You can tell I just discovered the tumblr web site I Love Charts
So glad to discover this more accessible description of the principal-agent problem afflicting aid (and domestic bureaucracy as well).
A new paper shows evidence that researchers manipulate econometric regressions to artificially attain a “real” (i.e. “statistically significant”) result. See Aid Thoughts for the key graphs, also Marginal Revolution. This is old news, but important.
So of course, at this point you are wondering: how can I do this so I can also get my papers published? Here is a pedagogical illustration: