Reading the Political Tea Leaves

Unlike every 2024 election post-mortem you’ve read (or listened to—this was a podcast election after all), this one was not written by AI nor was it penned while staring off into space while the scrambled eggs burned. We took some time to consult our data and allow for a healthy internal debate to arrive at our perspective on the American electorate as Trump parades down Pennsylvania Avenue rolls over to the CapitalOne Arena. 

As a company that prides itself on capturing the desires, drivers, aspirations, and attitudes of diverse populations, it’s imperative that we skeptically evaluate what we know about the populations we claim to understand and how we understand them. In sharing our perspective on the 2024 election, we hope to add some substantive clarity to the discussion by drawing upon survey experiments that we fielded prior to the election. 

Full disclosure: we are not election pollsters and have no aspirations to be (but bless those of you who do!). We exist to equip our client partners (whether they are a company, philanthropy, or non-profit) with multiple angles of intelligence so they can makemake high-stakes strategy decisions to better reach and speak to their audiences. Support for a presidential candidate is one valid form of intelligence (aka horse-race polling), but we think it’s fundamental to understand the constellation of context, values, perceptions, and behaviors that collectively contribute to voters’ choices.   

From our perspective, what we saw in pre-election research was that voters did not feel comfortable revealing their true opinions on various highly sensitive topics that dominated the discourse of both campaigns. As a result, the campaign consulting industrial complex recommended strategies based on an incomplete picture of what truly mattered to voters. 

From data that we collected with Populace over the summer of 2024 using a novel methodology called Private Opinion Research, we found that voters were self-censoring their internal stance on many of the Republican and Democratic parties’ signature positions.

It’s telling that some of the distinguishing tenets of both campaigns were, even through public polling, not popular. On a host of issues, however, we found a difference between what supporters of both parties would disclose publicly compared to where they stood on the same issue when guaranteed privacy in a survey experiment. The causes of this self-censoring still warrant further debate but could be due to a variety of reasons such as discomfort for being out of step with “like-minded” peers or pressure to say the right thing—even if a voter disagrees in their most private of thoughts.   

Knowing that Republican and Democratic voters didn’t feel comfortable voicing their deeply held opinions on some of their party’s most distinguishing positions could be a leading indicator that the parties’ positions are out of step with its base and with voters it thought to be persuadable.

For example, private opinion research reveals that Democrats have lower levels of support for policies that attempt to assist and protect minority populations, such as affirmative action and gender diversity quotas. Although support is relatively higher for protecting transgender Americans, Democrats are self-censoring the extent to which they support the government protecting this minority population. Perhaps most telling is the extent to which Democrats are publicly much more inclined to declare that the government can be trusted (keep in mind these data were collected during the Biden administration) but privately have almost no trust in the government.  

Republicans are also subject to self-censorship on policy positions that were prominent flashpoints in the campaign. To draw a common parallel with Democrats’ self-censorship on support for affirmative action and diversity quotas, Republican voters were also self-censoring the extent to which they want DEI efforts to be dismantled. Seeing that prominent companies were quick to disband their DEI programs after the election suggests that they could be responding to this latent loathing of DEI programs. On immigration, Republicans are not nearly as unanimous in their hope to see the U.S.-Mexico border closed as public opinion polls would suggest, which may have turned off some potential voters given how prominent that messaging was during the campaign.  

These results signal to us that the public opinion polling methodologies that political campaigns rely on are not a sufficient tool to understand the latent complexities that voters have to muscle through as they decide if and for whom to vote. 

That’s a wrap, folks

We'd love to hear from you. Do you have any comments, suggestions, or other pieces of feedback? Please don't hesitate to respond to this email, we don't bite!

Do you have a burning strategic question on which your organization needs objective clarity? Get in touch and we can see how we can help.

About Gradient

In a continuously changing world, intuition isn't enough. We are decision science partners who equip our clients with evidence-based clarity to answer their most challenging strategy questions and achieve their growth goals. We uncover critical objective realities for our partners with bespoke, consultative research programs that push the boundaries of custom statistical methodologies. We’ve partnered with Fortune 100 brands like Nike, Bacardi, and Brooks, startups, consulting firms, and political campaigns. Want to learn more? Visit our website!