Do Economists care about the real world?

I've been burning through my pre-downloaded Radio Economics Podcasts over the past 24 because I've been without internet! The most recent one was on what Economists can learn from Biologists.

A common criticism of economists, certainly by Tyler Cowen, is that they don't read enough intellectual writting from other disciplines. The reason this is disconcerting to some is because economists tend to high-jack other disciplines. Over the past 15 to 20 years many economists have focused on traditionally more behavioral based topics, things that would usually be deemed the territory of Psychology or perhaps Sociology.

An economist, by my definition, is a fan of the economic way of thinking. Often we feel so confident in the power of this way of thinking, that we would like to extend it to fields outside our core competency. Experts in those fields may claim that we have nothing news to add.

The economist does not argue that by pooring over data and looking at incentive structure, she will discover a new biological process that changes our perspective on Nature vs. Nurture. Though the economist may try to argue that, given we don't know exactly how the process works, let's use collected data and logic involving incentives of individuals and see if we cannot develop a model that helps us understand behavior better. This is a theory. If the theory helps us predict behavior, then so much the better. However, if we test the theory in the real world and prove the theory wrong, that is also progress.

A common problem in economics is to develop a theory that makes sense without supporting it adequately with empirical analysis. It is a lot easier to preach from the soap box these theories that sound so logical without putting them to the test. I would love for my ideas to be supported by empirical analysis, but I am usually not in the position to attain the necessary data (either it doesn't exist or I don't have access to it). Whenever some one develops a theory about the way the world works, even if it makes sense, there is a need to test it in the REAL world. Are the results statistically significant? If so, are they really significant enough to change the way we behave? The value added by economics rests on these two questions.