Random Thoughts…

… on averages, policing, and recidivism.

Averages. In all too many cases we see people basing their assessments on averages, even when the groups being averaged over are very heterogeneous. To me, that’s equivalent to ordering a dinner of onion soup, filet mignon, Caesar salad, a fine Bordeaux wine, and peach Melba – but instead of eating the courses separately, putting it all in a blender and eating it that way – after all, it goes into the same stomach, doesn’t it?

[Francis Galton criticized those who focused on averages as being “as dull to the charm of variety as that of the native of one of our flat English counties, whose retrospect of Switzerland was that, if its mountains could be thrown into its lakes, two nuisances would be got rid of at once.”]

– – – – – –

Policing. After my first few police ride-alongs in the mid-sixties I reached the conclusion, jokingly, that the police were report writers with the power of arrest. An exaggeration, of course, but it appeared to me that most of their activity seemed to be documenting the nature of their interaction with citizens.

Rather than look upon that as a joke, we can turn that around to some extent and realize that they are the only 24-hour-a-day representatives of the city on the street. And as such, they should note the deficiencies, not just in the people they meet, but in the city services that should be (but are often not) provided to the areas of most concern.

  • Kids congregating on street corners? Document the lack of athletic facilities and parks
  • Trash on, and potholes in, the streets? Notify the appropriate city agencies.
  • Truancy? How much is invested in the neighborhood schools?

I don’t mean to imply that police officers should all be renaissance men/women, but that they should be aware that the problems they encounter are not for them to solve using their police power. In other words, the police should be tasked with not only dealing with problem persons but dealing with problems that are exacerbated by municipal deficiencies. This would make the “thin blue line” a little bit stronger and more substantial.

– – – – –

Recidivism. Over thirty years I published a book on recidivism (called, appropriately enough, Recidivism). It focused, for the most part, on the way recidivism was measured at the time. In that era the effectiveness of a correctional program was evaluated based on the “one-year recidivism rate,” i.e., the fraction of those enrolled in the program who “failed” (choose a meaning: who were rearrested, or retried, or reconvicted, or whose probation or parole conditions were violated). That is, the status of the cohort at a single point in time was the criterion. So two programs with the same one-year recidivism rate were judged to be equally effective, which is a lousy measure. [1]

In our first paper on the topic, my colleague Dick McCleary (currently at UC Irvine) and I pointed out that a better measure would take into account the entire trajectory, not just one point in time. It was based on statistics originally developed in the biomedical field, for evaluating different treatment regimes for cancer patients.[2] In that field, rather than use the term “failure” for those who did not improve, the term “survival,” its complement, was used to denote those who did not fail. In other words, the flip side of the coin is considered as the important variable.

Now with a few more years under my belt, and a little more perspective, I realize that we should not only follow the lead of biostatisticians in the methodology to use, but also in the perspective we bring to evaluating correctional programs. Specifically, we should look at survival rather than failure as the more important consideration.

Words have impact, and these words promote different views of the context in which a failure may occur. When we say a person recidivates, we frame the situation using an active verb, as a deliberate action on the part of the offender. That is, we ascribe to the offender a willfulness to do bad. On the other hand, when we talk about correctional failure from the standpoint of survival analysis, we frame the situation as an action on the part of the offender due to his/her inability to withstand the pressures preventing rehabilitation.

Other correctional programs do not ascribe such willfulness to failure. As Alan Marlatt noted in his book Relapse Prevention, we realize that it may take a few tries before a person quits smoking, or drinking, or drugs. In these situations we say a person has thus far survived, and that s/he has not yet relapsed.

Obviously, there are people out there who are truly beyond correction and who, regardless of the circumstances, will continue to reoffend. But this is far from the norm, and tars every other person released from incarceration with the same brush.


[1] Suppose, for example, that both had one-year recidivism rates of 60 percent. In one program, 5 percent of the cohort failed every month; in the other program, all 60 percent failed in the first month, with no subsequent failures in the next 11 months. Obviously, they should not be considered to be equally effective. While this is an extreme example, it points out that the entire trajectory of failure should be considered when evaluating programs. That is what the book focused on, and it helped to change the way recidivism was measured.

