Skip to main content Accessibility help
Hostname: page-component-559fc8cf4f-6pznq Total loading time: 0.613 Render date: 2021-03-04T07:49:42.662Z Has data issue: true Feature Flags: { "shouldUseShareProductTool": true, "shouldUseHypothesis": true, "isUnsiloEnabled": true, "metricsAbstractViews": false, "figures": false, "newCiteModal": false, "newCitedByModal": true }

Making (and Sometimes Taking) a Difference: The Dynamic Career of Janet M. Box-Steffensmeier

Published online by Cambridge University Press:  16 October 2020

Anand Edward Sokhey
University of Colorado, Boulder
Rights & Permissions[Opens in a new window]


Image of the first page of this article
© American Political Science Association 2020


When you’re a graduate student you hear a lot of adages like “academe is a marathon, not a sprint,” or “the perfect is the enemy of the good.” One piece of advice that many of us recall faculty relaying goes something to the effect of “this business is about all about asking the right questions.” That last statement—like a lot of general truths— doesn’t always seem particularly helpful when you’re trying to learn matrix algebra or equilibrium concepts. However, it feels increasingly on point when you find yourself on the other side of the desk, and then somehow profound when you move into the role of evaluating the records of other professors. Indeed, a lot of people seem to carve out good careers by asking “the right questions.”

Of course, what many of us don’t hear during our training is that when you ask your questions is probably as important as what questions you’re asking. Dr. Janet M. Box-Steffensmeier has built an exceptional career by consistently asking the right questions at the right time. Appropriately, those questions have often been about time itself, covering temporal aspects of quantitative methodology (most notably, time series and survival analysis techniques). However, they also span tremendous substantive ground: How stable is partisanship in the United States? What factors influence high-quality challengers to run for federal office? How do interest groups coordinate, and how do they influence judicial decision making? Reading Box-Steffensmeier’s work provides insights into practically all aspects of American politics (and some aspects of AmericanaFootnote 1), from the behavior of the masses, to the actions of members of Congress and the Supreme Court, to the activities of the parties and interest groups that serve to link the public to these governing institutions. Her curriculum vitae could be the syllabus for a graduate proseminar in political methodology and/or the American political system; it’s a record that reads like someone who’s been sprinting for 30-plus years (with no signs of slowing), and most would agree that her work is considerably closer to “perfect” than “good.”

2020–2021 APSA President

Janet M. Box-Steffensmeier

Vernal Riffe Professor of Political

Science and Professor of Sociology,

The Ohio State University

That said, Box-Steffensmeier’s contributions extend well beyond the pages of the journals. Her teaching and service record is nothing short of remarkable, and has long sought to address an equally pressing series of issues: How can universities extend opportunities for rigorous methodological training in the social and behavioral sciences? How can we accommodate students who need flexible or non-traditional learning arrangements? How can we encourage participation by women and minorities in places—like the political methodology community and broader discipline—where they have and continue to be underrepresented? At the same time, Box-Steffensmeier might also be called “mentor to all,” given the number of people who have a story about her helping them with a project or professional situation. How can we be better scholars, better teachers, and better colleagues? One answer is to emulate Box-Steffensmeier’s behavior over the past 30 years.

Suzie Linn describes Box-Steffensmeier as “a tribute to the profession.” Alison Craig calls her “unfailingly kind and generous,” and “a model of the kind of mentor we should all strive to be.” Dino Christenson comments that “[w]hile few scholars are more intimidating on paper, a big part of what makes Jan such an amazing mentor is that she is anything but in person.” All of these statements are true, and if you’re meeting her for the first time, there are several things you should keep in mind. First, any adage she shares—and it is almost certain she will offer some form of advice—will always be helpful, both in that moment and down the road. Second, regardless of whether you’re talking about a theory of interest groups, properties of estimators, or good names for cats, she will very likely make you think of something you hadn’t considered previously. And third, any formalities you lead with will be met by an energy and friendliness that is refreshing and totally authentic. You might feel the need to address Box-Steffensmeier by one of her many (current or past) titles— “Distinguished University Professor,” “Dean,” “Vice-Provost,” “Middle School Golf Coach,” “Fellow of the American Academy of Arts and Sciences,” and now “APSA President.” However, she will most likely insist that you call her “Jan,” and then proceed to give you (a generous amount of ) her focused attention.


Like every superhero, Jan has a great origin story. Jan was raised on a small farm in Henry County, Iowa. Her parents both worked the farm; her father was also a brick mason. Jan recalls hearing her father talk and comment on politics when she was little, and her first political memory was attending a “meet the governor” townhall with him. Jan’s mother frequently worked elections, and Jan remembers her mother’s comments on one-party dominance in Henry County piquing her interest in politics.

