I was enthralled by a New York Times article about how the state of Missouri systematically provides judges cost information when they are deciding how to sentence someone convicted of a serious crime. “For someone convicted of endangering the welfare of a child, for instance, a judge might now learn that a three-year prison sentence would run more than $37,000 while probation would cost $6,770.”

All of these costs and guidelines are the result of the Missouri Sentencing Advisory Commission, which aggregates and summarizes the cost of justice.  It also provides other useful information, such as the expected time served as the recidivism rate (see this newsletter for some examples).  The system doesn’t capture indirect costs, such as the societal costs when someone on probation commits another crime.  But overall, this system makes excellent use of math and statistics aimed at making better decisions.  I could pore over the numbers all day, if I had the time.

With more transparent information, judges can make better information.  It’s win-win, isn’t it?

There are some downsides:

• Some critics “say that the cost of punishment is an irrelevant consideration when deciding a criminal’s fate and that there is a risk of overlooking the larger social costs of crime.”  I agree that you can’t put a price on being a victim, but at the end of the day, costs and budgets do matter.  Someone has to pay for putting someone behind bars–it doesn’t come cheap.
• Some judges do not really look at all the cost information.  I suppose we can’t expect everyone to enjoy studying numbers, but on the upside, it does sound like many judges do look at the numbers.
• “To some, the concept sounds crass, and carries the prospect of unwanted consequences. Might a decision between life in prison and a death sentence be decided some day by price comparison? … Could the costs of various sentences become so widely known as to affect the decisions of jurors?”
• “Some voiced concern about the ramifications, the methodology — even the price tag of calculating sentencing price tags.”  This is a sticky issue.  Transparency can uncover inequities in the system, and such a system aimed at cutting costs could lead to additional costs to fix any inequities.

Despite the downsides, I am always a fan of using analytical techniques to make better decisions.  Prison reform is one of my favorite topics, although I am the first to admit that I don’t have any good answers.  I am always looking for better numbers to help me shape my views, and I am glad to hear that the Missouri Sentencing Advisory Commission is helping to provide some of those numbers.  But this story is also a cautionary tale to nerds like me:  it is hard for politicians to acknowledge that we live in a world with limited resources, and they might not get so excited about better numbers and transparency.  Sad, but true.

If you were a juror, would you like to have access to sentencing numbers and statistics?

Related posts:

• How many people get arrested?
• Criminal background checks
• Sense in sentencing
• Criminal justice and the operations researcher

I recently wrote about how OR can be used to determine how employer criminal background checks could be conducted. This is a follow up post.

An excellent article by Carl Bialik in the WSJ summarizes arrest patterns in the United States. It reports that according to a report from the President’s Commission on Law Enforcement and Administration of Justice, 52% of American men will be arrested in their lifetime. This is consistent with the number that Al Blumstein reported in the 1960s.

Both then and now, the researchers who came up with this number were surprised that so many men get arrested. For a law-abiding citizen, it is hard to understand how so many people are arrested. The image from Bialik’s article (displayed below) helps in that regard. Note that the proportion of men that have been arrested (approximately 40%) is lower, since some men have not yet been arrested for the first time yet. The proportion of men who will be arrested in four times higher than the proportion of women that will be arrested.

Bialik explores the 52% claim in greater detail in his blog.

In January, Ralph Keeney’s paper called “Personal Decisions are the Leading Cause of Death” made quite a splash. As the title suggests, the article tackled the issue of how our choices affect our health. The paper asserts that “over one million of the 2.4 million deaths in 2000 can be attributed to personal decisions and could have been avoided if readily available alternative choices were made.”

In the same vein, a Marginal Revolution post summarized violent crime statistics. It seems that our personal choices affect our likelihood of being a victim of violent crime. MR reports the rate of victimization for violent crimes (per 1,000 persons aged 12 and over), summarized below. Married men and women have drastically lower rates of victimization than their unmarried counterparts. Interestingly, divorced men and women have drastically higher rates of victimization than their married counterparts. Even more interesting is that these trends appear to be much stronger for women than men. I’m not sure what to make of all of this. The general trends are not so surprising to me, the the strength of these trends (particularly for the divorced) is troubling. Now I know that correlation doesn’t imply causation, but these overwhelming statistics seem to suggest personal choices play a role.

