Do The Easy Neglected Task

Academics use many different methods of inquiry. While different disciplines use different methods, most are useful in some contexts. 

Academics present themselves as mainly just trying to understand things, and as choosing their methods for that purpose. But in fact they are more rewarded for being credentialed as impressive, and for letting customers (students, journalists, patrons) gain status by associating with them as such. 

These rewards substantially distort academic methods. To be credentialed, intellectuals need to use methods that their associates also use and understand. And to be impressive, they need to use methods that are hard for most to master. And for both purposes, they prefer methods that let associates easily and reliably rate the impressiveness of their work, even while paying minimal attention to its details. Also, in areas that value new results matching predetermined conclusions, they value methods with sufficient degrees of freedom to allow such matching.

Much of the task of understanding stuff boils down to matching theory to data. This can be assisted by the supporting tasks of theory, i.e., carefully calculating the implications of particular theories, and also data, i.e., carefully documenting key data patterns. In my home field of economics, each of these supporting tasks has been greatly elaborated with complex math and computation. On the theory side, people carefully calculate the equilibria of various math-specified game theory models, while on the data side people carefully fit math-expressed stat models to large datasets. 

The core intellectual task is to collect many possible theories and many related data patterns, and match them to each other, noting how many of the patterns each can explain, and at what cost of adding auxiliary model assumptions. As the best theory usually fits the data quite a bit better than other theories, one doesn’t usually need to look that closely at details of theory implications or data patterns. A crude qualitative comparison is usually sufficient.

However, the supporting tasks of theory and data, are usually more standardized and impressive, more easily rated by associates paying little attention to to their details, and more easily adjusted to ensure they get desired results. And thus it is much harder to be reliably impressive on the core task of mostly qualitative matching of theory implications to data patterns. So academics tend to emphasize these supporting tasks of elaborating theory and data, and neglect the core task of matching theory to data. 

If you are trying to impress academics, and gain entry into their world, you will probably need to do what they do. But if you are mainly trying to understand stuff, and don’t need to impress academics, or gain their credentials, then you have available to you a simple method to gain much insight fast. Just do the simple neglected task of matching theory to data. Survey those huge piles of theory papers, and data papers, piles which academics keep growing, but doing little with other than to pile higher, collect lists of plausible theories and relevant patterns, and just try to match them. Sure, it will help to learn some theory and statistics, so you can read all those papers; just don’t get sucked too far into those worlds.

That’s it. Other people don’t do it because it is too easy, and hard to judge quality for, even though it is quite effective at the key goal of figuring stuff out.