“Change in Source of Leader Support” Data

Co-PIs: Brett Ashley Leeds (Rice University) and Michaela Mattes (UC-Berkeley)

This data collection effort was sponsored by the National Science Foundation (SES: 0921781; Collaborative Research “Interests, Institutions, and Foreign Policy Change”)


There are few things more important than the ability to explain and predict dramatic changes in foreign policy.  Unfortunately, existing international relations theories are much better at explaining consistency in behavior rather than predicting change.  The primary reason is that scholars have tended to focus on the international system as the main source of foreign policy and have neglected the effect of changes in domestic political interests and preferences. Leaders have an incentive to pursue policies (including foreign policies) that are in the best interest of the particular societal coalition that keeps them in power.  Leadership transitions that do not change the representation of domestic interests should have relatively little effect on foreign policy.  But when leaders who depend on the support of different groups than their predecessors assume power, the opportunity for domestically motivated policy change arises. Whether changes in the interests and preferences a leader represents actually lead to shifts in foreign policy, however, depends on the domestic and international institutional context.  Some domestic political structures make it easy for leaders to redirect foreign policy, while others impose significant constraints on leaders by requiring them to find approval from other domestic actors.  In addition, some areas of foreign policy are more conducive to change than others; it should be harder to change formal international commitments than foreign policy that is not tied to a formal legal obligation. This project explains and evaluates the interaction between changes in the core societal groups with influence over policy, domestic political institutions, and international political institutions in affecting foreign policy consistency and change.  To know whether institutions constrain behavior, one must isolate instances in which there is reason to suspect that preferences have changed, and see if policy changes accordingly.

Data Collection:

We differentiate leadership transitions in which a new leader comes to power who depends on different societal groups for support than her predecessor from leadership transitions where both the current leader and her predecessor rely on essentially the same groups for support. The data cover all countries with a population of more than 500,000 and span the period 1919-2018.


The CHISOLS data v. 5.0 and all supporting materials can be downloaded at