The September PS symposium addresses the uses and abuses of political science data, and the rules or standards, if any, the profession should adopt to preserve the integrity of political science research and to advance knowledge. The ability to verify research findings by reanalyzing original data or to replicate findings by analyzing new, independently collected data has been the focus of debate among political scientists since Gary King distributed his initial version of “Replication, Replication” in 1992. The PS verification/replication symposium is the latest but most public round of the ongoing debate.

Political science is a community enterprise; the community of empirical political scientists needs access to the body of data necessary to replicate existing studies to understand, evaluate, and especially build on this work. Unfortunately, the norms we have in place now do not encourage, or in some cases even permit, this aim. Following are suggestions that would facilitate replication and are easy to implement—by teachers, students, dissertation writers, graduate programs, authors, reviewers, funding agencies, and journal and book editors.

Political scientists believe they ought to promote research as a social enterprise and develop a stronger sense of community within the discipline. They disagree on how to foster community and on the best way to promote research. The position taken in “Replication, Replication” (King 1995) is that these goals can be achieved by requiring researchers who collect data to surrender the products of their labor to others without compensation.

The public discussion over replication has been ongoing among political scientists for approximately three years. I welcome PS and the American Political Science Association to this debate. The American Journal of Political Science (AJPS) has operated with a replication policy since January 1, 1994. The first part of this essay reflects my views on replication; the second, my experiences as one of the first editors to adopt the policy.

Several important areas of agreement in the King-Herrnson debate suggest that a broad consensus exists on the desirability of changing disciplinary practice and norms governing the collection and release of original data. After reading both essays, I am convinced that both Gary King and Paul Herrnson would agree that (1) Collecting and organizing political science data is often a costly enterprise that requires a substantial commitment of valued resources, including money, time, and expertise.

Having discussed this issue at three meetings of our editorial board, we were already familiar with most of the arguments presented in the thoughtful essays by Gary King and Paul Herrnson. We can confirm Herrnson's supposition that the proposed policy is controversial, and that the chief difficulty has to do with the relinquishing of what might be called intellectual property before its creators have fully exploited it. Although we have decided to wait until a thorough discussion of the issue is conducted under the auspices of the APSA before finally deciding on a policy, any statement we might require of authors as a condition of publication in The Journal of Politics is likely to fall considerably short of what King recommends.

Providing good answers to the questions raised in the discussion between Gary King and Paul Herrnson requires paying attention to the full range of interests affected. Unless we pay attention, individual political scientists, research groups, departments, peer reviewers, funding organizations, data archives, journals, and presses will not be able to establish and uphold an effective set of norms relating to the collection, analysis, reporting, and preservation of data.

Everything I have published, apart from some ghost-written school speeches for my children, I have written as a social scientist. Having persuaded myself that I had discovered something worth reporting, I have attempted to persuade others that it was worth knowing. I have taken this to be consistent with the spirit of science, perhaps naively; a spirit encompassing, among other things, the value of public argument based on publicly reproducible evidence.

Gary King has done a great service to the discipline in raising the question of how we meet appropriate standards for assuring that our published work advances the collective knowledge concerning government and politics that is in our mutual professional interest. He has done so in a truly collegial manner, stating his view, raising concerns, and inviting discussion. The editors of PS are using this publication in precisely the proper way by providing a forum in which this professional debate can be aired.

Like Gary King and Paul Herrnson, we believe that making our data accessible can strengthen and advance scholarly knowledge. King and Herrnson, however, disagree about the measures necessary to ensure that data are widely available to achieve this collective good. Herrnson criticizes King's recommendations as potentially bad for the profession. As scholars who have collected data and relied on others' data for our own research, and—equally important—as relatively new Ph.D.s in political science, we have our own perspective in this debate over replication and data accessibility.

Gary King and the Methodology Organized Section of APSA have certainly done the profession an important service by point out some of the deficiencies of our current practices on archiving data sets. For 40 years now, political scientists have grumbled about the lack of availability of major data sets (with the McClosky (1960) data being perhaps the most prominent example), so it is high time that the profession as a whole adopts some guidelines and standards. King's article in this journal is an important way to start the discussion.

From one standpoint, it is quite easy to concur with Gary King's exhortations on behalf of researchers making their data sets publicly available. His plea is, after all, consistent with a basic canon of scientific work, namely, that results be capable of replication. Our disagreements with the position taken by King have to do both with practicalities and with an excessively narrow focus on what it means to reassess a given set of findings—issues that we think Paul Herrnson addresses quite tellingly in his statement.

over replication is comparable in many respects to controversy among policy analysts about the ends and means of regulation. Proponents of replication seek to provide the collective good of free-flowing information and to sanction the negative externality of sloppy scholarship. Similarly, opponents of replication claim property rights in the data and variables they have created through hard work and ingenuity. As with disputes over regulatory policy, the fundamental sticking point is whether the benefits of mandatory replication to the community of political scientists outweigh the costs of compliance to individuals.

