« Updated Segal/Cover Scores | Main | Abstract - Judicial Elections »

20 February 2006


Dwight Hines

Dwight Hines
150  Nesmith Ave
St. Augustine, Florida 32084

February 22, 2006

Dear Empirical Law Blog:

I’d like to see the blog emphasize practical empirical law that is actionable. Given the tremendous amounts of information now available in machine readable forms and thus eligible for meaningful analyses, as well as complex displays, I would argue that contemporary attorneys could define themselves as prevention specialists and remedial behavioral architects to their advantage and to the advantages of our small and large communities.

For example, I’ve been working on boring accident and traffic data and what is intriguing is that the ability to create colorful displays with high amounts of meaning are not limited to major or minor universities. Today, at the GIS Department of St. Johns County, I obtained a ‘spider’ map of St. Johns County Emergency Medical Services missions throughout the county for all of 2003. It is an attractive map and looking at it you realize that:

1) There are lots of data so you don’t have be concerned about sample size not being adequate or the costs of collecting data;

2) You have access to all the data, not a sample, avoiding problems in the understanding of your methods (“You mean to tell me that you only looked at thirty deaths in that study and blah blah blah?”), and your inferences (“WE can NOT justify spending 3 million or x million to save 10,000 lives based on a study of 30 deaths.”) NOTE: these are just fictional examples;

3) An attorney, a practicing attorney, not one who sits around joyfully reading law reviews (he expects us to do that), might like for us to suggest ways of breaking out the aggregate data that would allow for the prevention of future problems with the government funded fire rescue departments (I am using the fire department as an example, to my knowledge no one has made and placed any nooses on the equipment of black fire fighters in St. Johns County*) in many, if not most jurisdictions. So, calling the GIS department should take three minutes, asking for breakdowns by race and age and diagnosis (and amount of time for transport), and if the patient left the hospital alive, if they were alive on arrival, should take another three minutes. Making the computer runs and having the results emailed as pdf attachments should not take more than five minutes, with all of these things happening while the practicing attorney checks over her investment portfolio, will have a practical appeal to practicing attorneys.

What increases the appeal of these types of practical questions that lead to actionable answers is the results can be combined with results from other data sets the practicing attorney may not be aware of in her community, like predatory lenders (Steven M. Graves and Christopher L. Peterson , “PREDATORY LENDING AND THE MILITARY: THE LAW AND GEOGRAPHY OF “PAYDAY” LOANS IN MILITARY TOWNS”, www.responsiblelending.org/ pdfs/Testimony-Graves_payday-032905.pdf), and the growing number of catalogued illegal toxic dumps (http://www.streport.com/files/Duval_County/suprfnd.htm), or substantive and continuing violations of state public records statutes (www.sunshineweek.org) that directly precede and parallel most, if not all, illegal and harmful acts (I don’t have a reference or citation for this sentence but I think Stiglitz’s Nobel Prize winning work on development and media would support the general statement, “ Roumeen Islam, Simeon Djankov & Caralee McLeish (Eds.), “The Right to Tell: The Role of Mass Media in Economic Development” 2002, World Bank Institute, Washington, D.C., ). The practicing attorney could directly relate simple practices, shown repeatedly and in exquisite detail by the data sets, to individual harms and also show how combinations of exposures, cause some extremely specific harms, harms that are quite quantifiable and are amenable to accepted remedies. The remedies must be crafted to insure the responsibilities for the harms are in the correct places so changes are made that make past practices unlikely to occur. Relating general harms to specific causes, although societally desirable as one way of increasing the health of a community, at this time may not be practically possible, even with the powerful methods like the variations of Mantel-Haenszel tests. However, I think a good practical empirical law blog would discuss how Mantel-Haenszel tests could be used for cases other than those involving discrimination or cancer treatment.

4) An attorney, a practicing attorney, would need to know some good visual ways of explaining multiple regression, for example, that are not an insult to statisticians around the world or an insult to their lovely numbers around the universe, while at the same time making regular people exclaim: “Of course, it’s obvious, you ninny.” See Edward H. S. Ip,
“Visualizing Multiple Regression”,Journal of Statistics Education Volume 9, Number 1 (2001), http://www.amstat.org/publications/jse/v9n1/ip.html. To be fair, the practicing attorney would want to be assured that there is not some forgotten pesky statistical assumption that would be worse than a landmine when it was discovered.

