How much should politicians (and pundits) follow the polls?

This is the wrong question.

Dear readers,

I delivered some remarks at a conference on public opinion last week, focusing on how data can shape political journalism in a way that informs the public and brings political science to the masses. Preparing those remarks sparked this post on how much politicians should be “leading” vs “following” public opinion, which is a topic of renewed debate recently.

Like other Tuesday posts, this one is paywalled for paying members of Strength In Numbers. This is necessary to pay the bills, but I’ve left both the introduction to the piece and the key questions I’m answering above the paywall to try to contribute something to the public debate regardless. If a paid membership is in your budget, I think our newsletter is very high ROI, especially compared to the competition, and it supports the public mission of this business.

Many thanks for your support, shares, and comments! I’m having a lot of fun as an independent journalist, and I couldn’t do this without you.

Elliott

---

"To speak with precision about public opinion is a task not unlike coming to grips with the Holy Ghost" – V. O. Key Jr, 1961

A debate is raging among Democratic strategists and left-leaning commentators about how the Democratic Party should position itself ideologically to maximize its probability of winning the next election.

This debate centers on conflicting diagnoses of Democrats' electoral problems. One side argues that a failure to win the presidency, and especially the Senate, can be traced to the party moving too far from the median voter and alienating moderates. This side argues Democrats should immediately move back to the middle on key issues, such as immigration, in order to win back the pivotal center. They suggest moderating especially on issues that get a lot of attention from the press and voters, and being wary of raising aggregate attention on issues where Republicans have an advantage.

This group is sometimes referred to as “popularists.”

The other side argues that Democrats should focus more attention on shaping public opinion, rather than merely following it. They say what the party lacks is leadership, especially on issues of corruption, democratic institutions, and civil rights.

An illustrative case is the deportation of Kilmar Abrego Garcia, about which the popularists said Democrats were focusing too much on immigration, while the latter group argued public opinion could easily be changed with sympathetic figures and the right message. The “anti-popularists” charge the pro-moderation camp with affecting excessively "poll-tested" politics and lacking the courage and conviction to create a mobilized base of mass support on ideologically cross-cutting issues.

At the heart of this debate are two empirical questions, as yet unaddressed in the popular discourse. One question concerns how much opinion can change in the medium term. The other is about our ability to construct counterfactual elections against which to judge the party's current positions and future performance.

As an avid reader of political science and the author of a book about the link between public opinion and democracy, I have some ideas about how to answer these questions. Let’s look at the empirical evidence and see if we can find a way to resolve this debate — or if not, at least offer some much-needed data-driven advice.

---

Public opinion matters more closer to Election Day