WikiSummary, the Social Science Summary Database

Rosenstone and Hansen. 1993. Mobilization, participation, and American democracy.

Main Points

The book addresses two main questions: Why people turn out, and why turnout has declined over time.

Why Turn Out

Citizens participate based on their personal costs and benefits of doing so, but that's only half the story: they also participate when politicians mobilize them. And politicians are strategic in deciding whom to mobilize. Thus, why we turn out:

1. What about you affects the costs/benefits of voting?
   • Costs of voting: Income, education, age, race, efficacy, institutional factors (e.g. registration laws)
   • Benefits of voting: strength of partisanship, feeling thermometers for candidates
2. Are you being mobilized?
   • What social networks do you belong to: church membership, homeowner, years in community, current employment, etc.
   • Contacted by party?
   • Is the election close? Open/unopposed (for House)?

Why Turnout Has Declined

The authors attempt to explain the decline in turnout between the 1960s and 1980s:

1. Mobilization is the major cause: Less effort at mobilization: explains 54% of decline.
2. Voting age drops to 18: explains 17% of the decline in turnout
3. Weakened social involvement: explains 9% of decline
4. Declining feelings of political efficacy: explains 9% of decline
5. Weakened attachment to parties/candidates: explains 11% of decline

So why is mobilization falling? Party organizations gave way to candidate-centered campaigns. Politics became more impersonal; local party clubs waned. (This trend has begun to reverse.)

Implications

Personal-Level Variables: Two Paradoxes

Voting is costly; income, education, and other resources enable some people to overcome these costs. Voting also has benefits; material preferences, group identifications, and preferences/beliefs make these benefits more worthwhile to some people.

Still, the benefits will never exceed the costs for most people. Thus, two paradoxes: rational nonparticipation and rational ignorance. Moreover, a model based only on personal-level variables can't explain why participation peaked in the 1960s, dipped in the 1970s, then rose again in the 1980s--even while education, income, and so on rose steadily; thus, they don't explain participation.

Social-Level Variables

Social networks help us overcome rational ignorance (because talking with family, friends, and neighbors is a low-cost way of learning about what's going on), but they don't do enough. To understand trends in participation, we need to look at mobilization.

Politicians are strategic in who they mobilize (targeted mobilization). They mobilize groups that will provide the most benefit. They mobilize people they already know; people who are centrally positioned in social networks; people whose actions will make the most differences (the powerful); and people who are likely to respond (i.e. people with resources). Politicians apply these strategies when they attempt to mobilize unions (existing networks), business leaders, and the wealthy and educated.

Politicians are also strategic in when they mobilize. They mobilize people when salient issues top the agenda; when other concerns aren't on their minds (e.g. don't mobilize students during midterms); when important decisions are pending (e.g. during presidential election years more than during midterm years); and when outcomes will be closely decided.

Implications of Lopsided Participation

The conclusion shows that, no matter what type of participation you look at, white/wealthy/educated people participate far more than their share. As more people participate, however, participation becomes less biased. This has two applications. First, more popular modes of participation have a less biased sample (e.g. voting is more popular than letter writing; the voting population looks more like the whole population than the letter-writing population does). Second, when participation as a whole is up, it is less biased (e.g. if more people are voting, writing letters, and everything else, then it will be less biased).

Data

NES data. Lots of it. See chapters 3-7.