Comment by chordalkeyboard | Hacker News Reader

Comment by chordalkeyboard | Hacker News Reader

Parent Full thread

chordalkeyboard

Physics and chemistry.

Its possible to do science in biology with things like knockout mice and physical limits.

Other things are more a matter of classification and interpretation and it should be clear that they are socially constructed to a greater degree than physics or chemistry.

> What would you call all of what we now call the biological, social, medical, sciences at the moment?

I don’t know, I was originally defending the comment on the basis of “real science” so I don’t see a problem calling medicine an applied science.

> What would be your proposal for a new category for everything that uses the scientific method, but that may have variations in the individual elements of the treatment and control groups?

See, what I mean is actually a point about the scientific method. Its not really possible to use the scientific method without ensuring that your experimental subjects are indistinguishable. The reason I’m gatekeeping is because its not the same method, it would be pseudoscience. Like doing experiments on football games.

> I would propose as an alternative that we work to improve general scientific literacy so that the general public might be more familiar with the methods used in each field. For example, if more people knew how regression analysis worked they might be better able to understand it as the foundation for many of the conclusions in the social sciences, and so might be better able to understand the generalizability and potential limitations of what they are seeing when they are presented with a conclusion in one of these fields.

I think it would be great if people understood these things better, I guess I see this distinction as a way of articulating those fundamental differences in the mind of the public.

You ignore the variability of KO mice? You'll never get the exact. same response between two mice. nor will you between two cloud chambers. so the real science left is the one someone makes up in their mind maybe...

I’m not sure what you mean. Variability in physically identical specimens is handled as probability. KO mice are basically the same organism, when they react differently to the same stimulus it indicates something about all KO mice. When two humans react differently to the same stimulus it tells us something about two humans, but nothing about the rest of the humans.

I think you may be vastly overestimating the scientific literacy of the general public.

I don't mean to imply that most people are stupid or anything. Far from it I've met many tradesman (and tradeswomen) who I consider vastly more intelligent than many people I've met with PhDs.

But, I suspect the willingness of most folks to care about the distinction you're making is negligible at best.

It's sort of like being a car salesman who truly loves cars. And the type of car you're selling has a brand new oil-pump bearing design that is 30% more efficient than the prior bearing design. It doesn't make much of a difference in the overall performance of the car, but you're fascinated by the engineering prowess of it. Unfortunately, 99.9% of the people who walk in the dealership simply don't care about the oil pump bearing. They care about the heated seats, and the bluetooth audio, and if they can get it in seafoam green with the black trim package.

In a similar fashion, although I find knockout mice to be absolutely fascinating (to steal your example), my friends and family simply glaze over when I start to ramble on about such things. They simply don't care. But when they need the gene therapy that resulted from studies performed with the knockout mice for their weird cancer, well, then they're interested (but not too much).

And, so, most people don't know a for loop, from a plasmid, from a cloud chamber, from a difference-in-differences regression equation in economics research. And as a result, they're not even remotely interested in the degree of stochastic factors that may be influencing the research design of the study poorly-summarized in the Medium article that pops up on their Facebook feed.

I would love for people to be more savvy in interpreting the products of the scientific method (or any variation thereof) that they may encounter. But most people are so far behind in being able to approach such detours into the methodology that such a semantic distinction would prove useless. Hell, I have more years in grad school under my belt than I care to admit, and even I will tell anyone who will listen that I barely understand most stuff outside of my field. A political science paper utilizing network models to analyze trade fluctuations with third-party nations following bilateral trade agreements. Maybe I could get the gist of it, but I'm probably not confident enough to definitively say if it's bullshit or not.

Rather, I argue that we focus on piquing the public's curiosity, and doing our damndest to act as scientific ambassadors to the next generation. Rather than changing what we classify as 'science' in the colloquial sense, give a nephew or niece a copy of Calculus Made Easy (and actually help them get through it), show them how to use a metal lathe, give them a brief demonstration of OLS regression by hand, get them a decent microscope to look at pond water with, build a TTL flip-flop with them on a breadboard, and read them The Logic of Collective Action as a bedtime story for the hell of it. The only way we lift the ability of the general public to understand the application of the scientific method in all of it's forms, is by teaching the general public to give a shit about it in the first place. For most people, it's just too late, but it's certainly not too late for us as a collective society.

Thanks for your interesting reply and I’m glad that we’re both working towards scientific literacy in our own ways.

Reply on news.ycombinator.com