The purpose of this rejoinder is twofold. The first is to question the call to extend a “replication standard” for qualitative research in political science. (See Golden, 1995a). The second is to question how useful the focus on “replicability” is for research in political science in general (see King, 1995a, and the debate on “Verification/Replication” in PS, 1995 28(3): 443–99).

What underlies a replication standard for qualitative research? I believe that the call for a replication standard for qualitative research is applicable to only certain kinds of research, and that, even within its own limitations, is probably misguided. For the most part, when proponents of a replication standard have anticipated dissension over the requirement of replication for qualitative research, they have pointed to logistical, pragmatic, and privacy burdens that such a requirement may impose on researchers.

In contrast, my main contention concerns the very conceptualization of research as replicable. I argue that the assumptions underlying a replication standard are strongly and narrowly positivist. The replication standard itself is reductionist, and its use would marginalize a portion of the qualitative research now being conducted in political science.

How far in advance can one predict the next presidential election? Forecasting as late as election eve is trivially easy, at least since the advent of polling. Forecasting as early as two or four years in advance is virtually impossible (see, e.g., Norpoth 1995). It is at some time between the previous midterm election and election day that electoral expectations come into focus.

The evolving expectations of the 1996 election cycle offer a familiar example. A year ago, Clinton was widely considered unelectable. As of this writing (November 1995), Clinton is perceived as a mild favorite for 1996. By the time you read this essay in spring 1996, a new consensus may emerge. Expectations will harden by election day, when the verdict is not likely to surprise (Gelman and King 1993).

“Trial heat” polls are one way to predict elections (see, e.g., Campbell and Wink 1990), but they are notoriously unreliable when conducted more than a year in advance. A year in advance of the two Reagan elections, for instance, Reagan trailed Carter in 1979 and Mondale in 1983 in the polls. We can, however, examine other early indicators of election “fundamentals.” These indicators include the results of another poll question that is asked regularly throughout the electoral cycle: the president's approval rating in the Gallup poll. Various earlier models of the presidential vote have incorporated a measure of approval (Sigelman 1979; Brody and Sigelman 1983; Abramowitz 1988; Lewis-Beck and Rice 1992).

Since January 1995, when the National Voter Registration Act (NVRA) went into effect, people have been registering or updating their voting addresses at the rate of nearly one million per month in 42 states. Based on early figures, we expect that the rolls will rise by 20 million before the 1996 election and 20 million more by the 1998 midterm election when the full four-year drivers' license renewal cycle will be completed. The resulting increase would be far and away the largest among already eligible voters (as contrasted with the enfranchising of new groups, such as women and 18-year-olds) since the personal voter registration system was introduced in the closing decades of the nineteenth century.

The NVRA requires that states offer to register people to vote when they get or renew drivers' licenses (called “motor voter”), or when they apply for AFDC, Food Stamps, Medicaid, WIC, and disability services (called “agency-based” voter registration). Roughly eight million enrolled or updated their voting addresses in the first nine months—four million of them in drivers' license agencies, one million in public assistance agencies, and three million by mail.

Years ago, when Robert Caro's magisterial biography of Robert Moses was first published, I remember reading it with mixed emotions. On the one hand, it reinforced every lesson I tried to convey in my course on urban politics. On the other hand, it taught those lessons in a way that was so vivid, and so engaging, that it made my conventional textbooks and lectures seem hopelessly abstract and lifeless. There was only one thing to do: throw out the old texts and make The Power Broker (Caro 1975) the centerpiece of the course. I redesigned everything from scratch.

Recently, one of the students in my course on the history of city planning (offered in the University Honors Program at the University of Maryland at College Park) ambled up after class. “Have you ever played sin city?” I thought I heard him ask. He set me straight—“SimCity” he said, enunciating carefully—and then offered a demonstration, during which I was reminded of Caro's book. Here we go again, I though to myself as I purchased the Windows version of the simulation.