DATA

To answer these questions, I collected data on the most highly cited political science articles published in each decade from the 1940s through the 2020s.Footnote ² I operationalized influence with citation data because citations are a standard measure of academic impact (Aksnes, Langfeldt, and Wouters Reference Aksnes, Langfeldt and Wouters2019). The data are from the Web of Science,Footnote ³ which is the world’s oldest and most authoritative database of research publications and citations and the primary data source used in scientometrics today (Birkle et al. Reference Birkle, Pendlebury, Schnell and Adams2020).Footnote ⁴ Citations are from articles in the universe of works appearing in the Web of Science. I considered only journal articles; not books, book chapters, or other types of scholarly works. It is true, of course, that books are central to political science; however, assessing their impact is beyond the scope of this article (Strother Reference Strother2026).

I collected bibliometric data from the Web of Science Core Collection (webofscience.com) on October 15, 2025. Searches were conducted separately for each decade from the 1940s through the 2020s using the following query structure: (((((PY=(1940-1949)) AND DT=(Article)) AND LA=(English)) AND WC=(Political Science))).Footnote ⁵ Search results were sorted by total citation count in descending order, and the top 1,000 records from each decade were exported in the “Full Record” format.Footnote ⁶

Collecting data at the decade level allowed for the identification of temporal patterns in scholarly influence that could be obscured if only the overall most-cited articles were considered. This is a strategy that would disproportionately reflect recent decades due to the growth of the discipline, rising publication volumes, and evolving citation practices. In this approach, each decade’s dataset consists of the most-cited articles published during that decade, with citation counts reflecting all-time citations rather than citations limited to any particular period.

Analysis of the journals that publish the most influential articles in political science requires us to specify what we mean by “most.” Any chosen threshold (e.g., top 10 or top 25) is arbitrary. In light of this difficulty, I adopted and described four different thresholds so that readers can choose their preferred threshold: top 100, top 50, top 25, and top 10. That is, for each decade in the study period (i.e., 1940s through 2020s), I constructed lists of the top 100, 50, 25, and 10 most-cited articles, and then present information about which journals these highly influential articles were published in.Footnote ⁷

ANALYSIS AND FINDINGS

Table 1 presents an overview of the journals in which the most highly cited articles published in the past 85 years have appeared. All told, 79 distinct journals published at least one article appearing on a decade’s top-100 list; 64 distinct journals published at least one top-50 article; 44 journals published one or more top-25 articles; and top-10 articles appeared in 29 unique journals.

Table 1

Journals Publishing Most-Cited Articles of Each Decade, 1940–2024

Table 1. Long description

The table consists of five columns: Journal (Abbreviation), and four columns under the header N Total Appearances: Top 100, Top 50, Top 25, and Top 10.

Key data points include:

* American Political Science Review (A P S R): 218 in Top 100, 104 in Top 50, 54 in Top 25, 21 in Top 10.

* Public Opinion Quarterly (P O Q): 118 in Top 100, 62 in Top 50, 31 in Top 25, 11 in Top 10.

* American Journal of Political Science (A J P S): 49 in Top 100, 22 in Top 50, 11 in Top 25, 2 in Top 10.

* World Politics: 48 in Top 100, 23 in Top 50, 8 in Top 25, 3 in Top 10.

* International Organization (I O): 44 in Top 100, 24 in Top 50, 12 in Top 25, 5 in Top 10.

* Political Analysis (P A): 17 in Top 100, 14 in Top 50, 11 in Top 25, 8 in Top 10.

* Journal of Conflict Resolution (J C R): 15 in Top 100, 4 in Top 50, with no entries in Top 25 or Top 10.

* Annual Review of Political Science (A R P S): 14 in Top 100, 4 in Top 50, 3 in Top 25, 1 in Top 10.

* British Journal of Political Science (B J P S): 7 in Top 100, 5 in Top 50, 2 in Top 25, 0 in Top 10.

* Political Science Research and Methods (P S R M): 3 in Top 100, 3 in Top 50, 2 in Top 25, 1 in Top 10.