The statistics:

Never Married Males: 45.0
Married Males: 12.3
Divorced or Separated Males: 44.2

Never Married Females: 38.4
Married Females: 10.3
Divorced or Separated Females: 49.4

What do you make of this?

Links:

• Keeney’s article (published in the Nov/Dec issue of Operations Research)
• OR forum discussing the Keeney article
• Marginal Revolution post
• US DOJ Crime Statistics

Dr. Colleen McCue, president and CEO of MC² Solutions, gave a talk at VCU entitled “What can Amazon and Wal-Mart teach us about fighting crime in a recession?” She talked about using data mining and analytics tools to not just fight crime, but prevent it.

I really enjoyed McCue’s talk, since she  had some really, really cool examples of how she used analytical tools (data mining, statistics, and a little optimization) to fight and prevent crime.  (I am also a CSI jumkie).

For example, McCue used data mining to determine which armed robberies escalated to an aggravated assault. One challenge was that these events are rare (3% escalate) and data is unreliable, since it is mostly provided by eyewitnesses (If you’ve seen CSI or Law & Order once or twice, this should be no surprise). Most of the useful data is in a narrative, which is hard to process in an automated way. Once this challenge was met, the data were used to proactively located police officers in the riskiest areas (areas that were most likely for a robbery to escalate to an aggravated assault), rather than locate police areas with the highest rate of (non-escalating) robberies.  It turned out that one of the riskiest areas was where suburbanites drive into the city to buy drugs (as one police officer put it), where they were then used as “walking ATMs.”  Some resisted being robbed, and were subsequently shot.

McCue used a similar method to develop a risk-based method for locating and scheduling police officers on New Year’s Eve. Her method reduced random gunfire by 47%, increased gun seizures by 246%, and saved $15K. The key was noting that most of the violence in New Year’s Eve occurs in a two hour window around midnight–and being able to convince decision-makers to radically change shift assignments.

In her talk, McCue noted that most of the time, data analysis confirms what is already known. But sometimes, it identifies new things. For example, Wal-Mart was able to identify that before storms hit people stock up on bottled water and strawberry Pop Tarts. Her approach is similar, with McCue’s Pop Tarts being non-intuitive ways to fight crime. After scouring non-acquaintance rape data, for example, McCue was able to determine that prior property crime offenses (not prior assault or peeping Tom offenses) was the most likely indicator for rape.

The most interesting thing I learned about the talk is about how behavioral analysis should be used in Numb3rs-esque ways to fight crime. For example, segmentation uses behavioral analysis, much like marketing, since it, in effect, targets certain populations. It makes no sense to look at homicide, McCue argues. Instead, she looks at drug-related homicide, manslaughter, etc. McCue developed a segmentation tool to determine motive for homicides, in order to help develop a list of suspects early on. Once a list of likely suspects are formed, the perpetrator can be found. Her model only takes data that would be available early in an investigation, since it is critical to make progress on a case (with 48 hours) before it goes cold. Finding likely suspects is one of the most critical parts of an investigation.  Apparently, behavior models are more accurate than models that rely on the “hard data,” since hard data is unreliable whereas behavior is homogeneous. Being a numbers nerd, this makes me uncomfortable (I could barely type that last sentence), but I impressed with the results.

Related posts:

• Criminal justice loves OR
• Murder and the Operations Researcher
• Sense in sentencing

I attended the National Institute of Justice Conference this week, and was happy to see a few sessions that were bent toward operations. (I wrote about OR and criminal justice here and here and last year’s NIJ conference here). The opening plenary session was a panel on prison populations. The panel discussed questions regarding the effectiveness in prisons, since the number of people incarcerated has grown by a factor of six in the last 35 years.