Jan was valedictorian of her (small) high school class, though she notes that it didn’t seem like a sure thing that she would go to college or get a degree. As a first-generation college student, she interviewed for scholarships at Coe College in Cedar Rapids and felt at home in (what she later recognized as) a traditional liberal arts environment. She recalls being “unpolished” as she entered her undergraduate education, lacking the same writing skills as some of her peers and not having had exposure to topics like calculus in her small hometown. “My early papers at Coe had more red ink than typed words… The fact that Dr. [Fred] Wilhoite [a professor of political science] would take the time to write those detailed comments was critical to my later success. My professors taught me how to write, they didn’t give up on me just because I arrived without those skills.” Jan double-majored in mathematics and political science. The math degree was part of an initial plan to be a high school teacher; the political science one came about because the political science professors at Coe were such “captivating” teachers that she simply kept taking their classes. Jan credits her Coe College honors adviser, Dr. Peter McCormick (Dept. of Philosophy), with helping put her on the path to graduate school— McCormick suggested that she might be a good fit for the college classroom, as teaching at the high school leveled required being more of a disciplinarian (Jan has always been really nice).

Jan applied to graduate school in both mathematics and political science, passionate about each subject. She ended up choosing political science because the math programs did not care about her political science major, but the political science programs saw her math major as a strength—she realized that she could pursue both her interests in the social sciences. While interning at the US GAO in Washington, DC, she became interested in potentially attending the University of Texas at Austin for graduate school; an office colleague had graduated from there. UT’s separate specialties in methods and formal theory were of particular interest to her at the time, and UT fit the bill on other dimensions, with strong programs that matched her husband Mike’s career plans.

Jan hit the ground running in Austin, discovering a love for research that complemented her interests in teaching. She took comprehensive exams in empirical methods, formal theory, and American politics, and started to develop her professional identity and research interests. In 1990 she attended her first political methodology (“PolMeth”) meeting at Washington University in St. Louis, and describes it as a moment when she knew where she belonged: “I just felt electric excitement about the questions asked and the approaches to answering them.” She also credits that and other early methods meetings with forming a core part of her long-term professional network. The fellow graduate students she met over those summers have become life-long friends and colleagues (not to mention prominent figures themselves)—an incomplete list is comprised of Mike Alvarez, Nancy Burns, Liz Gerber, Jim Granato, and Simon Jackman. Jan also met several of her first coauthors at the PolMeth meetings, including Suzie Linn, Brad Jones, and Renee Smith.

Somewhere in this period, Jan found herself increasingly drawn to questions involving time. And her dissertation reflected this, beginning to articulate a research agenda that would define much of her career. Her thesis focused on the role of money in campaigns, noting that previous work had done a poor job looking at how contributions and expenditures should vary across election cycles (based on conventional wisdom about candidate strategy). In addition to making a general argument about the need for scholars of politics to give greater attention to temporal dynamics (and the inferential leverage that comes with it), her dissertation anticipated the focus on causal inference that has come to dominate many conversations in the decades since: she conceptualized campaign expenditures in terms of non-random assignment and proposed a series of equations for dealing with the problem.

Jan’s thesis was chaired by Melissa Collie. Jan notes that “Melissa was inspiring as a role model as she was one of the very best teachers I had in graduate school and she was a mom of four kids. I saw her and felt it was possible to be a mom and an academic… Melissa helped me navigate having my first child in graduate school, getting to conferences, etc.” Other committee members included Walter Dean Burnham, Mel Hinich, Brian Roberts, and Charles Franklin (then at Washington University, and whom Jan credits as getting her started on event history analysis). A later version of the dissertation work tackling endogeneity in campaign expenditures—“A Dynamic Model of Campaign Spending in Congressional Elections” (1997)—would become the first of her (five and counting) publications in Political Analysis (a journal that consistently has one of the highest impact factors in political science). Jan credits her coauthor on that piece, Tse-Min Lin (then an assistant professor at UT)—and another UT faculty member outside her committee, James Enelow—with also playing critical parts in her training and professionalization while in Austin. Marc Hetherington, another distinguished alumnus of the Department of Government at UT, was just starting the graduate program about the time Jan was wrapping up. Still, he shared the following reflection on his impression of Jan the graduate student: “In the years that I was at UT, and I am certain in the 23 years since I left, Jan Box-Steffensmeier has been held up as the model for what all graduate students should aspire to be. Not only were her professional accomplishments and career trajectory already clear, they came wrapped in a package of humility, generosity, and kindness.”


In 1993, Jan joined the department of political science at The Ohio State University, the institution where she has spent the duration of her career. Her first decade in Columbus would be marked by tremendous productivity (as have subsequent ones); by 2003 she would have a national and international reputation as a leader in the field of political methodology and American politics, and be named the Vernal Riffe Professor of Political Science. Paul A. Beck, who served as the chair of the department of political science at Ohio State from 1991– 2004, offered the following thoughts on her addition to the department and rapid rise within the discipline:

“[I]t is clear to me that Jan was my best hire ever. I am proud of many others, some in endowed chairs at Yale, Princeton, and Berkeley, but no one has achieved what Jan has achieved. I could see from the beginning that she was destined to greatness in how efficiently she organized her personal (raising four children!) and professional lives, how proactive she was in research and grant getting even in her early years, and how skilled she was as a methodologist, traditionally a more male domain. It was hardly surprising to me how quickly she advanced through the faculty ranks to become an endowed professor just 10 years after receiving her PhD.”

