Probability of being a serial killer: the case of Lucia de Berk

Here's a post on an extremely interesting story at the intersection of law, statistics and professional ethics in several disciplines. Inter alia, the issue has been covered in some of the best scientific journals. See for instance a). a report in the Science edition of the 16th of November (you need a subscription to access the PDF); b). a very good report on this case in Nature, January this year; you can read here that PDF (subscription free).

What is this about? One of the most interesting legal cases ever, involving multiple homicide, is now under review by the Supreme Court in the Netherlands. The case deals with the former trial of Lucia de Berk, a Dutch nurse, accused and sentenced to life imprisonment in spring 2003, for the alleged murder of seven hospital patients and the attempted murder on three others, in places where she had worked between 1999-2001. The issue is that there was never direct evidence to implicate De Berk (and she has always denied all accusations): she was condemned solely based on the fact that she happened to be always around when these patients died; basically the courts (an appeal court also upheld the initial verdict) decided that it was very unlikely, essentially only one chance in 342 million, according to the expert statistician who testified at the trial, that so many deaths could have occured accidentally while she was nearby.

What happens now is that the 2003 sentence is being challenged by several scientists who signed a petition to re-open the case (this is what the Supreme Court needs to decide on, after last months a justice department panel indeed recommended that the case be reopened). Richard Gill, a Leiden University mathematician and organizer of the petition, states that the previous conclusion, leading to the condemnation of De Berk, is based on "every statistical mistake in the book". Gill and others concluded that the previous statistical testimony was based on an incorrect analysis and that in fact the probability estimated earlier, of 1 in 342 million, is in fact as low as 1 in 48 or even 1 in 5, which are very unlikely to meet the criterion of "beyond reasonable doubt" needed for a criminal conviction. Here's the website of Gill dedicated to this case (with his detailed discussion of the statistical aspects in this case here and Gill's synopsis/reconstruction of the case + other interesting details here). There is even a whole book criticizing De Berk's conviction on scientific procedure, by Ton Derksen , a philosopher of science from the University of Nijmegen. Mark Buchanan, who wrote about the case in Nature (see the link above) summarizes the legal essence of the argument this way: "The court needs to weigh up two different explanations: murder or coincidence. The argument that the deaths were unlikely to have occurred by chance (whether 1 in 48 or 1 in 342 million) is not that meaningful on its own - for instance, the probability that ten murders would occur in the same hospital might be even more unlikely. What matters is the relative likelihood of the two explanations. However, the court was given an estimate for only the first scenario."

This is certainly not an easy case, despite the fact that it is not the first one that might involve wrong statistical evidence in a criminal sentence (the Nature report linked above also mentions another high profile case involving misuse of statistics, the case of Sally Clark, from 1999, in Britain). My opinions are the following. Firstly, I strongly believe that the Dutch Supreme Court has more than sufficient basis to re-open De Berk's case and carefully re-analyse all the previous evidence (I've obviously signed the petition as well). Further, one can only hope that a wrong verdict is overturned swiftly in De Berk's case, should the petitioners be right (and then I wouldn't want to be in the place of the 'expert statistician' who testified to start with, although it is true, as can be read on Gill's discussion, that this expert "always insisted that his analysis only showed that the observed coincidence could not be due to pure chance, not that Lucia caused the the deaths"; moreover, this expert himself wants the case to be reopen; but the scientific flaws would still be there, if the petitioners are right.). More generally, one can only hope that science will be used with the greatest care in any legal processes, especially criminal ones, given the extreme emotions and stakes typically involved (but not only: science has to be done properly, anytime, anywhere, anyway...).

O initiativa de laudat: Orizonturi Deschise

Gabriel Istrate posteaza pe Ad Astra despre noul program de burse pentru masterate si doctorate, in afara, Orizonturi Deschise, al Fundatiei Dinu Patriciu. Dincolo de faptul ca multi il suspecteaza pe Dinu Patriciu de fel de fel de nereguli si foul play-- unde raman neutru, in lipsa oricaror date empirice care sa sustina conjecturile si speculatiile, dar aceasta discutie e in afara scopului acestui post-- asemenea initiative sunt mai mult decat binevenite si de apreciat (si evident au de-a face si cu strategii de imagine si chiar strategii business 100%-- daca realizam ca asemenea sponsorizari se pot deduce din taxe). Ideal, as vedea o competitie in top 100 cei mai bogati romani in asemenea initiative. And then you show that you care...

