This article proposes weighted content analysis as a methodological extensi
on of traditional (unweighted) content analysis of political advertising. W
eighted content analysis distinguishes between political ads that were prod
uced but not aired and ads that actually aired in a particular media market
. To illustrate the advantages of weighted content analysis, this article e
xamines the 1996 presidential campaign in Columbus, Ohio. The tone of 132 p
residential campaign ads produced by both candidates is compared with the t
one of the 2,522 ads that aired on the three major network affiliates in Co
lumbus. The price of the ad buy is used as an alternative weight that takes
into account audience size and thus allows a further distinction between a
ired and watched ads. Both weighting procedures show that Republican advert
ising in 1996 in Columbus was more negative than Democratic advertising dur
ing the fall campaign, a result that remains undetected by unweighted conte
nt analysis. The article concludes by describing three different ways to ob
tain weighting data: political files kept by television stations and cable
providers, commercial tracking data, and internal campaign records.