LeTendre and Baker led a team of researchers who analyzed educational data collected in the fourth, eighth and 12th grades in more than 40 countries in 1994, as well as data from an identical study in 50 countries, conducted five years later.
Virtualy wherever they looked, the researchers found no correlation between the average amount of homework assigned in a country and academic achievement. For example, teachers in many countries with the highest scoring students — such as Japan, the Czech Republic and Denmark — gave little homework. At the other end of the spectrum, countries with very low average achievement scores — Thailand, Greece and Iran — have teachers who assign a great deal of homework, Baker noted.
Note that U.S. teachers have been increasing homework amounts, while Japanese teachers have been decreasing it. In neither country do general achievement levels appear to be responding.
That is from Richard Morin’s WP Unconventional Wisdom column, although this installment is not yet on-line (I added the link to the text). Here is a good summary, with more information, it notes that homework may place a special burden on poor families.
We need to be careful about drawing strong inferences from negative results on heterogeneous data, nonetheless this fits my priors. I worry about this more than grade inflation, although I suspect the latter, by making grades less informative, induces overinvestment in extracurricular activities.