5) A judge, who is definitely a practicing attorney, but not a private one, would benefit from a practical empirical law blog by seeing how scoundrels do use statistics, or have used them, to the detriment of all of us, as well as the complexities of modern free market economic law (Jennifer Alix-García & David Zilberman, “The Effect of Market Structure on Pest Resistance Buildup”, www.umt.edu/econ/papers/pest%20resistance.pdf)

6) A practicing attorney, when taking action on data that she learned about from the empirical law blog that are applicable to her cases or are her case, will need to know ways of explaining how her results show that it is better, much much better, for example, right now to have ten shares of Winn-Dixie Grocery stock than to buy 5 lottery tickets, even when the payoff for the lottery is ten million dollars and Winn-Dixie is in bankruptcy and the shareholders’ counsel appears to be cooked. In essence, we need words and sentences that are concrete, that are not minced or mealymouthed, about the power of modern numerical empiricism, that you can use to explain, in a sound bite, to the man in front of you in that long line of people at the convenience store who are there only to buy lottery tickets on a Monday morning and to keep you from getting your needed, much needed cup of coffee.

7) A defense attorney, reading the practical empirical law blog needs to see that he should go into tort law as a member of the plaintiff’s bar because given human nature and large data sets of accurate, reproducible data, the defense bar is going to be on the run. To be fair, to those who want to defend tortious practices, the practical empirical law blog should establish a robust index, a single number that would let us all know approximately where a specific community falls in the litigation potential spectrum. The requirements that such an index be conservative, only reflecting those cases that a solo practitioner with finite resources could take on, and that no novel interpretations of statutes or decisions or admissibility requirements be necessary, must be strictly followed. This would not be for forum shopping but for looking for a place to settle and raise a family.

There are more reasons but I’m responsible for supper tonight.



I'm not sure how much it's penetrated into the mainstream of ELS yet, but I've seen more use of reaction time measures of "implicit" responses (especially implicit stereotypes and prejudice) borrowed from experimental social psychology.

Pizzi, W. T., Blair, I. V., & Judd, C. M. (in press). Discrimination in sentencing on the basis of Afro-centric features. Michigan Journal of Race and Law.

The influence of social psychology in general on economics and law & economics will also be important.

For a nice example, see Some Well-Aged Wines for the "New Norms" Bottles: Implications of Social Psychology for Law and Economics
Yuval Feldman, & Robert J. MacCoun, University of California, Berkeley

And to the extent that economics, psychology, and social science in general are influenced by the neurosciences, those insights will be important to ELS as well.

For an introduction to Social Neuroscience see "Multilevel Analyses and Reductionism: Why Social Psychologists Should Care about Neuroscience and Vice Versa" by Gary G. Berntson and John T. Cacioppo in Essays in Social Neuroscience ( http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=10252 ).

For a wider ranging survey of important primary research articles Social Neuroscience, see Foundations in Social Neuroscience, edited by Cacioppo, Berntson, Taylor, & Schacter. http://www.amazon.com/gp/product/026253195X/ref=pd_sim_b_5/103-9101239-0203057?%5Fencoding=UTF8&v=glance&n=283155

Ross Cheit

I consider what I do to be quite empirical, so I particularly appreciate Dan's comment because my research is rarely quantitative. I study trial court records and trial court transcripts. One example: a little study I just did with some undergraduates looking at the contextual variables in every case in which statutory rape charges were filed in our state over the last five-plus years.

The key variables were contextual and the analysis was descriptive--but this was very empirical. And our data challenge lots of popular claims about how these laws are actually enforced and against whom.

I will consciously promote this research by pointing you to the pdf:

Thanks for this new blog!

Dan Filler

As far as I can tell, there has been a paucity of attention to qualitative empirical work within the law. Unfortunately, I suspect that the (valuable) push for more quantitative empirical work may cause some people to devalue qualitative work because it is non-numeric. This may be because of a common shorthand which conflates "empirical" with "quantitative." In fact, I've been surprised to hear people suggest that qualitative work is the OPPOSITE of empirical resarch. I'd like to see more conscious efforts to promote, and properly frame, qualitative research.

Jason Czarnezki

Lee Epstein (Wash U) has recently, at both Vanderbilt and AALS, been giving a talk on "Visualizing Data and Results". Apparently, a Vanderbilt Law Review article will also be forthcoming. I attended the talk at AALS, and Epstein focuses on how we can better explain our data and empirical findings. While not a discussion of empirical methodology itself, I think this is an extremely worthwhile endeavor, and we should all think harder about making empirical work more accessible.

The comments to this entry are closed.


April 2021

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30  

Site Meter

Creative Commons License

  • Creative Commons License
Blog powered by Typepad