Is TV bad for kids?
The best way to determine television's impact would be in a laboratory that put some toddlers in front of TV sets and kept some away from it -- an implausible undertaking. But over the past 30 years, economists have developed statistical methods aimed at teasing out cause and effect in complex relationships where controlled experiments aren't possible. That has allowed them to take advantage of "natural experiments," where variations that occur naturally are used in place of the controlled conditions of the laboratory.
Economists have used such methods to look into everything from daylight saving time's effect on energy consumption to how better teeth affect incomes.
The variation Mr. Gentzkow and Mr. Shapiro exploited was the timing of the introduction of TV into different cities. Television began taking off in the U.S. in 1946, after a wartime ban on TV production was lifted. But the Federal Communications Commission stopped granting new commercial television licenses from September 1948 to April 1952 while it made changes in allocating broadcast spectrum. There was a long lag between when some cities got television and when others did.
The economists then looked at results of a survey of 800 U.S. schools that administered tests to 346,662 sixth-grade, ninth-grade and 12th-grade students in 1965. Their finding: Adjusting for differences in household income, parents' educational background and other factors, children who lived in cities that gave them more exposure to television in early childhood performed better on the tests than those with less exposure.
The economists found that television was especially positive for children in households where English wasn't the primary language and parents' education level was lower. "We don't exactly know why that is, but a plausible interpretation is that the effect of television on cognitive development depends on what other kinds of activity television is substituting for," says Mr. Shapiro, 28.