During this time Jan focused on time; she pursued the core of the agenda envisioned in her doctoral thesis, asking substantive questions about temporal dynamics as they relate to campaign spending, position-taking by members of Congress, and the nature of partisanship in the American public. In doing so, she made significant contributions to the literature on time series and helped advance the use of event history analysis in political science. Time series techniques are appropriate when one has observations on something over regularly spaced intervals (e.g., quarterly presidential approval over 30 years) and wants to understand the properties of that series, and/or how different series influence one another. Event history modeling, sometimes called survival analysis, is the right tool when timing itself is of central interest—that is, when the researcher wants to gain purchase on when something will happen, and what factors, including previous events, facilitate/mitigate the risk of that event happening (e.g., a military conflict, an election being called, when politicians make announcements).

All empirical modeling involves making assumptions. However, those assumptions can be problematic if they are overly restrictive or researchers adopt them blindly— they can limit the questions researchers dare to ask and pursue and lead to erroneous substantive conclusions. One of the themes that emerges across Jan’s methodological work, particularly her contributions as they relate to time, involves the questioning of assumptions and the introduction of more flexible alternatives for dealing with problems. In doing so, time and time again she has cleared the way for researchers to explore new substantive avenues. Commenting on this ability to anticipate questions and unlock research agendas, Barry Burden notes that “Jan has been at the frontier of statistical methods in the social sciences before it was cool. As one of a few senior women in the formative days of the political methodology field, she was a leader in bringing time series, event history, and other techniques into mainstream use in the discipline.”

With time series analysis, scholars often want to know whether an event, or “shock” (e.g., the attacks of 9/11), has a long or a short-term effect on a series. Does a jump in presidential approval remain for a long time after a sudden crisis? Does it fade quickly? Put in the language of the literature we might ask, “what kind of persistence and memory does presidential approval exhibit?” In order to properly model a series, a scholar needs to know whether the data is stationary, meaning that the mean and variance of the series is constant over time.Footnote 2 Understanding whether this is the case is important, as a non-stationary—or integrated—series violates basic regression assumptions; analysis without additional steps can lead to spurious inferences. It’s also important because the choice of model and procedure depends on the assumptions a scholar makes about stationarity, which in turn dictates one’s inferences about the persistence and memory of a political process.

In the mid-90s, political (and other social) scientists often treated the question of whether a series was stationary in blunt terms, viewing this (necessarily) as a “yes/no” question. If scholars assumed a series was stationary, they might choose a specification that assumed external shocks have a short-term effect before returning to a mean valueFootnote 3; if scholars assumed the data was non-stationary, they would first difference their data, and then choose a model that assumed that shocks can have effects for a long time, but do not return to a mean value.Footnote 4 This dichotomous choice between modeling strategies meant that scholars had little flexibility in understanding long-term political processes, and that they might come to dramatically different conclusions about the way the world works depending on (sometimes arbitrary) decisions between these “two sizes fit all.”

In her 1996 article in the American Political Science Review (“The Dynamics of Aggregate Partisanship”), Jan (and her coauthor Renee Smith) introduced the idea of fractional integration to political science, arguing that scholars should reject the choice between treating their time series data as either stationary or integrated.Footnote 5 Moreover, she made this major methodological statement while addressing several long-standing debates over partisanship in the United States. Proponents of realignment theory had long argued that shifts in partisanship—in response to party performance and major events—lasted decades. Meanwhile, scholars of political behavior debated the stability of party identification, whether taking more micro- or macro-approaches to the question; they also puzzled over how to best reconcile research on individuals and aggregates.

Jan carefully identified the theoretical premises noted in the micro-level literature—notably, she discussed and incorporated the potential for substantial heterogeneity in behavior across individuals (an issue that she has continued to focus on, and that has received increased attention among behavioralists in the 20-plus years since her article’s publication). She then reported a set of expectations for what kinds of aggregate outcomes would be consistent with the deductions following from these premises. Analyzing measures of aggregate partisanship, she found such series to be fractionally integrated, meaning that shocks to them should last on the order of years, not months (as some previous work had concluded) or decades (as the party system literature had insisted). In their 1998 paper in the American Journal of Political Science, Jan and Renee Smith expanded upon the argument for fractional integration techniques; their primer on the method made a persuasive case for the widespread use of such techniques in political science: such models allow scholars to capture long term dynamics while permitting a series to return to a mean value; they are likely the right models for much political science data due to scholars aggregating across heterogeneous populations.

At the same time Jan was changing how political scientists model time series data, she was also calling attention to how scholars could better address questions involving “the number, timing and sequence of changes in a variable of interest” (Box-Steffensmeier and Jones Reference Box-Steffensmeier and Jones1997, 1414). Indeed, another set of ideas proposed in Jan’s dissertation would make its way into the American Journal of Political Science (1996) as “A Dynamic Analysis of the Role of War Chests in Campaign Strategy.” Looking at FEC reporting data for nearly 400 House races (spanning the 1990 election cycle, from its beginning through the primaries), she focused on the issue of when money matters, finding that large incumbent war chests deter high-quality challengers. The paper not only highlighted the need for scholars to focus more attention on temporal dynamics when studying campaigns and elections, but helped introduce event history analysis to a discipline that had had little exposure to such techniques up to that point.

