Research Spotlight: Why not both? The Liberal-Realist Model of Public Opinion

What drives Americans’ preferences for potential alliance partners? In an ideal world, an alliance partner would bring not only strategic benefits, such as increasing the United States’ national security, but would also share liberal values, such as a belief in the virtues of democracy. In reality, the demands of the geopolitical environment may push the U.S. to consider partnerships with countries that do not live up to either of these ideals. For example, while a potential alliance partner may have a good human rights record, it may have a weak military. How then do members of the U.S. public evaluate potential alliance partners when they fall short of these ideals?

Some political scientists, such as John Mearsheimer, argue that Americans base their foreign policy preferences on how well it aligns with their liberal values. In The Tragedy of Great Power Politics, Mearsheimer alleges that among the American public, strategic considerations, such as concerns about relative power, take a backseat to installing policies that emulate U.S. values abroad. Other political scientists, such as Daniel Drezner (2008), poke holes in this assumption about the U.S. public. In a 2008 review article called, “The realist tradition in American public opinion,” Drezner concludes that Americans’ commitment to liberal values in the abstract tends to fall through when setting their priorities for U.S. foreign policy. The literature is divided on how the U.S. public assesses its foreign policy with some arguing that liberal values matter most, and other arguing that strategy takes priority.

In my working paper, “Why not both? The Liberal-Realist Model of Public Opinion,” I challenge these characterizations of Americans’ foreign policy preferences. I argue that Americans care about both “liberal” values and strategic interests, and that when deciding whether to support a potential alliance, these factors may be in tension with one another.

I explore this question with a conjoint experiment using a sample of U.S. adults recruited from MTurk in 2021 who in total evaluated about 6,500 alliance profiles. In the experiment, I showed respondents two profiles of potential U.S. alliance partners and asked them to indicate which country would be their preference for an U.S. alliance partner. Each potential alliance partner had randomized values of four attributes: military strength, location, regime type, and human rights. Following respondents’ alliance choice, I asked a series of questions to understand the underlying reasons behind their choice.

Using regression analysis, I calculated the effect (average marginal component effect) of each attribute on a respondent’s preferred choice for an alliance partner. I find that a country’s respect for human rights had the largest effect on alliance choice, followed by regime type, military strength and location. There were several statistically significant reasons behind respondents’ alliance preferences. If a respondents thought that a potential alliance partner would be more likely than the other to honor its agreements and increase U.S. political influence and security, respondents were more likely to select that country for their preferred alliance partner. However, if a respondent thought that a potential alliance member would be more likely to pull the U.S. into unwanted wars, or if an alliance was deemed to be morally wrong, respondents were less likely to choose that country for an alliance partner.

In conclusion, my results show that when forming alliances, Americans care about both strategic concerns and liberal values being reflected in their alliance partnerships. In an ever-changing geopolitical climate, choosing an alliance partner that satisfies preferences for all four attributes may be a challenging feat; however, my results provide insight about the public’s priorities and the reasoning behind them.

Taylor Kinsley Chewning is a PhD student in the Department of Political Science. You can learn more about Taylor here.

Note: This summary is based on March 2022 version of my working paper which is subject to change. Please reach out for information about the most current version of the paper at or visit my website at

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.