is owned and operated by the Economic History Association
with the support of other sponsoring organizations.

History by Numbers: An Introduction to Quantitative Approaches

Author(s):Hudson, Pat
Ishizu, Mina
Reviewer(s):Roberts, Evan

Published by EH.Net (May 2017)

Pat Hudson and Mina Ishizu, History by Numbers: An Introduction to Quantitative Approaches (second edition).  London: Bloomsbury, 2017. xx + 339 pp. $35 (paperback), ISBN: 978-1-84966-537-7.

Reviewed for EH.Net by Evan Roberts, Department of Sociology, University of Minnesota.

For a field on the wane (see the discussion in Historically Speaking in 2010) quantitative history has a good stock of textbooks, but the flow of new entrants is slow. The revised edition of History by Numbers — originally solo-authored by Pat Hudson in 2000 and now co-authored with Mina Ishizu in its 2017 second edition — would make an excellent textbook for an upper division undergraduate history class. Anyone wishing to extend their students and themselves could add Charles Feinstein and Mark Thomas’s Making History Count for a more advanced undergraduate or graduate-level class in quantitative history. Of course, this supposes there is sufficient demand among history students to enroll in such a class, a critical question I return to later in the review.

The presumed audience for History by Numbers is history rather than economics students. Chapter 1 situates quantitative history within the discipline of history, while Chapter 2 discusses the nineteenth-century statistical movements in Britain and the epistemology of quantitative reasoning. Economics students would benefit from reading these chapters, yet they fit more squarely within the design of a history course and curriculum. The remainder of the book takes students and their instructors through a standard sequence of data management, exploration and analysis, written with the presumption that students begin with only a memory of high school mathematics. The exploratory data analysis chapters focus on graphical methods, and how to characterize the distribution of a single variable. Examples in all the chapters come from the published literature, but with a hand on the scale for modern British economic and social history. British universities have been more successful than their North American peers in maintaining quantitative work within history departments, so the choice of examples reflects the state of the field.

The chapter on time series and indices is particularly strong, with clear worked explanations of how to construct indices and the art of choosing the right base period. There is a very good explanation of cyclical fluctuations and seasonality, and how to work with this form of data correctly. The graphics and the text complement each other especially well here. Regression and correlation is then covered in one chapter, integrating both time series and cross-sectional data. For an introductory course this is appropriate, and the separation of cross section and time series approaches can await students in a subsequent course, where Feinstein and Thomas’s more advanced text separates these issues. Sampling and significance testing in Chapter 7 wrap up the purely statistical chapters, with more worked examples from published research. Chapter 8 outlines how economic historians put statistical methods to work in building models and testing theories, putting cliometrics in perspective for history students. It is not until Chapter 9 that we get to the foundational work of doing History by Numbers: getting numbers from manuscripts into the computer. Again, the authors do well in putting recent developments in creating quantitative data into historical perspective for a generation of students who have grown up in an era where important tools of quantitative history: digital cameras, laptops, databases, geographic information systems, and fuzzy matching of text are ubiquitous in their daily lives.

History by Numbers would make an excellent textbook for a course introducing students to quantitative methods in the context of historical examples, particularly for instructors who would integrate that introduction with a history of the British industrial revolution. A more general and topically eclectic course on quantitative history could also use this book as a core text with very little modification, as it discusses research on topics ranging from violence in nineteenth-century New Zealand to the wolfram market in World War II Spain. Two sections of discussion questions integrate the worked statistical examples in the text with substantive historical and economic questions, and could be used as the basis for labs or tutorials.

Alternatively, instructors might use History by Numbers as their secret guide to teaching statistics to history students without assigning it as a text (I apologize to the authors for the significant reduction in their royalties this implies). Indeed, this was how I used the previous edition of the book; as a script for teaching statistics to history students in a social history class using entirely American data. With IPUMS, EH.Net, and Historical Statistics of the United States it is not hard to find data that can be used to teach all the statistical concepts introduced in History by Numbers. Bearing in mind that book authors face harder constraints on length than online book reviewers whose words come cheap, there are some omissions in an otherwise quite comprehensive survey of modern quantitative history. There could be greater discussion of longitudinal data from cohort studies which are mentioned just briefly at the end of the text. Moreover, in an otherwise diverse set of articles for discussion I was surprised to see none from the strong Scandinavian quantitative history tradition (largely written in English these days).