In the following year, Jan (and her coauthor Brad Jones) provided an expanded overview of survival modeling in “Time is of the Essence: Event History Models in Political Science” (American Journal of Political Science [1997]). In the piece they worked through several substantive applications, prioritizing the intuition, set-up, and substantive interpretation of event history models. An especially helpful aspect of the paper was that it drew parallels and contrasts with the generalized linear modeling techniques with which many in the discipline were more familiar at the time.

That piece (recently reprinted in Advances in Political Methodology [2017]) is characteristic of Jan’s approach to teaching and researching in methods —it is accessible without sacrificing rigor; it is focused on helping people with different levels of training and mathematical comfort make the transition from theory to practice. It is also consistent with the aforementioned theme of questioning assumptions and introducing flexible alternatives, for the article makes it clear how different modeling choices should be guided by theory. In specifying event history models, researchers have to account for time dependence.Footnote 6 If one has a reason to suspect that this dependence should take on a particular form, then certain parametric choices may make sense. If one does not have a priori expectations about what this should look like, then semi-parametric approaches like the Cox proportional hazards model may hold advantages and help the researcher avoid mischaracterizing a process. Putting this advice to immediate use, in that same year Jan would publish an article in the American Political Science Review (with Laura Arnold and Chris Zorn) that used a Cox semi-parametric model to understand the timing of position taking by members of Congress on the North American Free Trade Agreement (NAFTA).

In the late 1990s and early 2000s, Jan continued to advance our understanding of time in politics, publishing papers— often in “the top-three”—that used time series analysis to further conversations about mass partisanship in the United States (e.g., Box-Steffensmeier et al. Reference Box-Steffensmeier, Knight and Sigelman1998; Box-Steffensmeier and De Boef Reference Box-Steffensmeier and De Boef2001), and that elucidated extensions to event history approaches (e.g., Box-Steffensmeier and Zorn Reference Box-Steffensmeier and Zorn2001, Reference Box-Steffensmeier and Zorn2002). Jan was promoted to associate professor in 1998, and by that point had laid the foundations for the other parts of her record that have come to distinguish her career: she had an established and ever-expanding network of collaborators, extensive external grant support (most notably, from the National Science Foundation), high citation counts, and was actively involved in organized sections of the Midwest Political Science Association and American Political Science Association. In 2001 she was named the top (“emerging”) scholar within 10 years of her PhD by the Elections Public Opinion and Voting Behavior (EPOVB) section of the APSA.

Of course, by this point she also had an impressive record of publishing with and helping graduate students, and a reputation for experimenting with innovative teaching methods that promote accessibility and support students who need flexibility (whether due to work/life commitments, or issues of disability/mobility). Jan was heavily involved in the founding and development of what is now called PoliSci U in Advanced Methodology—a cooperative venture between the CIC (Committee on Institutional Cooperation) member-institutions of the University of Illinois, the University of Minnesota, the University of Wisconsin-Madison, and The Ohio State University. The program started in the mid-1990s with interactive (ITV) video courses, and Jan team-taught the first one—on time series—with John Freeman (she continues to teach in the program). During this period Jan was also involved in other video course offerings (e.g., with Brad Jones, on event history), published on ways for technology to support undergraduate education (Box-Steffensmeier et al. Reference Box-Steffensmeier, Grant, Meinke and Tomlinson2000), and by the early 2000s had even developed online courses, anticipating this trend by many, many years.


Entering this period as a relatively young endowed chair, the next ten years of Jan’s career would be equally productive, but even more multifaceted. Jan continued to publish in top political science outlets, steadily pushing forward the literature on time series while addressing topics like “The Dynamics of the Partisan Gender Gap” (Box-Steffensmeier, Linn, and Lin Reference Box-Steffensmeier, De Boef and Lin2004) and “The Aggregate Dynamics of Campaigns” (Box-Steffensmeier et al. Reference Box-Steffensmeier, Darmofal and Farrell2009).Footnote 7 These pieces yielded important insights for scholars studying behavior and elections in the American context. For example, the paper with Suzie Linn and Tse-Min Lin both added to and qualified the seemingly ubiquitous discussions of “gaps” in the electorate that had emerged by the mid-2000s. While pundits, politicos and even many scholars discuss gender (and other) gaps without context and with little aim of explanation, Jan and colleagues took the long view on such questions, linked socioeconomic factors to observed patterns, and were able to give informed speculation on future dynamics.

At the same time, some of Jan’s scholarly work in these years started to take on a different hue—one that still mirrored the shape of her interests, but that leveraged her perspective and experience and sought to make varied, but no less wide-ranging, impacts. For instance, during this period she coedited three volumes, including the Oxford Handbook of Political Methodology (2008), and authored ten chapters for various edited volumes (often continuing to give graduate students valuable experience as coauthors). However, she also engaged in increasingly interdisciplinary conversations, becoming a professor of sociology (by courtesy) and an affiliate of the Institute for Population Research at Ohio State. Jan’s scholarship reflected her growing reputation in the larger academic community, as she published in venues beyond the typical political science outlets (e.g., Meat Science; see Note 1). Her 2004 Cambridge University Press title Event History Modeling: A Guide for Social Scientists (coauthored with Brad Jones) was aimed at a broad audience. The book stands as perhaps the comprehensive text for anyone interested in pursuing survival analysis, walking the interested reader through data set-up, model selection, interpretation, visualization, and diagnostics; its thousands of citations reflect its considerable impact across the social and behavioral sciences. Likewise, her work in developing the conditional frailty model for event history analysis (with Suzie Linn and Kyle Joyce)—a model pairing the flexibility of the Cox semi-parametric approach while accounting for unobserved heterogeneity among subjects and dependencies across repeated events (problems that often present together in real-world applications)— found a home not just in Political Analysis (2007), but Statistics and Medicine (2006).

Towards the end of these years Jan also pushed into new areas of inquiry, sketching out an ambitious research agenda focused on interest group coordination and influence. Her approach to these topics was innovative, as she conceptualized the interest group universe in network terms—as actors linked together (or not) by their legal activism, and more specifically, their co-signing of amicus curiae briefs submitted to the US Supreme Court. Per usual, Jan was asking the right questions at the right time, as in the mid-to-late 2000s network analysis was having “a moment” in the social and behavioral sciences. Political scientists who had been interested in such questions, theories, and methods found one another, first through a couple of initial summer meetings,Footnote 8 and then through the founding of an organized section of the APSA (“Political Networks”). Jan quickly became heavily involved in this community—one with a decidedly interdisciplinary flavor, and many of whose members are characterized by a mix of substantive and methodological interests. And, true to form, she also quickly emerged as a leader in the field, obtaining NSF funding to support her study of interest group networks (an ongoing collaboration with Dino Christenson), while also securing NSF support for the section to be able to hold its summer meeting and methodological workshops.

Jan’s efforts outside the journals during this time also reflected her values, priorities and long-held desire to help others and facilitate academic community. Across these years she was a constant presence in the political research laboratory in the department of political science at Ohio State, serving as either associate director or director; faculty and countless graduate and undergraduate students benefitted from her advocacy and oversight of the lab’s computing and other academic support services. In 2003 she created the Program in Statistics and Methods (PRISM) in the department of political science, providing stipends for research fellows and additional methods advice/training for faculty and graduate students alike.

Jan also took steps to build (and change) institutions to further support mentorship, particularly of women and minorities working in political methodology. In talking about the inspiration she drew from her dissertation adviser, Jan also notes how crushed she felt when that female mentor left the discipline, and remembers receiving less than encouraging advice early on about being a woman in academe (and about work/life balance). As a graduate student she recalls a prominent, tenured older woman telling her that it wasn’t possible to be a mother and professor. “Same story when I got to OSU, a senior woman administrator actually told me not to have kids until after tenure!” Jan is quick to point out that she was fortunate to have many mentors when she first arrived at Ohio State.Footnote 9 Still, she began and advanced her career in a political methodology community—and larger discipline—that was very white and male-dominated (and remains so). Moved to give back, she has worked consistently to create opportunities and to change norms and demographics in different corners of the discipline.

As the first female president of the political methodology section of the APSA (2005–2007), Jan oversaw a substantial expansion of the summer meeting (and with it, increased female attendance). While in this leadership spot she also directed the creation of the section’s Diversity Committee (2006), and helped secure funding from the National Science Foundation to support new initiatives like the ongoing Visions in Methodology (VIM) workshops for female methodologists. Jan co-hosted the first Visions meeting at Ohio State (with Corrine McConnaughy) and developed its basic structure, which includes scholarly presentations as well as readings and discussions related to gender and careers in academe. Of these and Jan’s other efforts to support women and minorities in the methods community (and the discipline more broadly), Sara Mitchell and Caroline Tolbert note that “[a]mong women in political methodology and American politics nationally, Jan stands alone as the leader… It is hard for words to convey how highly we evaluate Jan.” Suzie Linn adds that “[s]he works tirelessly as a mentor to students scattered all over the country and to advance the opportunities available to women and scholars in underrepresented groups. Her research has impacted the way scholars in multiple fields do their work and her many leadership roles in the profession have made our discipline stronger.”

Indeed, during this period Jan would take on many such leadership roles, from being an academic program reviewer, to completing multiple terms on the board of the Dirksen Congressional Center, to serving in myriad capacities for the National Science Foundation—these included as a member of the Committee of Visitors for the political science program (2010), a convener and selection chair for a program aimed at getting political science methods into middle and high school curricula (2012–13), and a committee member on the search for the assistant director of the Social, Behavioral and Economic Sciences Division (2012–13). Jan also served on multiple editorial advisory boards, as the series editor of the Legislative Politics and Policymaking series with the University of Michigan Press, as a member of the advisory committee to the ICPSR summer program (2007–2010), as an editorial board member for the American Political Science Review, American Journal of Political Science, and Journal of Politics (at one period appearing on all three simultaneously), and as an associate editor of the American Journal of Political Science (2006–2009). In 2010 she was elected president of the Midwest Political Science Association, serving as president-elect in 2010, president in 2011, and immediate past-president in 2012.

