Take the oft-reported decline in voter turnout since the 1960s, when more than 60 percent went to the polls in presidential election years. So low has the nation supposedly sunk in the intervening years that the 1996 election drew less than half of the American electorate to the voting booths.
But the widely reported "turnout rate" is not really the number of votes cast divided by the number of Americans eligible to vote, note political scientists McDonald, of the University of Illinois at Springfield, and Popkin, of the University of California, San Diego. The denominator researchers use instead (because it’s more readily available) is the Census Bureau’s calculation of the voting-age population. This figure includes noncitizens, felons, and others not eligible to vote, and excludes military personnel and other citizens overseas who are eligible.
Making use of government statistics on noncitizens and the other subgroups, McDonald and Popkin modify the votingage population figures to produce a more accurate estimate of the electorate and its turnout. Their calculations show that turnout did indeed fall after 1960—from a 1960 level of 63.8 percent to 61.5 percent in 1968 and 56.2 percent in 1972. But since then, the number of ineligible noncitizens and felons has been increasing rapidly, and when that and other adjustments are made, the post-1972 numbers show no clear trend up or down.
The turnout for the 1996 election, by these new calculations, was more than half (52.6 percent) of the eligible electorate, and for the 2000 contest, 55.6 percent. In the 1992 election, 60.6 percent of the eligible electorate voted—a figure that should warm the hearts of analysts who mourn a golden age they thought ended in 1960.
The alarmists still have the supposedly low level of trust in government to worry about (or at least they did before the September 11 terrorist attacks sent poll-measured trust in government surging to its highest level in decades). But Moore, senior editor of the Gallup Poll, says that even before the terrorist attacks there was no clear cause for concern.
There may have been a decline in "trust" over the years, he says, but it was unclear just what "trust" meant or how much of it there was. The level of trust varied widely with the wording of pollsters’ questions. The most often cited poll, conducted since 1958 by the University of Michigan’s National Election Studies, asked respondents if they could "trust the government in Washington to do what is right—just about always, most of the time, or only some of the time?" In 1997, only 32 percent gave one of the first two responses. Yet that same year, Gallup got a very different answer with a slightly different question: It found that 62 percent had "a great deal" or "a fair amount" of "trust and confidence . . . in the executive branch," and 54 percent did "in the legislative branch."
Even if the levels of trust in government fell as low as alarmists believed, observes Moore, American democracy did not seem impaired. Citing a 1998 Pew Research Center report, he notes that in surveys conducted between 1987 and 1997, about 90 percent of Americans consistently said they were "very patriotic." Other polls confirmed that. "If people remain committed to their country, even though they believe the government does what is right ‘only some of the time,’ what’s the problem?" asks Moore.