Overall then there is much to recommend in History by Numbers for instructors whose goal is to teach statistics using relatively clean data from already digitized sources. Working with small clean datasets allows students to focus on learning statistics, statistical software, and gaining confidence in making statistical inferences. The next step in students’ development, directly suggested by Chapter 9, is writing an Honors or senior thesis using methods from the book, and available sources tailored to student and instructor interest.  This supposes that students have been motivated to first take a course guided by History by Numbers. Yet, as the authors note quantitative history has not grown since the first edition was published.

Perhaps we are doing it wrong, and need to rethink how we introduce students to quantitative history. The same arguments apply mutatis mutandis to introducing students to quantification in other undergraduate social science programs such as sociology, political science, and anthropology. In all of these disciplines a meaningful fraction of students approach quantitative methods with some anxiety. In the same spirit as Joshua Angrist and Jörn-Steffen Pischke (2017) recommend changing the traditional sequence of introductory econometrics courses to reflect changes in empirical econometric practice, quantitative historians should also introduce students to their practices “by example rather than abstraction.”

For historians and their kin in economics and sociology departments, teaching by example means beginning with primary sources. It is now straightforward to lead with the sources, to begin where students are found, a little shy of quantification but probably willing to enroll in a class that offers an introduction to research methods and an immersion in primary sources. Leading with immersion in primary sources meets the modal student closer to their interests, and can be a powerful recruitment tool for a class, compared to others built around textbooks and reading. Tools for transcribing data can now be easily built using Google tools, or the Zooniverse Project Builder (  Such a course could also be pitched to students as “digital history.” The fashion for attracting students with “digital” may wane in the future, but the pedagogical underpinnings of beginning with collaborative digitization of sources (quantifiable sources!) are sound.

When students get their hands metaphorically dirty with the sources, see that their small sample of sources differs internally, and differs from their classmates’ sample of sources, the motivation to investigate questions statistically comes more organically.  They can then learn statistics with data they have created. Fitting data collection and analysis into one semester requires compressing analysis somewhat, the omission depending on the manuscripts and data at hand. If the data are cross sectional surveys, for example, index numbers may be omitted. My experience has been that beginning with the manuscript sources, creating a small dataset, and then analyzing it, leads to the greatest engagement from students who initially lack confidence in quantitative methods. Such an approach punts some of the statistics taught in History by Numbers to later semesters, but with the benefit of having engaged more students in quantification than a course framed explicitly as quantitative history.

Thus, the conclusion that History by Numbers is an excellent text for an upper division class is premised on the existence of a sophomore course that introduces students to quantitative methods quietly, and maybe just a little surreptitiously. Call it digital history, call it social history or historical sociology, whatever topic you think will attract students at your institution, and ideally the students will be so far into the course that they won’t realize they’re learning statistics until you’ve shown them they can do it. As Hudson and Ishizu emphasize many students have more of an aptitude for the common-sense tools of statistics than they realize. In short, while I can recommend History by Numbers to social and economic historians, I also recommend that we think carefully and creatively about how our curricular sequences can bring more students into quantitative history courses.


Joshua D. Angrist and Jörn-Steffen Pischke. 2017. “Undergraduate Econometrics Instruction: Through Our Classes, Darkly,” Journal of Economic Perspectives 31(2): 125-44.

Charles H. Feinstein and Mark Thomas. 2002. Making History Count: A Primer in Quantitative Methods for Historians. New York: Cambridge University Press.

Robert Whaples. 2010. “Is Economic History a Neglected Field of Study?” Historically Speaking 11(2): 17-20.

Evan Roberts is Assistant Professor of Sociology and Population Studies at the University of Minnesota. His best enrolling quantitative methods class went by the title “Living, Working and Dying in Chicago.” Recent publications include “Family Structure and Childhood Anthropometry in Saint Paul, Minnesota in 1918” (with John Robert Warren), History of the Family.

Copyright (c) 2017 by EH.Net. All rights reserved. This work may be copied for non-profit educational uses if proper credit is given to the author and the list. For other permission, please contact the EH.Net Administrator ( Published by EH.Net (May 2017). All EH.Net reviews are archived at

Subject(s):Development of the Economic History Discipline: Historiography; Sources and Methods