Age, Generation, or History?

March 23, 2010
Posted by Jay Livingston

On questions of interracial dating and homosexuality, older people are generally more conservative than younger people. But is that because people get more conservative as they age, or is it because different generations have different ideas? (The aging effect is also known as “life-cycle”; the generation effect is also referred to as “cohort.”)

When I blogged this last May (here) in connection with a story about segregated high school proms, I forgot to add a third factor – the effects of historical change. Regardless of age or generation, people who experience the same historical changes – the Great Depression, the Civil Rights movement, Vietnam, 9/11 – may be affected in a similar way whether they are in their teens or their seventies.

In that post, I wondered whether today’s kids will get more conservative as they age. Or will they retain the attitudes of their generation? The Pew Foundation has some relevant information. Its report on “Millenials,” has a graph showing changes in attitudes towards interracial dating in each of four cohorts.


Each generation is more accepting of interracial dating than are those that came before. And each generation itself becomes more liberal over time. As the report’s authors, Scott Keeter and Paul Taylor, say, the upward trend of all lines is probably not part of some general liberalization over the life cycle; it is almost certainly a period effect. When it comes to interracial dating, we’ve all become liberals.

(I don’t know what happened in 1991 to make the older generations become suddenly more accepting of interracial dating. A change in the wording of the question? A change in the formula for choosing the sample? Or was it a real change in attitude?)

Hat tip: Lisa at Sociological Images.

Accounting for March Madness

March 21, 2010
Posted by Jay Livingston

The powers at SUNY Binghamton tried to sell their academic soul to the devil. All they asked in return was a good basketball team. I wondered aloud (here) if, in these hard times of budget cutbacks, they might have thought a top team would bring in the big bucks, and I added my opinion that if that was their view, they were shooting from well beyond the three-point line.

But a story by Chris Isidore at CNN supports the idea that college hoops are a good investment.

it’s clear that men’s basketball is a major source of funding for many colleges, and that profits are still far more common than losses for the major teams in March Madness.

Isidore gives the basic accounting for 342 Division I teams.

• Total revenues $1.08 billion
• Total expenses $796 million.
• The bottom line: a profit of $281 million. (That’s either 26% of revenues, or a 35% return on the expenses.)

The CNN article has the complete list. In the graph below, I’ve grouped the schools according to net profit as reported in the article.

(Click on the graph for a larger view.)

Clearly, the winners outnumber the losers. Even SUNY Binghamton is in the black, with a profit of $29,000 on total revues of $1.6 million, about 1.8%.

But what’s up with that very large number of teams that broke exactly even? It suggests that we need to take a closer look at the Revenue column.

Victor Matheson of The Sports Economist and Holy Cross looked at the books at his own school, which had expenses of about $1.5 million and, miraculously, revenues of exactly the same amount. It turns out that $1 million of these “revenues”

came from direct institutional support. The team didn't break even. It lost about $1 million.

The same is probably true of many of these break-even teams. Their “revenues” are what the school shells out to cover their expenses. In this sense, the Sociology Department at Montclair State is a break-even unit.*  Our revenues (entirely in the form of money from the University) exactly matched out expenses.  We do not know just how far to the red-ink side of the graph those break-even schools really are, but it’s clear that the CNN reporter is wrong.  Losses are more common that profits.

Even some of the profitable schools may be using Enron-inspired accounting measures. UNC Chapel Hill, whose $12.3 million basketball profit is second only to Louisville,** counts only revenues from ticket sales, TV, and other legit sources. But on the expense side, as Matheson notes,

the University allocated exactly zero dollars in expenses to the basketball team for things like medical trainers, facilities and maintenence, promotion, or indirect institutional support. It’s pretty easy to have a profitable basketball team when all of your revenues count towards the bottom line but many of your expenses don’t.

* I’m going to suggest to our president that if the department could just sign a good power forward, we might actually turn a small profit.

** FWIW, UNC didn’t make it to the 65-team NCAA draw, and Louisville lost in the first round.

Australia and Capital Punishment

March 19, 2010
Posted by Jay Livingston

Australia killed the death penalty. Nobody had been executed in Australia since 1967, but a bill passed last week banned it permanently for all states and territories. It banned torture as well.
(A news story is here.)

In the US, the death penalty debate, when it does get to empirical questions, usually focuses on whether the death penalty has any impact on crime. The effect is probably stronger the other way round. Crime rates push the death penalty. In the US, crime rates began to climb steadily beginning in the early 1960s. Support for the death penalty started increasing about two years later. Crime rates started to decrease about 1991 and continued to fall dramatically. Starting in 1993, support for the death penalty declined from 80% to 65%.

Australia too has seen a decrease in murder.

(Click on the graph for a larger view.)

For some reason the chief government publication gives numbers not rates. Had it used rates, the drop would be somewhat more pronounced since the population during the period increased from about 17 million to about 21 million.

Other crimes have decreased as well.


At the risk of having the gunslingers descend again, I will mention that at one point during this period, Australia passed a stringent gun control law, chiefly in response to a mass shooting, the Port Arthur Massacre. Assault weapons were banned, the government bought back and then destroyed 650,000 guns, and stricter licensing and registration were required. If you don’t know when this happened, look at the graphs and see if you can guess which year.

(For the answer, and much more information on Australia’s gun laws, go here.)

A Shot of Ethnography

March 17, 2010
Posted by Jay Livingston

(Although this post is partly about drunkenness, its appearing on St. Patrick’s Day is purely coincidental and should not be construed in any way as related to ethnic stereotypes.)

I miss ethnography, an idle pursuit in my academic youth. Survey research results are more frequently cited, in the journals and even in the newspapers (as Contexts Crawler faithfully documents), maybe even a little snazzy graph in USA Today or elsewhere. Ethnographers less so it seems, though they might occasionally turn up on TV.* Or in a New York Times column.

Contexts Crawler missed it, but on Monday, “Our Towns,” by Peter Applebome reported on the ethnographic fieldwork of James Roberts (Sociology and Criminal Justice department, University of Scranton).

The field for his fieldwork was the bars of Hoboken. Dr. Roberts is a Jersey boy – degrees from Stockton and Rutgers – and he has tended bar down the shore. I haven’t read his work, but from the Times column I gather that it focuses on questions of how bar patrons become excessively drunk and violent. Who is responsible for feeding even more drinks to people who are far beyond three sheets to the wind? Not the bartenders, it turns out. I would also guess that Roberts is watching to see how some interactions escalate to violence, perhaps along lines of Luckenbill’s old research on scenarios that end in homicide.

Survey research shows the relation between variables. Ethnography tells you how things work. Ethnography is about knowing who the players are and how they think. I remember Robert Weiss saying that if you’re a survey researcher and you want to know about cars, you get a sample of cars, and you discover that a car has an average of 5.38 cylinders, 164.7 horsepower, etc. (this was so long ago that he also included something about carburetors). But if you’re an ethnographer, you get a car, you open the hood, and you try to figure out how all those parts fit together.

There are other differences, notably control of the data and the demands that the data make on the researcher. You can’t do ethnography on your own terms. If you want to do research on drunkenness in Hoboken bars, you have to go to Hoboken, even if you live in Scranton. And you have to do your research when people are going to be getting drunk, even if you usually go to bed after the 11:00 news.

*I myself was once on a morning show called “For Women Only” or maybe its later incarnation “Not For Women Only,” both distant ancestors of The View.