A pollster can oversample on age, race, gender and income as each of these typically stays the same throughout a person’s life and can be pegged to census data.
However, political party affiliation changes from year to to year and even from day to day. With that, political affiliation cannot in itself be a quota cell and cannot be an instance of over sampling.
“A pollster can oversample on age, race, gender and income”
Saw a white paper the other day analyzing different poll methodologies in relation to the 2012 election. One of the conclusions:
“Demographic post-stratification, similar to that used in most academic and media polls, is inadequate”
It basically concludes that the LA Times/Daybreak methodology is the most accurate.
http://www.stat.columbia.edu/~gelman/research/unpublished/swing_voters.pdf