Saturday, April 5, 2008

The fallacy of demographic political analysis

After each primary in the continuing campaign for the Democratic nomination for President, we are treated to an analysis by pundits based on the voting behavior of specific groups. This approach does a disservice to the issues debated, not to mention its dubious relevance to the campaigns for future strategic decisions.

On rare occasions, we can admit some groups do vote as a block. There is a small Orthodox community in New York State, and it always votes 100 percent the same as its Rabbi's recommendation. However, groupings based on race, gender and origin have fluctuated widely during the course of the campaign, an indication that analysis based on these characteristics is inaccurate.

There are some tendencies, but they are rather obvious ones. Of course, women and blacks would be inclined favorably to the first person from their group to become President of the United States.

However, when you look at other groups, the analysis falters. How could white men in Wisconsin favor Barack Obama while the same demographic in nearby Ohio voted for Clinton? The answer is quite simple. Demographic groups react to the overall momentum, media coverage and events in the campaign. So what looks like a shift in one group may be mirrored by similar movements in ALL groups.

And when you drill down to the detail used by some of these pundits to justify their time, you get little real enlightenment. I refuse to believe there is a common dynamic among Hispanic women over 30 years of age, with varying degrees of education.

While it would be easy for us to shrug off some of these pundits' statements, there is an inherent danger as well. Just like the arbitrary division of our country into red and blue states creates an impression of splintering, demographic political analysis does the same.

It would be far better to concentrate on the issues in the campaign, an approach political commentators seem to avoid like the plague.