are incompetent because you don't have something "new".

So, whine all you like about lack of testing replication of results, until the cultural hatred of replication is dealt with, it won't happen, because nobody will pay me to do it.

What science isn't completely flaw free and is vulnerable to human error, bias and such things and results aren't always infallible? who would have ever suspected...

Most modern 'scientific research' seems pointless.

Here is some recent scientific research:
- Pizza is the most addicting food in America.
- Late night snaking could damage your brain
- Hugging your dog is bad for your dog
- Drinking a glass of red wine is just as good as spending an hour at the gym
- Liberals are better than conservatives at 'smizing' (smiling with your eyes)
- Your cats might be thinking about killing you
- Bears engage in fellatio
- Smelling farts might prevent cancer
- Coffee might reverse cancer

source:
http://www.motherjones.com/environment/ ... ure-cancer

It appears 'p-hacking' (probability hacking) is the flaw in scientific research.
http://psychology.wikia.com/wiki/P-hacking

You play around the with the data until you show some statistical significance.

source:
http://www.motherjones.com/environment/ ... ure-cancer

Not really, I try to avoid them if at all possible, I link it because it's relevant to my view that when reason is rebutted with "but studies show..." it should be taken with several mouthfuls of salt.



Here are some other choice quotes of mine from that thread:

social science in particular has gone the way of newspaper polls, maybe it was never used in any other way. While polls are now commissioned to influence opinion instead of measuring it, scientific studies are now commissioned to change reality rather than observe it.

Hehe I just said it was ammunition for the debate. I read 2 rebuttals of this study before linking, I did note that neither tried to argue against the assertion that the majority of the often cited "pro" studies showing the opposite were seriously questionable.



But then you would have no place on this forum. I may be the Lord of all that is Evil, but it is too cruel to deprive a man of his station. Where would we be without your valiant arrival in every thread? Who else would ignore the overarching topic, heroically find one mistaken element out of a hundred and announce the entire thesis null and void? Who could nobly dig up and misrepresent the past and cast a shadow over a man's character? Who else could gallantly rebut any train of thought with a self-reported social science study and smear a person as misogynist from the most tenuous of links. It wouldn't be the same without you.

Such is life in a capitalist system, not that the communists did much better. I sometimes wonder if the older system was better, where science was a hobby rather than a career. Performed by outcast lunatics either with their own resources or with the charity of equally insane rich patrons. It's true a great deal of potential talent can be wasted, but at least it is driven more by true curiosity than the chasing of profit or favours for the government propaganda wing or backed by a degree mill conning millions of students into enormous debt for the rest of their lives.

I imagine any third option would involve a "committee" of some sort, which of course just switches the potential for corruption and bias to a different set of hands.

hasn't it been determined by science yet that psychology isn't a science?..

seriously. it's a very interesting field, and i'm pretty obsessed with it myself sometimes. but it's just not science

Science is just secular religion to a lot of people in how they put faith in it to guide their decisions, worshiping at the alter of learned sages in white robes teaching us irrefutable truths laid down in some book. People might call that anti-intellectual but I don't see putting blind faith in something to intellectual, you should always be skeptical and question people's real motivations. Follow the money is always good advice, people wouldn't believe a study about the effects of smoking cooked up by the tobacco industry so why would you believe a study from the pharmaceutical industry or from someone heavily invested in 'alternate energy'? If you already have a preferred result and personal stake in it then then confirmation bias no doubt takes over.

I watched this movie 'The Big Short' recently and there was a scene where they go I think Moody's(the credit rating agency) to investigate how they could still be rating these mortgage securities and CDOs AAA even tho they are filled with garbage subprime mortgages and the answer they eventually get is that if they didn't give them the AAA they wanted that they'd just go down to the street to Standard and Poor's instead which is essentially I feel like a lot of research amounts too especially when politics/policy is tied into it. If you can shop around until you get the answers you want then it kind of undermines the credibility of the whole thing. Everything is politics.

This thread recalls peer reviewing scandals exposed by Nature and other prestigious journals last year. The editorial boards of the journals had thought their processes of scrutiny were adequate, and were shocked to find who was doing what, how and how often.

A recent contributor to Nature wrote about a new aspect of scientific manipulation:

http://www.nature.com/news/watch-out-fo ... me-1.20246

In the links below is one that goes to the exposure of the peer review scandal as mentioned above.

The single most important thing is ending publication bias. We need journals which will publish anything, regardless of how "interesting" it is. We should also require declarations of intent before research starts, making it harder to hide inconvenient truths. Funnel plots are a hack.

It seems that the OP, who happens to be a British Eurosceptic, is less concerned about encouraging good scientific processes, and more about disregarding facts and expert opinions when they don't suit him.

Laugh if you wish, but I've just poured millions into a study that will prove my opinions are 112% correct on 123% of topics 102% of the time.

I wonder if I can get one of those "studies" to "prove" my opinions are -12% correct on -23% of topics -2% of the time and be paid for it.