issues are avoided in laboratory studies (e.g., Neuman et al., 1992; Norris and Sanders,
2003
) that find that people learn from watching news in a laboratory. However, it is hard
to generalize these results to the effects of years of daily media exposure on voters’ knowl-
edge or choices on Election Day.
Snyder and Str€omberg (2010) argue that some variation in the congruence discussed
above is exogenous and can be used to identify media effects. One specification inves-
tigates the consequences of the changing congruence between media markets and con-
gressional districts due to redrawing of the district lines. For example, a newspaper’s
coverage of a congressman may fall because part of its readers are moved to a different
congressional district. ...