In “Replication, Replication,” Gary King convincingly argues that publications by political scientists should adhere to what he calls a replication standard. Although King's article explicitly embraces qualitative as well as quantitative research, the policy statement that he suggests editors and reviewers of books and journals endorse exclusively imposes standards on studies based on quantitative research. Qualitative work is treated in a simple sentence and one that enforces no standards: “Authors of works relying upon qualitative data are encouraged (but not required) to submit a comparable footnote that would facilitate replication where feasible.”

The Social Science Quarterly (SSQ) is the major interdisciplinary journal in the social sciences, typically receiving more than 400 manuscripts per year. We publish about 10% of these, so we are in the same ball park as the major disciplinary journals in sociology and political science. More than, perhaps, even the American Journal of Political Science (AJPS), we publish empirical studies. Economic sociology, environmental studies, political behavior, ethnic studies, gender studies, urban affairs, and several other areas are among our specialties.

Replication of scholarly analysis depends on individual researchers being able to explain exactly what they have done. And being able to explain exactly what one has done requires keeping good records of it. This article describes basic good computing practices and offers advice for writing clear code that facilitates the task of replicating the research. The goals are simple. First, the researcher should be able to replicate his or her own work six hours later, six months later, and even six years later.

As evidenced by the thoughtful symposium participants, political scientists have a variety of opinions about verification, replication, and data archiving. I continue to have strong reservations regarding the proposed replication, verification, data relinquishment rules debated here. The symposium, however, has convinced me of one important thing: a still broader discussion is needed before editors or organizations change the norms governing research or publication. Walter Stone's (1995) proposal to survey political scientists to obtain their opinions on verification/replication is, therefore, a good one.

Receiving thoughtful comments in this symposium from 19 of the discipline's most innovative empirical researchers is gratifying and goes a long way toward bringing issues of replication and data accessibility into public view. The progress made in the preceding pages clearly demonstrates the virtues of a community of scholars willing to share sufficient information to benefit from each other's work.

Computer models hold a continuing fascination for political scientists. Part of their attraction derives from being working miniatures like mechanical toys. They also hold the less enjoyable but more scientific promise of allowing something akin to laboratory experimentation—denied to political scientists in most circumstances because of cost or the nature of the subject matter.

Spreadsheets are powerful tools for the construction of simple or complex models. At their core, spreadsheets are nothing more than cells arranged in columns and rows into which labels, numbers, or formulas can be placed. The numbers and formulas can be arranged in infinite variety, and in modern spreadsheets they are supplemented by such features as built-in formulas called functions, macro languages which are highly flexible programming environments, data query features, and graphics capabilities.

Edith Stokey's and Richard Zeckhauser's clearly explained difference equation models lend themselves especially well to rendition on a spreadsheet (1978, 47–73). Stokey and Zeckhauser furnish the case of public housing units in a city deteriorating at the rate of 10% per year to the point of being uninhabitable. If 800 units per year are built, will the total number of public housing units reach an equilibrium and, if so, will it be stable?

Such a problem is expressed in difference equation form as follows:

Unitsn = Units n − 1 – .1

* Units n − 1 + 800

or

Units n = .9 * Units n − 1 + 800

The number of housing units in any year n equals .9 times the number of housing units in the previous year plus 800. The beginning of a spreadsheet solution to this problem is shown in Figure 1.

In a recent article in this journal, James McGregor (1993) criticizes political scientists for practicing regression analysis without recognizing the limitations in the regression technique. McGregor argues that the technique is inappropriate for political science research in many cases and suggests that we pay close attention to both the substantive (e.g., linearity and additivity) and statistical aspects of the technique.

While acknowledging that many of us have not done enough to meet the statistical assumptions of regression analysis, George Krause (1994) argues that McGregor has underestimated the value of the technique to political science research. To support his argument, Krause directs our attention to a number of complicated regression models, including those designed for situations in which the dependent measure is a discrete variable (logit, probit, ordered probit, multinomial logit, Poisson, negative binomial and generalized event count) or the independent effects on the dependent variable are lagged distributed (polynomial distributed lag). He also mentions the recent developments related to event history regression, bootstrapping (a nonparametric approach to statistical inference), and vector autoregression (VAR, a creative use of OLS in simultaneous time series data). Krause deserves credit for his attempt to show us that the scope of the regression technique is not as narrow as McGregor might have assumed.

However, in rejecting McGregor's pessimism, Krause does not realize that McGregor's major concern with regression analysis, as he illuminates by applying OLS to certain laws of nature, is not about the type of dependent variable but about relationships.