Daca e ceva in care merita si trebuie investit, in Romania, din toate punctele de vedere, aceasta e educatia si posibilitatea educarii valorilor in centre de (reala) valoare din intreaga lume. Ar mai fi si cercetarea, deci e scop amplu de intreprinderi similare si acolo. In genere, investitii in educatie si cercetare ar fi de fapt de asteptat din partea oricarui antreprenor de succes, la noi. Nu pe baza morala sau de caritate, ci pur si simplu pe baza de strategie pe termen lung.

Evident cele de mai sus sunt scrise presupunand o functionalitate perfecta a mecanismului acordarii de burse, incepand de la selectia strict pe baza de merit (de preferat selectia facuta de comisii specializate, complet neutre) si pana la primirea sumei in timp util (ceea ce nu se intampla de exemplu, ori rareori in timp rezonabil si fara probleme, in cazul burselor oferite de universitatile noastre pentru vizite scurte ale studentilor/masteranzilor/doctoranzilor romani la universitati din afara: lucruri care se stiu foarte bine la noi). Ar mai fi poate de discutat si cuantumul identic al burselor- ar trebui diferentiat in functie de regiunea/universitatea unde candidatul ar fi admis (altfel, sa dau doar un cons, ai avea motivatii sa cauti doar locuri unde cuantumul de 15000 USD pe an ar fi suficient, ori asta ar putea fi suficient in unele tari europene, dar in nici un caz la universitati celebre din state, dar chiar si la cele din Londra, de exemplu); acelasi lucru vis-à-vis de numarul fix de 100 de burse, asta ar trebui variat in functie de numarul candidatilor cu adevarat meritorii, deci intr-un an ai putea avea 50 si in altul 200 etc. In ceea ce priveste conditia de intoarcere in Romania si, in caz ca nu, restituirea integrala a bursei, mi se pare fireasca, dar ar fi minunat ca Fundatia sa faca si un lobby acerb pentru acesti candidati, potential elite la intoarcere, pe piata muncii de la noi: ce nu vrem sa se intample e sa avem someri in Romania (ceea ce e echivalent cu a fi angajat intr-un job cu mult sub potentialul tau) cu, e.g. MBA-uri de la Wharton sau INSEAD. Deci motivatia trebuie urmarita si asigurata pana la capat. Dar, ceteris paribus, o initiativa de toata lauda. Ad majora!

And Nicole says...(plus the song of the day)

Ok Seb, okok! I've got to admit it. You've earned it deserve it u'r da man, I owe you this one! :) [...] Nico, at your service...

(of course I only took the relevant parts from Nicole's corresponding comment to this previous post of mine)

She might not be Paulina Porizkova (and after all I am not Steven Colbert :-)), but I promise you: you would not be ashamed with her :-). This is Nicole (turning into Ms. Turner, above :-)). And I swear I did not even have sex with that woman (unfortunately...yet... allright, allright, I stop right here, you did not read my last phrase), to echo somebody whom we all know.

Oh, wait, let me paste that once more (Colbert forever!) :-). For future reference :-)

Ok Seb, okok! I've got to admit it. You've earned it deserve it u'r da man, I owe you this one! :) [...] Nico, at your service...

Nicole, muchas gracias, I really feel flattered! (should stop though, it starts getting too much to my head already...). And looking forward to reading the first posts on your blog!

PS. And why not, let's make Tina's (and Nicole's) song of the day for today. "Simply the best", once again...

Econlinks for 27-11-'07

• Becker and respectively Posner, on tax evasion. Why is tax compliance higher than one would expect (with a reservation here: it also follows from Posner's argument than in fact tax evasion isn't necessarily lower than rationally expected...) and is that mainly due to the rational fear of punishment or to the taxpayers' feeling of moral duty/ fairness etc? For me Posner wins this argument (they do oppose each other on the essence herein, which after all rarely happens on the Becker-Posner blog...): it is much more the deterrence effect of the fear of punishment plus the costs associated with getting to know how to evade the taxes 'properly', than a feeling of moral duty or fairness, particularly when no other individual is directly involved; after all Posner's got the comparative advantage in this area and it is very very difficult to beat that...

• From now on (until I'll exhaust all notable examples), I'll present within the 'econlinks' posts my favourite examples of 'markets in everything' from Tyler Cowen's recent book, "Discover Your Inner Economist"-- one of my best reads this year (that does not mean that its chapters cannot be ranked: there are very good and also not so good, parts). Check for instance some quotes from it previously taken over for my quote-of-the-week rubrique, here or here. For today about the drinking-and-dialing-problem and equally exotic market solutions to prevent it (and from a more personal perspective, I think there is scope for a drinking-and-emailing-problem resolution as well :-)).

