‘Dangerous UCLA’ and other problematic rankings

How much should you believe a new survey that calls UCLA the most dangerous university in the country? About as much as you believe all those eye-catching lists that rank the best place to live, or the most pedestrian-friendly place to live or -- you get the point.

Rankings almost inevitably leave out key information or inflate other data. The results aren’t necessarily invalid, they’re just not necessarily useful either.

In the case of the “danger” rankings by Business Insider, the data have a certain, but limited, validity. The publication examined FBI data of violent and property crimes reported over four years; in measuring danger, it gave more weight to violent crimes. Fair enough. But it only examined crimes for large universities with their own police departments. (And its timing, through 2011, left out some tragic events involving USC students.) It didn’t examine whether those reports were considered to be true or whether they occurred on campus, outraged university officials said.

Yes, but -- since very few students stay strictly on campus, isn’t what happens to them in the surrounding neighborhood relevant? When parents send their offspring to school, you can bet they have an eye on the nearby community. And there’s little reason to believe that one large university would be more likely to receive false reports than another. There could, however, be differences in what sorts of reports different campus police departments might take. If one department will take reports of relatively minor thefts and another won’t, that could result in big differences.

Some of the criticisms of the Business Insider report have been more off base than the report itself. No, it did not leave out private universities, unless MIT recently switched categories. And it didn’t rank them just by the numbers of crimes; it appears to have developed a crime rate for each one -- though it doesn’t report those rates, just raw numbers. Still, you can see that some universities with lower numbers of actual crimes are rated more dangerous -- they’re the smaller of the large universities.

In truth, the universities are all pretty safe, even if you look at the supposed worst of them. So it’s unclear why these numbers have any meaning anyway.

It’s similar to many rankings of “most livable cities.” They leave out factors that are important to families -- yes, maybe it has jobs, but does it have jobs in your field? -- or weigh factors such as pedestrian-friendliness the same as the quality of the schools, items that mean different things to different people. Even more problematic are the ratings done at the local level -- say, the best city in the Bay Area or San Diego County. Among the factors commonly taken into account are how many jobs the city has, how much shopping and how rich the cultural offerings are. But the town next door might have better schools and lower prices, yet there’s nothing to keep people in that town from taking advantage of the shopping, culture and jobs next door.

Rankings can be fun. They make headlines. But they’re seldom worth taking seriously -- and pretty much never useful as the basis for an important life decision.

ALSO:

HIV testing for all

Give charter schools their due

No more curtain calls for elephants