Corrections and Medicaid are the only programs that are a growing part of state spending (as a percentage of all spending). Due to the exploding prison population, sentencing reform is a hot topic. Many states established sentencing commissions in the past two years in order to address sentencing issues such as minimum sentences, truth in sentences, three strikes laws. Some of the issues include:

• Parole populations are skyrocketing as a result of more people being sent to prison and as a result of reduced sentences for non-violent offenders.
• There are usually two extremes in sentencing: no sentence or a prison sentence. Other sentences in the spectrum (such as rehabilitation) are rarely used.
• Before you accuse the panel of being soft on crime, consider that evidence shows that it is not necessary to put all criminals in prison to reduce crime.
• Dora Schriro (from the Arizona Department of Corrections) observed that “good” inmates often become “bad” ex-offenders because prison life is not like life on the outside.

Of course, this is a very political issue. Politics is part of the problem (no politician wants to be soft on crime, resulting in long sentences) and must be part of the solution. Hopefully, OR will also play a role in the solution.

The Supreme Court recently decided to give more leeway and discretion to judges imposing sentences for federal crimes that have mandatory minimum sentences. The sentences for crack cocaine seem particularly harsh to many, but the discretion can be used for other crimes as well.

What does this have to do with OR? Well, minimum sentences not only impose stiff penalties on criminals, but they also result in more people being in jail (see this chart). Al Blumstein has published in the area of criminal justice extensively (see my posts about him here and here) and has argued against minimum sentences. Read his testimony here and read his article in Federal Sentencing Reporter here. In addition to being costly, having so many prisoners have high social costs and unintended consequences. For example, the Chicago Tribune reports that nonviolent drug offenses are overwhelming the criminal justice system. Not a surprise.

If you want to learn more about prisoners and our criminal justice system, check out the fascinating Act V podcast episode of This American Life that details the story of prison inmates performing Hamlet (about an hour), and a 60 Minutes piece about maximum security prisoners earning college degrees.

Because my flight arrived a little late for the INFORMS Annual Meeting, I missed the tutorial entitled “Community-Based Operations Research” by Michael Johnson (U Massuchusetts Boston) and Karen Smilowitz (Northwestern). Fortunately, they have a corresponding paper on the CD that came with conference proceedings, and it is excellent. Beg, borrow or steal this paper!

What is Community-Based OR?
Community-based OR is a subfield of public-sector OR, and it has the following characteristics:

1. It focuses on human stakeholders and is interested in human versus physical resources
2. It focuses on disadvantaged, underserved, or vulnerable populations
3. It focuses on community-level characteristics, such as socio-economic status

One of the key characteristics is that there are multiple stakeholders, and hence, multiple decision makers. These decision makers are often local or are disadvantaged. Looking at tradeoffs in efficiency/effectiveness and equity are a necessity. Johnson and Smilowitz recommend that OR practitioners emphasize models that provide specific, theory-based guidance to local decision makers that can be easily applied to different applications, rather than develop high-level models and then apply them to the specific scenario at hand.

What are Some Examples of Community-Based OR?
Johnson and Smilowitz group community-based OR in four catogories:

1. Human Services: services to senior citizens, humanitarian logistics, public libraries, public education, and family support services.
2. Community Development: housing, community/urban planning, transportation.
3. Public Health and Safety: health care, criminal justice, emergency services, hazardous facility location, and the correlation of chronic diseases and individual deprivation or social externalities.
4. Nonprofit Management: management of community-based service providers.

What is Some Further Reading in Community-Based OR?
The list of references in the paper is extensive, so I will summarize just a few here.

• Urban Operations Research by Larson and Odoni (1981). [Read the book online here]
• Facility Location with Price-Sensitive Demands: Private, Public, and Quasi-Public by
  Donald Erlenkotter [abstract]
• Adventures in policy modeling! Operations research in the community and beyond by Edward H. Kaplan [abstract]
• Modelling Effectiveness-Equity Trade-Offs in Public Service Delivery Systems by
  Marvin B. Mandell [abstract]