Hacker News Reader

Comment by forgot-im-old | Hacker News Reader

Many people are pointing to the replication crisis as indicating that Psychology is not a science. However, Medicine is also facing a replication crisis to the same degree. Would people also suggest that Medicine is not a science?

I think it comes down to the fact that both disciplines work with people and behaviour, and people are not homogeneous nor are behaviours easily predicted. For that reason, I think that a distinction can be made between “hard sciences” and “soft sciences” - we just can’t get the same level of precision that the hard sciences can, but that doesn’t mean that Psychology isn’t scientific. We still apply the scientific method to discover phenomena and develop testable theories, just like other sciences. And meta-analyses allow for much greater certainty in findings..

View on HN

It's possible that fields can fluctuate in how scientific they are over time.

Physics has in the past been a case study of how to successfully deploy the scientific method, but modern physics is often criticized from within for spending all its time on mathematical "theories" that can't actually be tested, something conventionally considered unscientific. Epidemiology was once built on basic scientific observation, it's now also disappeared down the drain-hole of endless mathematics without real world hypothesis testing.

Medicine in contrast spent a lot of time being unscientific in the past, and is now much more rigorous. The problems here are usually fraud done in aid of avoiding the scientific method without being detected, rather than the actual lack of a scientific method at all.

And some fields have got more scientific over time, or at least more quantitative. Economics and education are like this.

It isn't very scientific if a discipline persists in bad research practices, and that's what psychology is still doing. Indeed, it's very hard to get experimental evidence for a theory by measuring people, but if you know you can't, then don't do it. Wait until the means to do so become available, and try to find those ways instead. Unfortunately, that's not the state of psych. research.

There is of course also a more practical side. For psych and med interventions, you don't need to know how it works. If method A works better than method B, it's a good idea to try A before trying B. However, that research is also difficult, and many papers have been found wrong afterwards. While that's the way to go, it shows that even when attempting to establish superficial effects, the disciplines are failing.

I think you are ignoring a massive driving factor in both medicine and psychology; they are businesses.

Both fields, scientific as they often try to be, are subject to the sway of funding and profits more-so than other sciences like physics (as an example). Vague and often unsupported claims sell medical and mental health products, making it a profitable venture whether the product actually works or not.

I say this as a former student of psych, degree and all. The mental health industry has exploded in the last ten years or so, yet not much new thinking has been brought to the table at all. It's a lot of old standards being repackaged as a revolutionary solution (looking at you CBT) but being sold as mobile apps, hooking people with hope for relief in the form of a convenient, easy-to-use package, the same way medicine does when they add a bit of caffeine to acetaminophen and call it a solution for migraines that you don't have to visit a doctor to obtain. Once a field is driven by profit growth, it becomes this twisted, ersatz version of it's former self and this works because people are so desperate, they'll buy into just about anything you offer that might make them feel slightly better.

I would absolutely say that medicine is not science - or at least that it is very bad science with a low probability of being reproducible - and anyone with even tangential exposure to most medical research would do the same. Small sample sizes, confounding variables in treatment, and non-random assignment are the norm, simply because practicing medicine is not the same as performing scientific research.

Medicine is the practice of healing sick people, and (even if they wanted to, which most do not) it is very, very hard for doctors to convince an IRB that it's a good idea to give sick people a placebo. Also, despite some attempts to perform placebo knee surgeries [0], there are many medical interventions where it is simply impossible to do things like a basic RCT.

[0] https://www.nejm.org/doi/full/10.1056/NEJMoa013259

Would people also suggest that Medicine is not a science

I think that depends on whether those people believe that evidence-based medicine is scientific.

I believe evidence-based medicine is possible, maybe, with enough practice.

When the evidence is sketchy or fabricated (or p-hacked), as in psychology, the label "evidence based" doesn't mean very much.

Just look at decades of research into nutrition, and note that we haven't moved very far beyond "vitamin c prevents scurvy" and "too much of anything is bad". Even identifying at what point the amount of things like "too much" of salt or cholesterol crosses the line of "too much" remains contentious.

Certain diagnostic fields have obviously grown leaps and bounds, as have certain categories of medicines. On the other hand, there are counter examples aplenty.

The fact that the term "evidence-based" science exists in medicine research is an indication of a big problem to start with.

I understand the premise of the idea and that more scientists in the field are trying to make their research more rigorous. But, this also indicates that the research that has been done until recently was NOT "evidence-based", hence not very credible and reproducible.

I doubt it's to the same degree, maybe if you include cases like this https://x.com/cremieuxrecueil/status/1844596459162812869

That shouldn’t surprise anyone on the medicine end. People across the industry, from reviewers to those writing textbooks, are receiving money from those they’re reviewing. They’re incentivized to lie to make their sponsors look good. It’s the opposite of science.

Now, we come to the root of the failure. God’s Word tells us to test the source and the content of a message. Certain sources are all about ego, money, or pleasure. They’re willing to manipulate others for selfish gain. Simultaneously, God’s Word teaches habits and traits that honest people should possess. So, we look at their behavior to spot the warning signs that their content should be dismissed or simply reviewed more.

Applying the ancient wisdom to organized science, I found that this problem was pervasive. The institutions, both funding and research, had biases that caused more of specific types of work to be created. Then, there were biases at the individual and group level. These aren’t always bad but are rarely considered.

Further, there were three incentives driving lots of bad science: financial incentive called “publish or perish;” citation index or scores; sensational media. Two of those incentivized cranking out lots of low-quality papers that pressed the right buttons to get more money or citations. The third incentivizes specific claims that receive fame. All three usually penalize, by finance or fame, both steady, grunt work and replication of existing results.

So, rather than the scientific method, what’s happening is more like a game of TV show with winners and losers. It always looks like it uses the scientific method with some amount of real science in it. There’s also some portion of useless work, necessary work not happening, fraud, and censorship. These aren’t random: they’re baked right into the incentives and biases of the system.

Realizing this has made me wonder what I can trust or what I even know out of prior, scientific reporting. Fortunately, we’re blessed that most facts are so unimportant that being wrong doesn’t hurt us. The fields we depend on day to day are close enough to the truth that the products work acceptably well or don’t hurt us. Past that, I wonder what it will take to get to a point where organized science is actually doing science. Consistently.

I hope we achieve this. I worship Christ, not science. It is one of many forms of knowledge, many gifts, that make our lives better if used correctly. I thoroughly enjoy real science. I just think it’s in huge decline with society getting more and more dependent on the fake forms of it.

>Would people also suggest that Medicine is not a science?

Yes, I would. If one looked closely at it it's got the trappings of religion. To give you a guesstimate I'd say 3/4 of medicine is pseudoscience. either completely useless or actively harmful.

Reply on news.ycombinator.com