[2] We thought that we had developed this completely new way of assessing correctional failure, but found out later that others had gotten there first. Sic transit gloria mundi.

Unpublished Op-Ed

Mark Kleiman and I wrote this in February 2017, but never had it published. I thought that it might be worth posting at this time.

Some Words of Advice for Federal Employees

Receiving directives inconsistent with good government – if not worse – creates one of the most difficult situations a civil servant can face. As former Justice Department staffers, we have some advice to offer Federal employees when such situations arise, as they seem likely to do often under the current regime.

1.       When told to implement a policy that is counter to statute, regulation, or the stated and authorized goals of the agency, take good notes; such directives rarely come in writing. Then go back to your office and write down your understanding of the recommended policy, making sure you have correctly described what you were told. Then send that account as a memo to your superior.

2.       Whether or not you receive a reply, follow up with a detailed list of issues and concerns, both pro and con, involved with proposed policy or action. Describe them in full context and cite the relevant legislation, executive orders, and constitutional issues. Send that, too, up the chain of command.

3.       You may also be at the receiving end of threats or other problematic situations that are meant to intimidate you. Write a memo to yourself and share it with a trusted friend as soon as possible, to establish a time line.

4.       Do not use your office phone or computer (or cell phone while in the office) for personal reasons, least of all to complain about these situations, as this may open you up to attack. If your agency expects you to be available for phone calls and text messages around the clock, get a cell phone that you use only for official business. You might want to use a text messaging app that encrypts the message, and ask your recipients to do the same.

5. Maintain a contemporaneous, written log on a ruled ledger with a sewn binding, so removal of any page will show. Enter every meeting, call, and significant email on successive lines in ink, leaving no spaces. Fill in any space on the right with a slash, so nothing can be added. Note the date, time, attendees, subject, and conclusions. Absent minutes, no one else will remember what happened a day later, so your record will become dispositive. This approach, laborious though it is, can provide valuable protection for anyone from a GS-1 to a cabinet officer.

6. If you decide to talk to a reporter, get the ground rules clear first. “On background” means you can’t be identified, but your agency can; “deep background” means that even your agency isn’t mentioned.  Any communication to the press about official business not previously cleared by your agency’s public information office will probably put you out of bounds; consider whether you’re willing to take the consequences. If you’re later asked about whether you were the source of a story, either tell the truth (and be prepared to find a new job) or refuse to answer.

There are already reports that White House Chief Strategist Steve Bannon has taken steps to erase the paper trail behind various Executive Orders. All the more reason for career civil servants and the political appointees more loyal to the country than to the ruling cabal to make as much of a record as possible.

Michael Maltz is Emeritus Professor of Criminal Justice and of Information & Decision Sciences at the University of Illinois at Chicago. He was a research analyst with the National Institute of Law Enforcement and Criminal Justice during the Nixon administration and had to deal with some questionable directives.

Mark Kleiman was Professor of Public Policy at the New York University Marron Institute of Urban Management. He served as Director of Policy and Management Analysis for the Criminal Division in the Carter and Reagan Administrations, never receiving an improper order.

Fear of Crime and Information Theory

An appropriate lead-in to this post is the quote attributed to Stalin: “A single death is a tragedy; a million deaths is a statistic.” This fits very nicely into the structure of information theory, as formulated by Claude Shannon sixty years ago. His measure of information is equivalent to the reciprocal of the probability of occurrence of an event (specifically, the logarithm of this number). That is, if the probability of an occurrence is 100% (1.0), that is, if it’s certain to occur, then its information content is the logarithm[1] of 1, or 0; if the probability is 10%, its information content is 1; if its probability is .001, its information content is 3. In other words, the less likely the event is to occur, the higher the information content of its occurrence. That’s why it’s also called a measure of “surprise,” because the occurrence of a highly unlikely event is more surprising than one that is highly likely to occur.