During this time Jan would also be honored for her many contributions to the fields of political methodology and American politics, as well as for her impacts in the classroom. In 2012 Ohio State presented her with a distinguished scholar award, followed by a distinguished teaching award in 2013. That same year Jan was named the winner of the Warren E. Miller Award for Meritorious Service to the Social Sciences by the Inter-University Consortium for Political and Social Research (ICPSR). And, last but not least, in 2013, Jan was also given a career-achievement award by the Society for Political Methodology—a recognition of her 20-plus years of accomplishments in the field, including twice winning the Gosnell prize for the best work presented in political methodology, being named an inaugural fellow of the society (2008), and previously having had an ICPSR graduate student award named after her.


The phrase “productive administrator” may strike some as an oxymoron. However, it is a perfect—if insufficient—description of Jan’s career since 2014. In that year Jan moved desks at Ohio State, becoming Dean for the Social and Behavioral Sciences and Graduate Dean for Arts and Sciences. She would hold these positions until she moved into a year of service as Interim Executive Dean and Vice Provost for the College of Arts and Sciences, and lead Dean for Translational Data Analytics (2018–2019). The Social and Behavioral Sciences division at Ohio State has eight departments, all of which are ranked in the top-25 nationally; the College of Arts & Sciences consists of 38 departments and schools that are home to 17,000 undergraduates, 2,500 graduate students, and an annual operating budget of roughly $250 million dollars. In July of 2019, Jan would choose to return to the department of political science, remarking that “[it] was fun and exciting to work in administration at OSU, but also took me away from my love of working closely with students on their research and being a teacher. There are few things as rewarding to me as seeing a student light up when understanding a concept, getting their code to work, or finding results.”

Of her years in university leadership, Jan states that:

“My vision was to create an intellectual community that no one wanted to leave. After all, that was why I stayed at Ohio State for my entire professional life—I was part of a department that was always striving to be one of the best. I wanted to leverage our diversity, tout our inclusiveness, and celebrate our excellence, because our faculty, staff, and students are generating knowledge that changes the world.”

By all accounts, Jan’s time in administration at Ohio State could only be characterized as an unqualified success. In a letter nominating her for Distinguished University Professor, Professors Greg Caldeira and John Casterline describe Jan’s service to the university as marked by inclusion and clear communication: “The emphasis on transparency and dialogue was to build trust, improve morale, lower hierarchy, and improve the culture of the college. She also dedicated time and energy to building needed partnerships and advocacy for policy change. How decisions occur in the college was remade, built on a structure that embodies collaboration and teamwork.”

Of course, while helping run one of the largest institutions in higher education, Jan was also incredibly productive on another front, somehow keeping her research agenda not just active, but as successful as at any period in her already illustrious career. In 2014 she published a second methods text with Cambridge University Press, Time Series Analysis for the Social Sciences. Coauthored with John Freeman, Jon Pevehouse, and Matt Hitt, the book encapsulates the wisdom accrued through the team’s decades of collective experience teaching and researching on time series topics; it is nothing short of a definitive statement on the methodology, in the same way Jan’s book on event history was a decade prior.

While still addressing the state of the art in time series (e.g., Box-Steffensmeier and Helgason Reference Box-Steffensmeier and Helgason2016) and event history analysis (e.g., Box-Steffensmeier, Linn, and Smidt Reference Box-Steffensmeier, Linn and Smidt2014), during these years Jan’s work on interest group networks also advanced rapidly on multiple fronts, giving scholars not just insights into what such networks look like across industries and periods (e.g., Box-Steffensmeier and Christenson Reference Box-Steffensmeier and Christenson2015), but how they matter for different branches of government. For example, in a piece in the American Political Science Review, Jan and her coauthors use the position of interest groups in the amici network as a measure of power, finding that groups influence judicial decision making when an equal number of briefs have been filed on both sides of a case (Box-Steffensmeier et al. Reference Box-Steffensmeier, Christenson and Hitt2013). In a paper in the AJPS, they turn their attention from the Judiciary to Congress, finding that Dear Colleague letters from more central interest groups serve as valuable cues to members of Congress and encourage the co-sponsorship of bills early in the legislative process (Box-Steffensmeier et al. Reference Box‐Steffensmeier, Christenson and Craig2019). At the same time, while providing these substantive insights, Jan and colleagues have developed innovative methods for modeling network data, proposing a model and R package— the frailty exponential random graph model (FERGM)—for handling unobserved heterogeneity (Box-Steffensmeier et al. 2017). A key assumption of exponential random graph modeling is that our models are correctly specified. Consistent in her advice as ever, Jan’s paper reminds us why we should be wary of this assumption, and then provides a flexible alternative for moving forward.


In her roughly 30 year career, Jan has published two books and nearly 50 peer-reviewed papers; she has edited multiple volumes, obtained millions in grant funding, and garnered well over 8,000 citations.Footnote 10 Additionally, she has served in nearly every role imaginable in the discipline, and as a high-level administrator. In 2017 she was elected a Fellow of the American Academy of Arts and Sciences.