We have all known people who make phone calls when they shouldn't, especially when they are drunk. A survey of 409 people by Virgin Mobile found that 95 percent had made drunk calls, mostly to ex-partners (30 percent), 19 percent to current partners, and 36 percent to others, including their bosses. Fifty-five percent of those people looked at their phones the next morning to see whom they had called-- similarly, someone is waking up in the world this minute and checking to see who it is he or she is sleeping with.
To alleviate the drinking-and-dialing problem, a phone company in Australia started offering customers blocked "blacklist" numbers, which they select before going out to drink. In Japan they sell a mobile phone with a breathanalyzer, to see if you are really fit to drive home, or for that matter to make a phone call. If a bus driver fails the test, his location is sent immediately to his boss by GPS.

• Some wisdom from the one and only Milton Friedman. More actual than ever (in fact I recalled this old interview with Friedman after mentioning the very recent one with Richard Freeman in a previous post and disagreeing with some details in Freeman's address). The most spontaneous and brilliant economist you've ever heard, I promise. So listen and relisten and... relisten. To all of it. And learn. About 30 min clip on YouTube.

• And finally, I will be a millionaire (albeit in DK Kroner, but after all one has to start somewhere) for the next two years. I have been awarded (mange tak!) a prestigious (and generous, academe-wise) independent postdoctoral research grant of The Danish Social Science Research Council (Forskningsrådet for Samfund og Erhverv) for my project "Wages, Productivity and Firm Sizes in Imperfectly Competitive Markets", submitted for the grant applications' contest last August. So yes, you can congratulate me :-). And no, they won't let me buy Belgian beers for all that money. Though I guess that some top wine for research inspiration is allowed :-). Cheers!

Best phrase I've read today

"no one cares about inequality. People care about injustice, unfairness, poverty, sexual predators, family values, gay marriage, terrorism, and many other problems of everyday life. People don't care about Gini distributions."

This is by Herbert Gintis, via Greg Mankiw.

I would add that even when one thinks/is convinced s/he cares about inequality (I know some people that claim that loudly...), that is not direct, it is not unconditional, it is always a cover for something else, e.g. something likely to be on Gintis's list above...

One hour conversation with Richard Freeman (on YouTube)

Shortly after having posted a link to an IZA clip about Richard Freeman among a recent 'econlinks' post, I discovered, via Greg Mankiw, this other link to a very recent (October 31st) whole one hour interview with Freeman on YouTube, within the "Conversations with History" series, having Harry Kreisler as host. Richard Freeman talks, inter alia, about his own career path and decision to become a labour economist, about globalization and its implications, about the 'feminization' of the labour market, about the relationship institutions-market and the role for governmental policies etc. Very clear talk, aimed to a general audience, hence if you want to learn more about all these topics (although the discussion is mostly within a US context, it is generally applicable everywhere), the easiest thing you can do is take some time and watch this video (tip: watch it in more sessions, as I did, that way you don't have to allocate a full hour in one shot...).

Now, inasmuch as my opinion is concerned, I agree 95% with the views of Prof. Freeman expressed in this interview and I think he is one of the current best labour economists able to explain them clearly also to a non-specialist audience. The disagreement arises in some of the details (precisely where a general audience- including his otherwise witty and sympathetic host- would in all likelihood feel lost): namely, a. with respect to Freeman's (seemingly uncritical) support for a raise in the federal minimum wage (see here a sequence of posts I had on that debate), though he stated previously that he strongly favours the Earned Income Tax Credit (EITC)- well, so do I, and I see that that as the best alternative to the raise-the-national-minimum-wage policy- and b. regarding the possible implication of his saying that there is a lot of scope for governmental intervention in 'leaning against the wind' inasmuch (potentially 'heavy') corporate regulation is concerned (suffice to say that US has one of the highest corporate taxes in the world: I am not sure whether Freeman hinted to anything in this regard, but I can hardly see how you can eventually make that even higher or enact similar measures...). With respect to this latter point, maybe we should all remember- and once again give the last word to the one and only Milton Friedman- that "corporation conscience" is not possible (and neither desired...). Come on, join the Milton Friedman choir: "Freedom to choose, says Friedman, or you will lose, says Friedman, freedom is freedom, says Friedman..." :-).

Quote for the week 18th to 24th of Nov '07

During America's experiment with Prohibition, it was common to see the following label on grape juice: Caution: May Ferment into Alcohol. The harvesting of grapes rose dramatically.

Tyler Cowen, "Discover your inner economist"

Read the quote of the previous week.

