I agree with most of what you say, except for the conclusion that there is "a world of a difference between their usefulness". there is a world of a difference in the reliability of the theories and predictions, yes, but this doesnt make social sciences useless.

simply put, due to the nature of the field's subjects, social sciences are much more demanding in terms of statistical methodology if you want to be able to draw valid and confirmable conclusions from their studies. equivalently, the results of social science theories are always statistical statements, not definitive ones.

and yes, way too many students and researchers in social sciences suck at math and stats and apply inadequate statistical methodology in both their survey design and in the process of deriving conclusions from their data. but this fact does not make the entire field of social sciences "useless" or "unreliable". there is solid, statistically reliable research done in social sciences, it's just that the share of crap is higher than in physics/chemistry. but thats no reason to have an arrogant stance towards social sciences. for example, the work of isaac newton was legit even though the majority of ""scientiests"" at the time were shams who ""researched"" turning iron into gold and stuff like that.



btw I'm a STEM guy.

This post was edited by Black XistenZ on Sep 11 2017 01:04pm

It totally affects the reliability.

The scientists in Newton's time weren't really researching alchemy. Alchemy could be described as a side-project for them rather than the main body of their work. He was contemporaries with people like Hooke and Halley.

thats why I wrote "statistically reliable" and not "reliable [in absolute terms]".

I think a good comparison would be weather: everything that happens in weather is the result of natural laws. theoretically, a "quasi-Laplace Demon" could reliably predict the weather hundreds of days into the future.

but in reality, there are so many complex interactions and side conditions that weather is de facto a chaotic system. even though weather follows the consistent principles of natural sciences, there is still a big margin of error and it is to this day still impossible to have definitive instead of statistical forecasts beyond a timeframe of a couple of hours.

it's the same situation in social sciences: tons of initial conditions to consider, tons of complex interactions, so that the best outcome one can hope for are statistical predictions and conclusions, not definitive ones.

Being bad at statistics will definitely make experiments they design unreliable


One issue is that a large portion of social scientists doesn't even have replicability in mind.

I'll add a bit more nuance then, since I don't think we disagree that much on this topic. I am describing Sociology as it largely exists right now. There's nothing by definition that requires it to be as unreliable as it presently is. Thus, when I say it's not nearly as useful, I mean that a lot of the experiments it has produced thus far are not useful, because usefulness is directly tied to the reliability of predictive models, and Sociology has a huge issue right now with reliability of results.

The theory of gravity would not be "useful" if it was only capable of predicting the consequences of dropping a bowling ball off a tall building 60% of the time, now would it? It IS useful because it's prediction rates are damn near 100% under the current framework. Sociology is thus significantly less useful in its present form than say Physics because it predicts things significantly less than 100% of the time.

This post was edited by Magicman657 on Sep 11 2017 01:59pm

true dat, but I've personally worked with social scientists who were competent at math/statistics, so I refuse to play into the typical stereotype that all students/researchers from non-quantitative fields suck at math.





that I can agree with.



when one equates usefulness with prediction rates, then what you say is correct. but I'm not so sure if near 100% prediction rates are really necessary when analysing trends in, for example, psychological conditions or in societal change. but there are also counterexamples, e.g. microeconomics where we indeed want near-perfect predictability.

it's not a black and white topic, some differentiation is needed. even though my field belongs to the STEM category, I just dont like this oversimplified "all social science fellows suck at math and their subjects are useless anyway" narrative.

Not too much tbh. Just that we don't yet know how the supposedly 22 genes interact to form intelligence. When it comes to heritage, the randomness of these genes being passed over, making it more difficult. Perhaps in the near future we are able to connect more dots to have a better argument on how these genes are passed over to create more intelligent children. Yet even then we can say something about the average of a defined group, we cannot say anything about the new individual being born. he/she will just be a part of the new group average, with its own unique features and possible extremes..

There is also the Flynn effect. Stating that IQ raises significantly with generations in our current age. Especially visible in developing countries, less visible in Scandinavian countries, whom are near the end of the curve.. Or are they? Basically intelligence is more demanded in developed countries, which is creating more impulses for its inhabitants to grow. These changes are too fast to be considered genetic. (changes in food patters could cause this shift as well).



This post was edited by Knaapie on Sep 11 2017 03:55pm

Is sociology used in anything but marketing though?

I'm a social worker and we don't use it. We have our own theories that fall into line with allied health science.

Pretty sure only business uses sociology....and business science isn't precise at all....sociology and economics are great, important disciplines for sure, just not uses for a lot.

One master of sociology I know runs a call center for gallop poll. Other does fund raising and development. My professor in my intro classes was a marketer. Afaik there is no job title sociologist. This was one of the reasons I went with an applied science instead of a research one like economics and sociology.




Biology is as much social as physical. And medicine is an area things have to be very exact yet playa between the physical and social sciences....particularly psych meds. Unless you have some model of personality already you wouldn't even know what you're looking for.

It all runs together. This phallic compartmentalization youre doing from hard to soft only exists in academia. Everything where it really counts is interdisciplinary.

This post was edited by Skinned on Sep 11 2017 04:30pm

I stepped away from jsp for a few days and looks like I didnt miss much. Just more dishonesty and character attacks on moronic grounds.
Cant refute our arguments? better pretend we should be against charity, despite charity being part of a free market and not mutually exclusive with recognizing basic market economics.

Can't refute my argument?

Better just straw man it to death instead of owning up to the logical consequences of your own beliefs.


Can you make an argument that it is morally acceptable to bail out people who made poor economic decisions?

I don't see a distinction between using public funds and using private funds, the result is the same. You distort market forces to subsidize poor economic decisions.

This post was edited by Thor123422 on Sep 11 2017 06:18pm