Wednesday, October 26, 2011

How to judge campus safety?

A few days ago I was emailed a pdf document: the 2011 Annual Security Report for George Mason University. As mandated by the Jeanne Clery Disclosure of Campus Security Policy and Campus Crime Statistics Act (yeah I hadn't heard of it before either) it provides a breakdown of all criminal activity which occurred on campus, by year, and with special columns for "Hate Crimes." The picture I attached has the numbers for Fairfax. This is the most interesting part of the document to me because it contains some raw figures on different offenses committed in the campus I attend. Statistics for the other George Mason campuses (Arlington, Prince William, Loudoun, etc.) are also available but are a lot less edifying, because the columns have just a bunch of zeroes. Coincidentally, Fairfax also happens to be the only campus with attached undergraduate housing -- make of it what you will.

The most exciting table I've seen since breakfast.
This report is obviously intended to increase public awareness about crime rates on campus, allowing potential students and their parents to make an informed decision when comparing different universities. What I wonder is, how does someone look at this report and get any sense of the probability that they themselves will be victimized? This blog post is a rough attempt at answering that question.

Some useful figures to get started with:

So, to get a quick back-of-the-envelope calculation on risk of a certain crime you could take the number reported above (averaging across the three years to be fancy), multiply by three for property crimes or two for violent crimes, and then divide by 30,000. That would give the raw chance for any one individual becoming a victim of that crime. For example, burglary: 20.67 burglaries were reported per year, and if that's only 35% of the total we expect the true number to be about 62. Divide that by 30,000 and you find a 0.2% chance of being burglarized at GMU in a given year.

If a student was wondering about the chance of being caught for drinking on campus, they could perform a similar exercise. An average of 647 students get referrals or arrests for alcohol per year. From a resident student population of 7,000 we might assume 75% are underage (if everyone starts college at age 18, and stays 4 years) and nationally, 82% of underage college students report drinking in the last year. Taking one extreme, if we assume that all students caught drinking are underage, the chance of being caught for drinking is about 15% as an upper bound. If instead we assume that students above the legal age may also be arrested or referred for drinking, and that everyone drinks (an extreme belief indeed) then the chance of being caught drinking in a year is 9%.

If caught, about 40% are arrested if caught on campus, with the rest getting referrals. Of course, common sense would indicate that acting drunk and stupid in public can sway those probabilities in a dramatic fashion. It's also worth noting that students caught for drinking in the student residences were much more likely to get referrals, rather than being arrested.

In other respects, these figures are less helpful. Looking at "Nonforcible Sex Offenses" there are none reported in any of the last three years. While I would like to say this proves that GMU students are paragons of virtue and would never take sexual advantage of each other, it is more likely that these offenses just go unreported. How many such crimes occur at any college is a hotly debated question, which is important to address but very hard to answer conclusively.

Overall, I'd say that GMU's Annual Security Report paints a fairly pleasant picture of crime risks on campus, and justifiably so -- anecdotally, the George Mason area and Fairfax in general are probably the safest places I've lived yet. Compared to my undergraduate institution, Gonzaga University, which had a rash of stick-up robberies by two teenage thugs with shotguns during one year, GMU-Fairfax seems pretty rosy. It's no reason for complacency, but safety-conscious parents or students should take some comfort in knowing that their lives and property are pretty well-protected when attending George Mason.

Thursday, October 6, 2011

Shirking, Malingering, and other Unpopular Terms regarding the American labor force

So, the title might be overly sensational.

For an outside observer, finding measures of low conscientiousness on the job (shirking) or active deception to evade work (malingering) is a bit of a challenge, because workers have an incentive to conceal that sort of activity.

This rough study addresses worker injury on the job and attempts to determine whether outside incentives motivate changes in sick time and injury rates. Using some unsophisticated econometric techniques, surprising results are found. Surprising if you think unemployment levels and interest rates will not influence worker absences, anyway.

The paper:

Incentives for Health and Safety: Do Economic Conditions Influence Worker Absences?

Click the link to access the paper, in PDF form, off Google Documents.

Bonus material:

Malingering and/or exaggerating a real injury to avoid work is not a new phenomenon. Here are some illustrative anecdotes, documented roughly a century ago, which didn't quite make it into the paper.

Case I. W., aged 54, a labourer, met with an accident in February, 1907. He was crushed between a wagon and wall. He returned to light work eighteen months later, but after three months stopped, as he said work was causing him vague indefinite pains in the abdomen, especially on stooping, and that work made him sweat. On examination there was marked abdominal obesity due to his recent easy life, and the waist band of his trousers was too tight, so that the three upper buttons could not fasten. No other abnormality existed. This man could not be persuaded to work till the court stopped his compensation. He has worked well since and reduced his weight.

Case II. W., aged 26, a collier, was said to have sprained his back lifting a tub of coal. When examined fourteen months later he localized pain in a certain spot in the gluteal region; the spot was apparently excessively tender, but when his attention was withdrawn he did not notice the pressure. He also varied the locality of the tender spot. No other symptoms were present. He had bluffed the insurance representative for months by the vivid description of an injured spine, whereas there was no injury to the spine or lumbar muscles.[1]

[1] Gullan, Gordon. (1912, August) Medical Training For The Detection Of Malingering. The British Medical Journal. p. 222