"Age of the Empirical"

John McGinnis (Northwestern) describes some of the causes and effects (political and other) of the emergence of the "Age of the Empirical," in the recent Policy Review (No. 137; June & July 2006).  Quite interesting and well worth a read.

Comments

[buy viagra]

hello friend excellent post about Age of the Empirical thanks for sharing

[Kaly]

Hello im 25,he's 57,we have been friends for over 3 years,and now we have become best freind's.He's just a wonderful person,smart,very mature and sensitive.
I met him on the net Agelesscupid.com three years ago. I am now a happy woman who is deeply in love and planning a wedding ceremony before the end of this year by the grace of God.

[Foundation]

Experience the Annual Scientific Meeting in a whole new way… Donors who contribute a minimum of $1,000 to the REF by June 30, 2007 are eligible for exclusive donor benefits at the 2007 ACR/ARHP Annual Scientific Meeting in Boston. Click here for more information. http://big-tits-fucking.blogspot.com It is estimated that by the year 2020, 18.2 percent of all Americans will be afflicted with some form of arthritis. What if someday soon there aren't enough rheumatologists to meet the growing demand for rheumatologic care?

[Dan Kahan]

okay, here (with permission) is what McGinnis wrote in response to comments essentially identical to the ones in my last comment (but also referring to Lott v. Levitt and the recent Sunsten/Vermule, Becker, Posner et al. vs. Donohue/Wolfers debate on the death penalty):

"I would like, if I might, to probe, for the nub of our disagreement, because I agree with much of what you say. Everyone is influenced by priors and many are influenced by biases. The question is whether empirical data nevertheless has an effect. I would concede that for many people, particularly people unschooled in empiricism, it has no effect. But that does not means that empiricism cannot help form a consensus if it changes the views of some, particularly of the elites and experts that have an influence on policy. That is my view of how consensus is slowly made.

"I am doubtful that existence of priors and biases prevents experts and academics over the longer term from reaching consensus about the effects of policies on the world. As I suggest in the essay, there are norms in empirical work and scholars have incentives to comply with these norms, reducing the influence of bias. Of course, at any given time, there will be strong disagreements about some results (witness Lott and Levitt), but over time more studies are likely to bring consensus. I think it has in the past on issues such as sentencing and crime.

"In my view you seem to be conflating trust, which is necessary to reaching consensus in an empirical community, and moral consensus, which is not. I think the culture of empiricism generates norms that gives scholars reasons to trust one another and the changes in information technology are making it ever easier to policy those norms. I do not, however, think a thicker moral consensus that you mention is necessary to evaluate effects of policy on the world. I think it likely that a social scientist working in the culture I describe will have incentives to find that the death penalty deters (if that is what it does–my prior admittedly), even if he has other reservations about the death penalty. Professional incentives and culture often separate the behavior of individuals from being driven by their moral imperatives, narrowly construed.

"Perhaps I am not clear enough about what consensus is. It does not mean that there is no dissent, simply that empirically informed experts substantially favor one position or another. And then the consensus among experts radiates outward, crucially changing the political information mix.

"I quite agree this is a long run effect. One of the points of my essay was to point out reasons–blogs, information markets, and ever faster computers, that the longer run will be shorter (not immediate of course) in the future. Just as some economists, like Romer, argue that technology is accelerating economic growth, so is technology accelerating the forces for consensus.

"Perhaps I can put the question concretely: Do you believe that in twenty years there will be no consensus among experts on the deterrence of the death penalty and on the influence (separately considered) of vouchers and charter schools on student performance? If there becomes an expert consensus, what effect would it have in forging a consensus in the rest of society?"

[Dan Kahan]

I also was happy to read McGinnis's piece -- which was extremely well done (characteristically so). And I certainly wish that he were right. But I think the empirical evidence runs decisively against him. There is expert consensus among social psychologists that individuals conform their evaluation of empirical evidence to their morally grounded prior beliefs about disputed policy matters. Classics include Charles G. Lord, Lee Ross & Mark R. Leper, Biased Assimilation and Attitude Polarization: The Effects of Prior Theories on Subsequently Considered Evidence, 11 J. Personality & Soc. Psych. 2098 (1979); Robert J. Robinson, Dachter Keltner, Andrew Ward & Lee Ross, Actual Versus Assumed Differences in Construal: “Naive Realism” in Intergroup Perception and Conflict, 68 J. Personality & Soc. Psych. 404 (1995); Lee Ross, Reactive Devaluation in Negotiation and Conflict Resolution, in Barriers To Conflict Resolution 27 (1995). Sophisticated individuals, like scientists, aren't immune from this effect. See Jonathan J. Kohler, The Influence of Prior Beliefs on Scientific Judgments, 56 Org. Behavior & Human Decision Processes 28 (1993) (finding that scientist subjects deemed fictitious ESP study to be sound or not depending on result it reached!). This is definitely *not* to say either that empirical work is of no use or that political consensus cannot be obtained on facts relating to the death penalty, gun control, global warming etc. On the contrary, the point is that creating political consensus in support of sound empiricism itself requires the development of a separate, *empirically grounded* science of policy advocacy.
BTW, I shared these thoughts with McGinnis, who drafted a very generous & thoughtful reply, which I will happily post if he gives me permission.

[William Henderson]

I was happy to read John M's optimism; I hope he is right. Consensus on empiricism requires a threshold of sophistication among political actors to separate valid from invalid studies and conclusions. Absence this greater sophistication, more empiricism arguably could make matters worse.

bh.

[frank cross]

I wouldn't hold out too much hope that empiricism would create widespread consensus. However, I suspect it accomplishes two things:

a. creation of relative consensus among certain influential subgroups (e.g., many may believe in creationism but scientific study has most of the elite sold on evolution) and

b. shift the terrain of the public dispute (e.g., the debate is no longer over whether to ban teaching evolution but whether to require additional teaching of ID).

I also suspect that empirical results have affected policy in other ways. For example, I think they have moved liberals toward the right/center on various economic issues and moved many conservatives toward the left/center on other issues.

[Dan Kahan]

Hasn't McGinnis's claim that empiricism generates political consensus been shown, empirically, to be false?

[Chris Zorn]

And, as if on cue, we have these nuggets from the Souter dissent in today's Kansas v. Marsh decision:

"Today, a new body of fact must be accounted for in deciding what, in practical terms, the Eighth Amendment guarantees should tolerate, for the period starting in 1989 has seen repeated exonerations of convicts under death sentences, in numbers never imagined before the development of DNA tests....We are thus in a period of new empirical argument about how “death is different,” Gregg, 428 U. S., at 188 (joint opinion of Stewart, Powell, and STEVENS, JJ.): not only would these false verdicts defy correction after the fatal moment, the Illinois experience shows them to be remarkable in number, and they are probably disproportionately high in capital cases."

(Tip of the hat to Orin Kerr for pointing this out).