Econlinks for 23-11-'07

• Remember my post about names in Denmark? At that time I could not find which were really the most frequently used first names. But Denmark Statistics even has those available. So "Jens" (male) and "Anne" (female) are the most common first names in Denmark, to date. In terms of recents trends, the most popular names given to newborn children in 2006 were "Lucas" and "Sofie", respectively. PS: read also some relevant comments to an older post of Dan, with several interesting links concerning (though some of that is about futuristic trends:-)) names elsewhere in the world.

• Here's a short video about one of the most famous labor economists of the moment, Richard Freeman. (wmv file, about 55 MB). The clip is with relationship to his winning this year's IZA Prize (after he won a bunch of other relevant prizes during the years). IZA award statement, Freeman's website at Harvard (or the one at NBER). And yes, the hat is part of Richard Freeman's personality (by the way, did you know that before JFK all US Presidents had hats? I didn't. Freeman makes that clear in the video linked above:-)).

• WineSpectator's top 100 wines for 2007 is now also available (PDF). See also my previous post about the winner and top 10 of this year. I am happy to say that, at a first glance (did not go thoroughly through the list yet) I have tried at least 7 wines in top 50 (but none in top 10). Among those, number 26, "Marchesi de’ Frescobaldi, Brunello di Montalcino Castelgiocondo, Ripe al Convento Riserva 2001" is the one I prefer (2001 is by far the best year for that wine). Try it out. And if you really want the sublime experience, do this while indulging in an Italian classic dish: "lepre in salmì".

• Yes, Bloomberg goes ahead with the "cheap talk for better grades" plan (see my previous post on this) in New York, which is partly due to the initiative of Harvard's Roland Fryer. Excellent, let's see if incentives of this kind elicit better students' outcomes. Ex ante it is not clear how this will work, hence the more interest around these types of studies ( for instance, in a different but related context, recent research by Edwin Leuven, Hessel Oosterbeek and Bas van der Klaauw at SCHOLAR, University of Amsterdam & Tinbergen Institute, suggests that financial incentives do not work as intended (at least not for everybody), ie. extrinsic rewards might be detrimental for intrinsic motivation).

Replication in Economics

Here's my most interesting read in the last weekend, a recent article by Dan Hamermesh, published in the Canadian Journal of Economics, on "replication in economics". It is also downloadable as PDF from Hamermesh's site (Dan Hamermesh's website contains much more information, potentially useful for any economist, albeit junior or senior-- that as a remark for those of you who did not know about this excellent online resource...). But going back to Hamermesh's viewpoint in this context: the short article linked above is extremely interesting and informative. And it contains also a proposal ("modest proposal" in the text), which I also support as very welcome: the only way to see more replication studies done is to have them commissioned (to senior researchers, with tenure...) by journal editors.

While I'd advise any empirical economist (or better: any economist that is currently/ plans in the future to also do some empirical work) to read the paper, I'll put forward below some possible omissions in Prof. Hamermesh's article, which would have been particularly interesting for me:

a. What to do about registered (say, matched employer-employee) datasets that cannot be made publicly available due to confidentiality agreements (though clear steps for accessing them can be always provided as information; in practice, some of these might take too long etc. for the replication to be worth undertaking); the use of such datasets is increasing day by day, particularly within applied microeconometrics, hence I'd say they would merit some separate discussion. Caveat lector: although their case is somehow similar to that of any proprietary database, these administrative datasets do have one common characteristic, namely that usually they are provided by the official statistics bureaus of individual countries. This might suggest possible (ad-hoc) agreements between journals and such bureaus to allow replication of studies under the same very strict confidentiality rules as those the initial author had been subjected to.

b. There is no discussion in Hamermesh's paper about structural approaches: in those cases replication, as commonly understood (particularly "scientific replication" in Hamermesh's terminology), is typically not an issue (subject to correct coding and analysis, to start with). In fact, one way to avoid frequent requests for replication using other datasets and other time periods etc, would be to have a structural framework to start with...

Quote for week 11th-17th of Nov '07

How can my muse want subject to invent,
While thou dost breathe, that pour'st into my verse
Thine own sweet argument, too excellent
For every vulgar paper to rehearse?
O! give thy self the thanks, if aught in me
Worthy perusal stand against thy sight;
For who's so dumb that cannot write to thee,
When thou thy self dost give invention light?
Be thou the tenth Muse, ten times more in worth
Than those old nine which rhymers invocate;
And he that calls on thee, let him bring forth
Eternal numbers to outlive long date.
If my slight muse do please these curious days,
The pain be mine, but thine shall be the praise.

William Shakespeare, Sonnet XXXVIII

See quote for the previous week.