And yet, if you ask Jan what she is most proud of, she will quickly say “the relationships—being able to help others. Paying it forward and paying back the mentors who made all the difference in the world to me along my educational journey.” As she takes on the presidency of the APSA, Jan is excited about serving, but also aware of the challenges that we face in the present. Still, she expresses a confidence and optimism that those of us who know her have come to expect:

“More than ever, political science is positioned to address pressing questions of this moment and beyond, provided we embrace and promote the rich intellectual pluralism of our discipline—in methodology, methods, behavior, institutions, and perspective. In addition, we must recognize that that the diversity of our scholars in terms of racial and ethnic background, nationality, gender, sexuality, gender expression, institutions, and professional career stage contributes to knowledge and ways of understanding the world. It is also important to be inclusive of all political scientists working on critical questions whose careers are in industry, government, non-profits, and academia. Our association will be more vibrant when being more inclusive of all political science career routes. My goal is to further bring people together to celebrate the heterogeneity of approaches and to advance topics of diversity and inclusion.”

Jan’s scholarship has moved literatures and set agendas. Her teaching has inspired countless students. Her mentorship has helped make careers. Her leadership has promoted openness and positive transformation. In short, over three decades Jan has done nothing but make a difference. Another Fellow of the American Academy of Arts and Sciences, Gary King describes Jan this way:

“With a disarming and understated style, Jan has remade important parts of the scholarly literatures in American politics and political methodology, all the while managing to be a role model for the rest of us in teaching, mentorship, community service, and administrative leadership. I’ve learned so much from Jan; I only wish I could learn how to convince her to move so we could be colleagues, an effort I’ve now failed at four times!”

Fortunately, the career of Janet M. Box-Steffensmeier demonstrates that we don’t have to be in Columbus to benefit from her wisdom—Jan is truly a colleague to us all. ■

Supplementary Materials

To view supplementary material for this article, please visit


The title of this article refers to “taking a difference” or “differencing,” which is a technique used when time series data are non-stationary.

1. You should follow the advice you’ve heard about cooking pork tenderloin at a lower temperature – Jan and colleagues’ articles in Meat Science (Moeller et al. Reference Moeller, Miller, Edwards, Zerby, Logan, Aldredge, Stahl, Boggess and Box-Steffensmeier2010a, Reference Moeller, Miller, Aldredge, Logan, Edwards, Zerby, Boggess, Box-Steffensmeier and Stahl2010b) present compelling data and evidence on this point.

2. This is an informal explanation. For a more complete discussion, see Box-Steffensmeier, Freeman, Hitt, and Pevehouse (Reference Box-Steffensmeier, Linn and Smidt2014).

3. When scholars assumed data were stationary, they often used a basic AutoRegressive Moving Average (ARMA) specification. In time series notation, this meant they assumed d=0.

4. “Differencing,” or “taking a difference” of a series is accomplished by subtracting an observation in the previous period from an observation in the current period. When scholars would difference a series prior to analysis, they then employed an AutoRegressive Integrated Moving Average (ARIMA) specification. In time series notation, this meant they assumed d=1.

5. That is, they argued that d should be allowed to take on any value between 0 and 1, and that scholars should estimate d to obtain an “objective measure of the properties of a time series rather than the subjective information from visual inspection of the autocorrelation function or from often contradictory diagnostic tests” (1996: 667–68). Fractional integration specifications are commonly called referred to as AutoRegressive Fractionally Integrated Moving Average (ARFIMA) models.

6. In event history modeling, time dependence is accounted for with what’s called a “baseline hazard.” A researcher can select a model that specifies a distributional form for the baseline hazard, or take an approach that avoids making assumptions about its shape (e.g., the Cox Proportional Hazards model).

7. This paper was named top article in the Journal of Politics for 2009.

8. These summer meetings—which would become the annual “PolNet” Conference—were the brainchild of David Lazer and James Fowler.

9. For example, she notes that “Nothing I published pre-tenure had not been read in detail by Paul Beck and Herb Weisberg. How amazing is that?!”

