(Written in Oxford, OH): There is already too much post-Iowa punditry, but here are two points I would stress:
The two candidates who were endorsed by labor unions each got about 20 percent of union household votes. Unions were thus completely unable to deliver their own rank-and-file in a Democratic contest with a small turnout. This puts an exclamation point on a long sentence about the decline of labor power in America. I already thought that the SEIU and AFSCME endorsements of Dean reflected weakness. He had no significant prior relationship with them and owed them nothing. But they thought he had the best chance to win, and wanted to back a winner. In the old days, unions aspired to make winners, not ride along behind them.
There were substantial differences between the preferences that people expressed in “entrance polls” and the actual Caucus results. As reported, people changed their votes once inside the Caucuses. One letter to today’s New York Times claims that this is evidence of deliberation. “The heart of democracy is not pulling levers or punching out chads; it is government by discussion.” Another letter-writer complains that people were pressured by the “awkward social occasion” of a caucus to change their votes in favor of popular, mainstream candidates. Instead of deliberation, this writer sees peer-pressure. And many commentators detect strategic voting. People shifted to their second choices when they saw that their first choices lacked sufficient support to win delegates. I would say that deliberation is great; strategic voting is OK; and peer-pressure is bad.
If we assume that there was some deliberation in Iowa (and not much peer-pressure), then we face a classic tradeoff. The Caucuses are especially valuable forms of democratic participation, but also especially difficult: lengthy, complicated, and “socially awkward.” Thus they trade quantity of participation for quality. This is often a choice in the design of democratic institutions.