Comparative historical research is a method of social science that examines historical events in order to create explanations that are valid beyond a particular time and place, either by direct comparison to other historical events, theory building, or reference to the present day. Generally, it involves comparisons of social processes across times and places. It overlaps with historical sociology. While the disciplines of history and sociology have always been connected, they have connected in different ways at different times. This form of research may use any of several theoretical orientations. It is distinguished by the types of questions it asks, not the theoretical framework it employs.

Some commentators have identified three waves of historical comparative research. The first wave of historical comparative research concerned how societies came to be modern, i.e. based on individual and rational action, with exact definitions varying widely. Some of the major researchers in this mode were Alexis de Tocqueville, Karl Marx, Emile Durkheim, Max Weber, and W.E.B. Du Bois. The second wave reacted to a perceived ahistorical body of theory and sought to show how social systems were not static, but developed over time. Notable authors of this wave include Reinhard Bendix, Barrington Moore, Jr., Stein Rokkan, Theda Skocpol, Charles Tilly, Michael Mann, and Mark Gould. Some have placed the Annales school and Pierre Bourdieu in this general group, despite their stylistic differences. The current wave of historical comparative research sociology is often but not exclusively post-structural in its theoretical orientation. Influential current authors include Julia Adams, Ann Laura Stoler, Philip Gorski, and James Mahoney.

Moore's Social Origins of Dictatorship and Democracy influenced Daron Acemoglu and James A. Robinson to apply comparative methods to economic history in a 2006 book called Economic Origins of Dictatorship and Democracy. In 2024, Acemoglu, Robinson, and Simon Johnson won the Nobel Memorial Prize in Economic Sciences for their work.

There are four major methods that researchers use to collect historical data. These are archival data, secondary sources, running records, and recollections. The archival data, or primary sources, are typically the resources that researchers rely most heavily on. Archival data includes official documents and other items that would be found in archives, museums, etc. Secondary sources are the works of other historians who have written history. Running records are ongoing series of statistical or other sorts of data, such as census data, ship's registries, property deeds, etc. Finally recollections include sources such as autobiographies, memoirs or diaries.

There are four stages, as discussed by Schutt, to systematic qualitative comparative historical studies: (1) develop the premise of the investigation, identifying events, concepts, etc., that may explain the phenomena; (2) choose the case(s) (location- nation, region) to examine; (3) use what Theda Skocpol has termed as "interpretive historical sociology" and examine the similarities and the differences; and (4) based on the information gathered, propose a causal explanation for the phenomena.

The key issues in methods for historical comparative research stem from the incomplete nature of historical data, the complexity and scale of the social systems, and the nature of the questions asked. Historical data is a difficult set of data to work with due to multiple factors. This data set can be very biased, such as diaries, memoirs, letters, which are all influenced not only by the person writing them, that person's world view but can also, logically, be linked to that individual's socioeconomic status. In this way the data can be corrupt/skewed. Historical data regardless or whether it may or may not be biased (diaries vs. official documents) is also vulnerable to time. Time can destroy fragile paper, fade ink until it is illegible, wars, environmental disasters can all destroy data and special interest groups can destroy mass amounts of data to serve a specific purpose at the time they lived, etc. Hence, data is naturally incomplete and can lead social scientists to many barriers in their research. Often historical comparative research is a broad and wide reaching topic such as how democracy evolved in three specific regions. Tracking how democracy developed is a daunting task for one country or region let alone three. Here the scale of the social system, which is attempting to be studied, is overwhelming but also the complexity is extreme. Within each case there are multiple different social systems that can affect the development of a society and its political system. The factors must be separated and analyzed so that causality can be attained. It is causality that brings us to yet another key issue in methods for historical comparative research, the nature of the questions which are asked is attempting to propose causal relationships between a set of variables. Determining causality alone is a difficult task; coupled with the incomplete nature of historical data and the complexity and scale of the social systems being used to examine causality the task becomes even more challenging.

Theda Skocpol and Margaret Somers argued that there were three types of comparative history research:

A lot of comparative historical research uses inductive iteration (as opposed to purely deductive methods) whereby scholars assess the data first and reformulate internally valid explanations to account for the data.