Statistical Modeling, Causal Inference, and Social Science

Economics has more potential for being politicized, I think. A vast majority of people can't get their heads around what you do so it's safe from blatant partisanship.

I reckon I could referee an article in 15 minutes - but only if it is obvious rubbish. The great bulk of papers deserve and require far more attention than this.

Derrida,

I see your point--that's why I said that it would probably be a disaster if all referees followed my approach. But I think it's the best way for me to make a useful contribution to the process. Other people might be better off reviewing fewer papers and taking longer per paper.

Interesting! Personally, I adopt the strategy in the second point. I find that a week of letting the paper's ideas rumble around in the background makes it much easier to identify what I do and do not like. Perhaps the papers are more clearly written in your field.

I'm a junior academic who has refereed a fair number of papers. I find myself being very critical of most of what comes across my desk and wonder if there is a life-cycle effect going on here. I've heard several respected "elder statesmen" in my field remark that people starting out always seem more critical than their more experienced colleagues. Those of you who have been doing this for many years, do you find yourselves mellowing as time goes by!? I suppose this goes for marking as well.

Daniel,

I think you start to realize how hard it is for journals to find good submissions. Even top journals struggle with this. If you compare a submission with your own favorite work, you'll be inclined to reject, but if you compare with the other stuff being submitted, you'll be more mellow. If all the crappy submissions and all the mediocre submissions got rejected, the journals would be nearly empty.