The final row shows the total N including unlisted journals: Top 100 is 787 (850), Top 50 is 392 (425), Top 25 is 199 (213), and Top 10 is 80 (85).

Source: Data from Web of Science as of October 15, 2025.

Regardless of which threshold for “most” influential one chooses, a central fact emerges: the American Political Science Review (APSR) predominates these lists, consistently accounting for more than 20% of the most-cited articles. Perhaps it should not be surprising that the flagship journal of the discipline leads the way, but the story is different for the discipline’s other “top-3” journals. Articles appearing in the American Journal of Political Science (AJPS) account for less than 6% of the top-100 lists and only approximately 2% of the top 10, and the Journal of Politics (JOP) accounts for approximately 3% of the articles appearing on a decade’s top-100 and top-10 lists. Other generalist journals, including Annals of the American Academy of Political Science (Annals), Annual Review of Political Science, and Political Science Quarterly, were home to numerous highly influential articles as well.

Several highly regarded but non-generalist journals were home to many of the most influential articles published during the period of study. International Organization (IO) and World Politics each published approximately 6% of the top-100 articles and 4% to 5% of the top-10 articles. Political Analysis (PA), Journal of Conflict Resolution, and Comparative Political Studies also appeared relatively frequently. Many subfield journals published one or more highly influential articles, including Political Behavior, Political Communication, Political Psychology, Policy Studies Journal, and Review of International Political Economy (RIPE), among many others.

A striking finding revealed in table 1, however, is the frequent appearance of non-disciplinary journals. Indeed, the journal that published the second largest number of highly influential articles was the expressly interdisciplinary Public Opinion Quarterly (POQ). It published approximately 13% of the most-cited articles—far more than any other journal except APSR. Other non-disciplinary journals published many most-cited articles as well, including Philosophy & Public Affairs (interdisciplinary); Public Choice (interdisciplinary); New Left Review (multidisciplinary, including social sciences, humanities, and the arts); Social Science Quarterly (interdisciplinary); and Journal of Public Administration Research and Theory (public administration). We reasonably might ask whether articles appearing in these journals should be included in the analysis. Ultimately, I included them because the Web of Science expressly categorizes them as “political science”; because each journal lists political science (or “politics”) as constituting at least part of its topical focus; and/or because each journal is consistently indexed as “political science” by professional indexing services.

Figures 1 and 2 depict over-time trends in where these highly cited articles—the top 100 and the top 10, respectively—appeared.Footnote ⁸ Figure 1 plots the number of top-100 most-cited articles in each decade published in the 16 journals that had the most top-100 papers.Footnote ⁹ The data reveal substantial changes over time. In the 1940s and 1950s, the lion’s share of the most influential articles were published in POQ. Indeed, as discussed previously, POQ accounts for the second-most highly influential articles, but more than 80% of those articles appearing on a decade’s top-100 list were from the 1960s or earlier.

Figure 1

Journals Publishing Top-100 Most-Cited Articles (Count per Decade) Over Time

Notes: These 16 journals collectively published 78% (N=666) of the top-100 articles for the decades covering 1940–2024. The remaining 22% of top-100 articles (N=184) were published in 63 different journals. Data from the Web of Science.

Figure 1 Long description

A multi-panel figure containing 16 individual bar charts arranged in a 4 by 4 grid. Each chart shares a common Y axis representing article count from 0 to 50 and an X axis representing decades from 1940 to 2020.

Row 1:

* A J P S: Low counts peaking at approximately 10 in the 1980s, 1990s, and 2010s.

* A P S R: High counts starting in 1940, peaking at nearly 50 in 1960, and steadily declining through 2020.

* A R P S: Near zero until a small rise to approximately 5 in 2010.

* Annals: A peak of nearly 20 in 1940, dropping to near zero for all subsequent decades.

Row 2:

* C P S: Consistently low counts under 5 across all decades.

* E J P R: Near zero until a slight increase to approximately 5 in 2020.

* I O: Low counts peaking at approximately 15 in 1990.

* J. Confl. Resolut.: Consistently low counts under 5 across all decades.

Row 3:

* J O P: Low counts peaking at approximately 8 in 1950.

