Spanish economist Josep Colmer has a neat piece on the way that political science uses equations.
What other sciences look like
Josep M. Colomer (University Pompeu Fabra, Barcelona)
Prepared for a Symposium on “Why Political Science is Not Scientific Enough?”
In order to have references for discussing mathematical menus in political science, I review the most common types of mathematical formulae used in physics and chemistry, as well as some mathematical advances in economics. Several issues appear relevant: variables should be well defined and measurable; the relationships between variables may be non-linear; the direction of causality should be clearly identified and not assumed on a priori grounds. On these bases, theoretically-driven equations on political matters can be validated by empirical tests and can predict observable phenomena.
Colmer’s piece is a useful counterweight to the common notion that good social science is about chucking in variables to maximise R-squared.
I think about a hundred and fifty years ago it was established that social sciences, unlike natural sciences, was not susceptible to abstract formulae because the subject was also the object. Surely in Australia we don’t have to go through the same argument all over again do we?
TPS, abstract formulae are just a way of expressing a hypothesis. My frustration with a lot of political science is that their theories aren’t clear.
Before we all throw around the word science too much, let’s remember
that “hard” scientific method is about observing phenomena, evolving an hypothesis that abstracts the phenomena in a way that allows reliable predictions to be made, then testing that hypothesis in such a way that the manner of testing doesn’t affect the result. This is the nature of “objective” science.
Then we must remember that the “hardest” science of them all, physics, has spent the last century slowly realising that the nature of matter at the quantum level may very well defy objective observation. Is the very best “observation” of our most fundamental substance telling us subject and object are at best constructs?
It can be argued that as all of the social sciences are “observed” from inside of the phenomena (i.e. by people participating in the phenomena either directly or indirectly), there can be no “hard” hypotheses and certainly none that can be tested in a truly objective way without affecting the results.
Is there such a think as a true social “science” double blind?
If not, perhaps it’s time to suggest social “science’ is “quantum” in character and not “classical”, meaning any so-called empirical validation of an equation is a fundamental furphy.
Or… all this could be sophistry from one not in the business?
“Then we must remember that the “hardest†science of them all, physics, has spent the last century slowly realising that the nature of matter at the quantum level may very well defy objective observation.”
… but macroscopic quantities (eg pressure) can be observed with minimal disruption.
So, Sacha, the question is whether the phenomena being abstracted are of a classical character (i.e. can be observed in a context with “minimal” disruption” – there is NO undisrupting measurement) or quantum character (i.e. either the thing is disrupted by observation or the thing being observed is like spin and doesn’t “exist” until it is observed).
I think it’s important aspect of the discipline of rigourous thinking to ask whether the assumption of objectivity is appropriate and to account for the scale of observation’s disturbance. The price of thinking is eternal vigilance?
I’d suggest each poll infects the next one and every trend study reinforces the trend. Our underlying is always folks, and we can’t forget how queer they are…
They certainly are!
To put it bluntly, I think it is a conceit of political scientists to pretend that they are observing society outside of it, rather than having their view determined by their position in society and their political response to it. Presenting prejudices as an abstract formula is a way of concealing that conceit!
It has been so – look at the Adorno F scale for instance.
Doesn’t have to be always. If people are willing to pursue intellectual honesty, then maybe they could be trusted to make observations in social science?