What, pray tell, does this have to do with fear of crime? A lot, it turns out. As Steven Pinker has shown, violence and violent death has declined markedly over the past few centuries, and we are much, much less likely than our ancestors to die at the hands of others, either through wars or by crimes of violence. But this very fact means that (per the quote at the top of this post) each violent death nowadays has a greater surprise value than it did in the past. And that greater surprise value translates directly into greater fear – fear that oneself or one’s family is going to be harmed by others.

So we have this paradox: the safer we make ourselves, the more fear we have – of the unknown, of “them” (any outsider), of MS-13, of the person walking toward you (“Quick, get out your gun before he gets his out”). And of course, it is all so easy to stir up fear in a population, especially when those in power, whom we expect to be responsible adults, are the ones stirring it up.

‘Nuff said.

[1] To make it easy to follow, I’m using logs to the base 10. For those who slept through math class, the logarithm of a number goes up much more slowly than the number itself, so in the examples above the log of 1 is 0, of 10 (the reciprocal of 1/10) is 1, and of 1000 is 3. End of lesson.

Recycling and Packaging

One area of modern life that I think needs to be addressed in greater depth is the way packaging (and its attendant stress on landfill) has increased over the past few decades. When a package consists of paper, mylar, plastic, and other substances glued together, how in hell does a recycler deal with that? Since it’s a worldwide concern, are there other countries that deal with it in better ways than we do? Is there a way of incorporating the cost of (near-impossible) recycling this kind of packaging into the equation?

And more generally, what kind of research is being done to create packaging that is more amenable to recycling? I remember seeing something about using fungi (mushrooms or other mycological substances) for packaging. Of course, this material is not transparent, but if the rest of the package can be recycled it’s a start.

Stolpersteine

I have only seen the National Memorial for Peace and Justice on television and the web. As you probably know, it is a memorial to the over four thousand African Americans who were killed between 1877 and 1950 for the crime of having dark skins. It is certainly overpowering to see the stones representing every single victim whose death could be documented.

Last summer, while walking the streets of Würzburg, Germany, we saw a more prosaic, but just as heart-gripping, memorial to victims of a different Holocaust. Gunter Demnig, a German artist, created Stolpersteine (stumbling blocks), brass plaques cemented on cobblestones. The plaques are engraved with the details of the victims – name, birthdate, date of death, location (concentration camp) of death – who might be Jews, Roma, homosexuals, or mentally or physically handicapped persons. The stumbling blocks are set into the pavement in front of the former homes of the victims, so a stroll down the street is a reminder of what and where it happened.

While a trip to the various Holocaust Museums leaves a person with a profound feeling of rage at those who perpetrated these crimes against humanity, it may be forgotten with the passage of time and a return to one’s daily life. I have the feeling that the Stolpersteine may have a more subtle effect on people, in their pervasiveness throughout the cities where they were installed. I wonder if Demnig or someone else might be encouraged to do the same with those four thousand victims of racism, showing how pervasive and widespread it was throughout this country.

On Social Networks

John Kasich was elected governor in Ohio in 2010 as a strong Tea Party advocate. One of his first legislative campaigns in 2011 (Senate Bill 5) was to restrict collective bargaining for public employees: police, firefighters, and teachers. After it passed the hue and cry was huge: before the year was out a referendum put its repeal on the ballot, where it was soundly rejected – and since then Kasich has been a more moderate governor.

From 2002 to 2012 I spent a lot of time in Columbus, Ohio, and played handball at an athletic club there, with mostly Republican members. One of the regulars there was a retired state policeman who was on Kasich’s security detail. I remember him saying to us, “We told him, don’t go after the police and fire, just the teachers,” because he assumed that it would be an easy win to focus on a mostly female profession.

This is no longer the case. The strikes in West Virginia, Oklahoma, and Kentucky, coupled with the Parkland students’ activism, make me think about how social media has changed the way people organize – and that unions may be strengthened (or even superseded) by social networking, Facebook, and tweets. When a union calls a strike, it’s often a top-down decision. True, the leadership polls its membership to make that decision, but then it issues a proclamation. With social media involved in strikes it’s based on networking, which to my mind is a much more powerful way to rally support.