* J P R: Consistently near zero across all decades.

* P A: Near zero until a peak of approximately 10 in 2010.

* P O Q: High counts starting at 42 in 1940, declining steadily to near zero by 1990.

Row 4:

* P S Q: A peak of 10 in 1940, dropping to near zero for all subsequent decades.

* Philos. Public Aff.: Near zero except for a peak of 10 in 1970 and 5 in 1980.

* Public Choice: Near zero except for a peak of approximately 8 in 1980.

* World Politics: Consistent low counts between 5 and 10 from 1940 to 1990, then dropping to near zero.

Figure 2

Journals Publishing Top-10 Most-Cited Articles (Count per Decade) Over Time

Notes: These 13 journals collectively published 81% (N=69) of the top-10 articles for the decades covering 1940–2024. The remaining 19% of top-10 articles (N=16) were published in 16 different journals. Data from the Web of Science.

Figure 2 Long description

The figure consists of 13 small-multiple bar charts. Each chart has an X-axis labeled with decades from 1940 to 2020 and a Y-axis ranging from 0 to 6.

Row 1:

* A J P S: 1 article in 1980 and 1 in 2000.

* A P S R: Peaks at 6 articles in 1960, with a steady presence from 1940 to 2020.

* Annals: 4 articles in 1940 and 1 in 1990.

* B J P S: 1 article in 1970 and 1 in 1980.

Row 2:

* I O: 2 articles in 1980 and 3 in 1990.

* J O P: 1 article in 1940 and 2 in 1950.

* J P R: 2 articles in 1960.

* P A: Peaks at 5 articles in 2010, with 2 in 2000 and 1 in 2020.

Row 3:

* P O Q: Starts high with 4 articles in 1940, decreasing to 3 in 1950, 2 in 1960, 1 in 1970, and 1 in 2010.

* Philos. Public Aff.: 2 articles in 1970 and 1 in 1980.

* Polit. Stud: 2 articles in 1990.

* S S Q: 1 article in 2000 and 1 in 2020.

Row 4:

* World Politics: 2 articles in 1950 and 1 in 1970.

APSR consistently has accounted for a large percentage of the most-cited articles published each decade, but its prevalence has changed considerably over time. APSR dominated the most-influential lists in the midcentury, accounting for 38 of the articles in the 1950s, 46 in the 1960s, 36 in the 1970s, and 31 in the 1980s. In subsequent decades, APSR continued to account for a relatively large but declining share of the most-cited papers. For example, in the 2010s, APSR published nine of the top-100 articles—far fewer than its share in earlier decades but nevertheless the third-most of any journal in that period (AJPS and PA each published 10 top-100 papers in the 2010s).

This does not mean that the importance of APSR has abated much. It remains the discipline’s flagship journal; it is highly visible and article placements still are coveted. Recent reports by APSR’s editors indicate that its acceptance rate hovers around 5%. It has one of the highest impact factors in the discipline, which indicates that—on average—papers placed there are cited relatively more than most other political science journals. APSR is home to relatively fewer of the most influential papers published in the discipline today than it was in the past; it currently looks more like a first among equals rather than a hegemon.

A few other journals merit particular attention here. Since the 1970s, AJPS has consistently been home to approximately 10% of the top-100 articles. JOP and World Politics have been fairly consistent over time, although the latter produced fewer highly cited articles in the most recent decades. IO consistently produced a significant percentage of the top-100 articles in the later decades of the twentieth century but has since fallen off. Both Political Science Quarterly and the Annals were home to a significant percentage of the most influential articles in the very earliest decades in the data but rather fewer since then. PA, conversely, quickly rose to prominence after its creation in 1990.

Focusing on only the top-10 most cited articles published in each decade, a slightly different pattern emerges, as shown in figure 2. Again, APSR predominates, with its share of top-10 articles peaking in the 1960s and then slowly waning. AJPS has been home to many top-100 articles but few top-10 articles and JOP published three top-10 articles but none in the past seven decades. The Annals and POQ published a significant percentage of top-10 articles in the 1960s and before but very few since. As with the top-100 articles, PA again published several top-10 articles in recent decades—indeed, it was home to fully half of the top-10 articles published in the 2010s.

