19
poorly in mathematics was actually about the United States falling behind the Russians in
terms of technology and progress. Similarly, merely knowing the number of times the
word “nuclear” was mentioned in a campaign did not seem to be of much objective
interest.
6
Some concepts are more difficult to understand than others and require more
explanation, thus more words. Counting words associated with more complicated
problems as being more important to a campaign’s message did not seem to have much
construct validity.
Without counting key words, how can the content of an ad or speech be
measured? Many possibilities come to mind, including counting the number of
paragraphs that mentioned important concepts or following John Geer’s (1998) lead and
counting the number of “appeals” in each ad or speech. Counting the number of
paragraphs with important key concepts seemed to fall prey to the same weakness as
counting the number of times each word was mentioned – difficult or confusing concepts
may be counted more frequently. Thus, I turned my attention toward Geer’s work on the
content analyses of campaigns.
Geer content analyzed advertisements from presidential campaign ads between
1952 and 1996. His unit of analysis was the “appeal.” For example, one advertisement
might contain many issue appeals and equally many trait appeals. In this method, an
advertisement is not about only one thing, and that attracted me. Through this method,
Geer is able to capture the priorities of the candidate making the ad without losing
6
A good example of the pitfalls of using search engines to do content analysis came when I experimented
with one, asking it to search for the word “war.” It returned an amazingly large number of hits, much to
my surprise. As I investigated them, I learned that candidates use the word “war” to describe a lot of
different things – cultural wars, a war for our future, war on poverty, war on drugs, war on values, etc . . .
relying on a search engine to tell me how many times a candidate spoke about war, in its intended foreign
policy meaning, would have been a mistake.