10. This according to Google Scholar.


Box-Steffensmeier, Janet M. 1994. “Candidates, contributors, and campaign strategy: It’s about time.” Dissertation. University of Texas at Austin.Google Scholar
Box-Steffensmeier, Janet M., and Lin, Tse-Min. 1996. “A dynamic model of campaign spending in congressional elections.” Political Analysis 6 (1): 3766.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M. 1996. “A Dynamic Analysis of the Role of War Chests in Campaign Strategy.” American Journal of Political Science 40 (2): 352–71.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., and Smith, Renee M.. 1996. “The dynamics of aggregate partisanship.” American Political Science Review 90 (3): 567–80.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., Arnold, Laura W., and Zorn, Christopher J.W.. 1997. “The strategic timing of position taking in Congress: A study of the North American Free Trade Agreement.” American Political Science Review 91 (2): 324–38.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., and Jones, Bradford S.. 1997. “Time is of the essence: Event history models in political science.” American Journal of Political Science 41 (4): 1414–61.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., and Smith, Renee M.. 1998. “Investigating political dynamics using fractional integration methods.” American Journal of Political Science 42 (2): 661–89.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., Knight, Kathleen, and Sigelman, Lee. 1998. “The Interplay of Partisanship and Ideology: A Time-Series Analysis.” Journal of Politics 60 (4): 1044–62.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., Grant, Tobin J., Meinke, Scott, and Tomlinson, Andrew R.. 2000. “Virtual Field Trips: Bringing College Students and Policymakers Together through Interactive Technology.” PS: Political Science & Politics 33 (4): 829–34.Google Scholar
Box-Steffensmeier, Janet M., and De Boef, Suzanna. 2001. “Macropartisanship and Macroideology in the Sophisticated Electorate.” The Journal of Politics 63 (1): 232–48.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., and Zorn, Christopher J.W.. 2001. “Duration Models and Proportional Hazards in Political Science.” American Journal of Political Science 45 (4): 951–67.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., and Zorn, Christopher J.W.. 2002. “Duration Models for Repeated Events.” The Journal of Politics 64 (4): 1069–94.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., and Jones, Bradford S.. 2004. Event history modeling: A guide for social scientists. New York City: Cambridge University Press.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., De Boef, Suzanna, and Lin, Tse-Min. 2004. “The dynamics of the partisan gender gap.” American Political Science Review 98 (3): 515–28.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., and De Boef, Suzanna. 2006. “Repeated Events Survival Models: The Conditional Frailty Model.” Statistics in Medicine 25 (20): 3518–33.CrossRefGoogle ScholarPubMed
Box-Steffensmeier, Janet M., De Boef, Suzanna, and Joyce, Kyle A.. 2007. “Event dependence and heterogeneity in duration models: the conditional frailty model.” Political Analysis 15 (3): 237–56.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., Darmofal, David, and Farrell, Christian A.. 2009. “The aggregate dynamics of campaigns.” The Journal of Politics 71 (1): 309–23.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., Christenson, Dino P., and Hitt, Matthew P.. 2013. “Quality over quantity: Amici influence and judicial decision making.” American Political Science Review 107 (3): 446–60.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., Freeman, John R., Hitt, Matthew P., and Pevehouse, Jon C.W.. 2014. Time Series Analysis for the Social Sciences. New York City: Cambridge University Press.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., Linn, Suzanna, and Smidt, Corwin D.. 2014. “Analyzing the Robustness of Semi-Parametric Duration Models for the Study of Repeated Events.” Political Analysis 22 (2): 183204.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., and Christenson, Dino P.. 2015. “Comparing Membership Interest Group Networks Across Space and Time, Size, Issue and Industry.” Network Science 3 (1): 7897.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., and Helgason, Agnar F.. 2016. “Introduction to Symposium on Time Series Error Correction Methods in Political Science.” Political Analysis 24 (1): 12.CrossRefGoogle Scholar
Box-Steffensmeier, Janet M., Christenson, Dino P., and Morgan, Jason W.. 2018. “Modeling unobserved heterogeneity in social networks with the frailty exponential random graph model.” Political Analysis 26 (1): 319.CrossRefGoogle Scholar
Box‐Steffensmeier, Janet M., Christenson, Dino P., and Craig, Alison W.. 2019. “Cue‐Taking in Congress: Interest Group Signals from Dear Colleague Letters.” American Journal of Political Science 63 (1): 163–80.CrossRefGoogle Scholar
Franzese, Robert J., editor. 2017. Advances in Political Methodology. Northhampton, MA: Edward Elgar Publishing.CrossRefGoogle Scholar
Moeller, S. J., Miller, R. K., Edwards, K. K., Zerby, H. N., Logan, K. E., Aldredge, T. L., Stahl, C. A., Boggess, M., and Box-Steffensmeier, J. M.. 2010a. “Consumer perceptions of pork eating quality as affected by pork quality attributes and end-point cooked temperature.” Meat Science 84 (1): 1422.CrossRefGoogle Scholar
Moeller, S. J., Miller, R. K., Aldredge, T. L., Logan, K. E., Edwards, K. K., Zerby, H. N., Boggess, M., Box-Steffensmeier, J. M., and Stahl, C. A.. 2010b. “Trained sensory perception of pork eating quality as affected by fresh and cooked pork quality attributes and end-point cooked temperature.” Meat Science 85 (1): 96103.CrossRefGoogle Scholar

Altmetric attention score

Full text views

Full text views reflects PDF downloads, PDFs sent to Google Drive, Dropbox and Kindle and HTML full text views.

Total number of HTML views: 114
Total number of PDF views: 184 *
View data table for this chart

* Views captured on Cambridge Core between 16th October 2020 - 4th March 2021. This data will be updated every 24 hours.


Send article to Kindle

To send this article to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

Note you can select to send to either the or variations. ‘’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Making (and Sometimes Taking) a Difference: The Dynamic Career of Janet M. Box-Steffensmeier
Available formats

Send article to Dropbox

To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

Making (and Sometimes Taking) a Difference: The Dynamic Career of Janet M. Box-Steffensmeier
Available formats

Send article to Google Drive

To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

Making (and Sometimes Taking) a Difference: The Dynamic Career of Janet M. Box-Steffensmeier
Available formats

Reply to: Submit a response

Your details

Conflicting interests

Do you have any conflicting interests? *