An additional note: it seems to be going worldwide. Today’s NY Times has articles about the Dalit (formerly “untouchables”) in India and physicians in Togo using social media to push for change. While we may deplore its use by Cambridge Analytica to promote lies and influence elections, it can also be used to foster positive change.

McArdle on Denmark

My cousin alerted me to a post Megan McArdle wrote about Denmark and Danes, about trust and what we could learn from them. I have a slightly different take, although it’s from over fifty years ago. But I think that my perspective still has some validity.

In September 1963, with my newly minted PhD, I accepted a one-year postdoc position at the Technical University of Denmark. It was a time of ferment in the US, especially after the Kennedy assassination that November, and Johnson’s pushing for action on civil rights. A lot of racist bile cropped up in the media and was published in Denmark as well. My colleagues Gunnar and Erling were constantly on me about how terrible we treated Negroes in our country, implying that such a thing would never happen in Denmark.

I, of course, tried to explain that, yes, it was terrible but that we were working on it — as the 1964 Civil Rights Act subsequently showed. But before that passed, I brought up to them something that I noticed locally, that I hope they could explain: why were all of the menial workers, street sweepers and the like, apparently Greenlanders (recognizable due to their Inuit descent rather than northern European descent)? After I brought that up I never again heard about the mote in our eye. Yes, I’m sure we can learn a lot from Danes about mutual trust, but let’s be a bit moderate with our praise.

The NY Times Posted a Lott of Crap

So John Lott is promoting guns again, this time in an op-ed piece in yesterday’s New York Times. But this time he’s taking a different tack. Some years ago Lott maintained that a survey he conducted on defensive gun use showed its benefits. However, no one could check it; he said that he lost the data in a hard drive crash – but he couldn’t even provide evidence that he hired and paid interviewers to perform the survey.

He also used published crime statistics to promote his idea that relaxed gun laws prevented homicide. He subsequently was found to have misused the statistics in, shall we say, “innovative” ways, to “prove” that more guns leads to less crime.

Abandoning data and surveys to promote guns, this time he uses a couple of anecdotes relating to individuals. That is, one person who was improperly denied a concealed carry license is more salient to him than the deaths of dozens of schoolchildren across the country.

I first encountered Lott when he began to use crime data improperly and wrote to him explaining the issues. When he did nothing about it, I wrote an article criticizing his research. To counteract my criticism, a woman named Mary Rosh started appearing on the web, who vilified me and who praised Lott as one of the best teachers she ever had. Then it turned out that Mary Rosh was a fiction, a persona created by Lott to debunk his critics. [He even implicated his four children: he admitted that the name “Mary Rosh” was cobbled together using the first two letters of his kids’ names.] In other words, he hid behind the skirts of a woman he created out of whole cloth, just to promote himself and his pro-gun ideology. Here is my take on his actions in 2003.

At the time Lott was a resident scholar at the American Enterprise Institute, an organization with which he is no longer affiliated – which makes me look more kindly on AEI. Now he hangs his at the Crime Prevention Research Center, where he is president. I have no idea who funds this center, but I can guess. Those who want to learn more about the organization and Lott should read this article.

I realize that the New York Times is trying to do its best to look at both sides of controversial policies, but this really takes the cake. To publish a person who admitted to lying about his professional life, and who is writing about the policies he lied about, is offensive to me and should be to all those who look upon the Times as a credible source of information.

Bleg on Guns

Some years ago (can it really be 14 years?!) I guest-posted on this website a screed on John Lott, whose integrity, shall we say, leaves a lot of room for improvement. In that same time period I recall seeing a cartoon (it might have been in the NY Times) which had everyone in the street, including babies in their carriages, packing guns. Do any of you remember it, have the URL for it, or have a scan/copy of it? It unfortunately seems to be appropriate once again, with kids shooting kids at an unimaginable pace. If you have copied it but can’t post it here, please send it to me at my gmail address, maltzmd.