The recent prominence of PA reflects the growing centrality of method-focused work in political science. The first article making a top-10 list the chief contribution of which was methodological was published in 1955: Scott’s (Reference Scott1955) “Reliability of Content Analysis” (POQ Reference Scott1955); next was Sartori’s (1970) “Concept Misformation in Comparative Politics” (APSR Reference Sartori1970). Another such article would not make a top-10 list until Beck and Katz’s “What to Do (and Not to Do) with Time-Series Cross-Section Data” (APSR Reference Beck and Katz1995). Since then, methodologically oriented articles consistently have ranked among the most-cited papers published each decade. Three such articles count among the top-10 most cited of the 2000s (two in PA and one in Political Research Quarterly); five more appeared on the list for the 2010s (all in PA); and two appeared in the 2020s (one each in PA and Political Science and Research Methods).

Methods papers have potential citational advantages over other types of papers. First, they cut across subfields. Even an excellent paper that substantially advances the theoretical state of the art in, say, American politics, is relatively unlikely to find much purchase among scholars of international relations—at least in part because of the fundamentally different orientations of those subfields (Jackson and Nexon Reference Jackson and Nexon1999). But scholars working across political science have reason to want to understand linear regression models with multiplicative interaction terms (Brambor, Clark, and Golder Reference Brambor, Clark and Golder2006) and to think about the ways they could incorporate quantitative text analysis into their work (Grimmer and Stewart Reference Grimmer and Stewart2013). Second, methodological innovations typically can be taken up fairly quickly by the broad audience just discussed. Indeed, scholars publishing such work have incentives to make their innovations easy to use with open-source resources, accessible and well-documented code, user-friendly software packages, and the like (Arel-Bundock and McCrain Reference Arel-Bundock and McCrain2023).

Big-picture theoretical papers, on the other hand, may have more staying power and thus be cited long after publication. Consider, for example, how frequently we see citations to “Two Faces of Power” (Bachrach and Baratz Reference Bachrach and Baratz1962) and “Some Social Requisites of Democracy” (Lipset Reference Lipset1959) in contemporary research. Aristotle continues to receive scholarly attention some 2,300 years since his most recent publication. In contrast, not many of us cite Galton or Gauss to justify our regressions. At the time of writing, the second-most-cited paper of the 2010s was a methods paper: Berinsky, Huber, and Lenz’s (Reference Berinsky, Huber and Lenz2012) article on the utility and attendant risks of Amazon’s Mechanical Turk workplace for drawing convenience samples for experimental studies. It seems unlikely (although who knows!) that scholars will still be debating the relative merits of MTurk into the 2050s. More generally, as the methodological state of the art advances, papers that were highly cited for a while (e.g., for a decade or so) may quickly lose relevance. The upshot is that it is entirely possible that, in the fullness of time, theoretical papers could overtake more technical and methodological papers.

It may be that these trends—the decreasing centrality of APSR and increasing prevalence of methodological articles among the most cited—are all part of larger trends evident across academic domains. Citations are increasing everywhere; yet, at the same time, new scholarship is becoming less disruptive—especially in the social sciences (Park, Leahey, and Funk Reference Park, Leahey and Funk2023). That is, more recent papers, on average, do less to push the field in new theoretical directions compared to older papers. In such a world, it makes sense that theoretical papers would become somewhat less central in the discipline.Footnote ¹⁰

Finally, the data demonstrate the dominance of American journals. The overwhelming majority of the most-cited articles identified here are published in American journals—those sponsored by American associations of scholars, housed at American universities, and edited by teams of scholars based predominantly in the United States. Only 10 of the 85 articles identified here as top-10 papers within the decade they appeared were published in international journals: two in British Journal of Political Science and one each in Australian Feminist Studies, European Journal of Political Research, Government and Opposition, Journal of Peace Research, New Political Economy, Political Science Research and Methods, Political Studies, and RIPE. Scholars at international institutions routinely publish in the leading American journals (e.g., APSR, AJPS, and IO), to be sure, but they are American journals nonetheless.Footnote ¹¹