by Patrick Murray
The Monmouth University Polling Institute recently provided research support for a new book by John Dean and Bob Altemeyer entitled Authoritarian Nightmare. Our objective was to further public understanding of what drives voter attitudes and behavior in the 21st century. This is how our association with the authors came about. Thoughts on the data in the survey can be found in separate posts (Part 1 and Part 2).
I have a long-standing interest in the role of identity in politics – how people see themselves and how that translates into political attitudes and behaviors, be it through the lens of race, class, geography, etc. The rational choice models typical of political science tend to miss the more visceral need to be part of a group and how that group identity can subsume other political calculations. This is the main reason why I studied political psychology in graduate school and eventually found my way into the polling business.
These factors became especially evident with the ascendance of Donald Trump’s 2016 candidacy. It is worth remembering that Trump had only a 20% favorable to 55% unfavorable rating among Republican voters before he got into the race. Within weeks of his June 2015 announcement, that opinion had flipped to 52% favorable and 35% unfavorable. He hit 70% favorability by Election Day and now routinely tops 80% among his fellow Republicans. I remarked at the time that a universally known candidate suddenly upending what voters think of him on the basis of a campaign speech was unprecedented.
Trump, however, was no typical politician and his appeal could not be readily explained by the usual political paradigms – no matter how much academics and pundits tried to do so. It wasn’t that a critical mass of Republican voters were looking for certain policy boxes to be ticked. These voters were not saying to themselves, “The lack of a border wall is keeping me up at night. Which candidate can I trust to build one?” They were looking for someone who could articulate why they felt unsettled in a changing world; someone who could identify the culprits responsible and vilify them. They were looking for someone who would ostensibly allay their fears – ironically by stoking those fears.
Clearly, something more than a rational choice model of political behavior was at work. But standard public opinion polling has a difficult time tapping into these psychosocial dimensions. These factors are always present but have become much more prominent – perhaps even overwhelming – in the Trump era. There were a few attempts to measure these extra-political factors during the 2016 campaign. None was particularly robust, though, largely because the lengthy question sets needed to tap into these dimensions cannot be administered reliably in a telephone poll. Surveys with psychological batteries tend to be long, wordy, and intentionally provocative, while standard polls are short, succinctly worded, and intentionally bland.
I began looking into ways to obtain better measurements of these phenomena as Trump supporters’ loyalty solidified once he took office. Enter John Dean.
Dean has spoken on the Monmouth University campus twice in the past five years and I had the opportunity to join him for lunch on those occasions. During a conversation in October 2018, I mentioned my concern that standard political polling and the media coverage of voter attitudes was not up to the task of fully explaining the current political climate, and specifically Trump’s ability to maintain his core support. John contacted me shortly after that meeting to discuss a book idea he was working on. In 2006, he published Conservatives Without Conscience, a play on the title of Barry Goldwater’s seminal work, The Conscience of a Conservative. Dean’s book profiles the evolution of the Republican Party as he saw it, drawing on the authoritarianism research of Altemeyer, a retired professor of psychology from the University of Manitoba. While the 2006 volume profiled Republican leaders, Dean and Altemeyer were collaborating on a new book examining Trump’s rank-and-file supporters in the context of the typologies Bob started developing four decades ago.
One thing they were missing, Dean explained, was a broad-based survey of American voters that could directly test their hypotheses. Upon hearing my own interest in developing a fuller understanding of voter attitudes and behavior in the Trump era, Dean asked if Monmouth might be interested in conducting a survey using Altemeyer’s psychological batteries. I saw this as an opportunity to further our body of knowledge by combining academic research with public polling methodology.
We came to an agreement where the authors underwrote a representative online survey of American voters (although we slightly oversampled Republicans to ensure a sufficient group size in the study). Monmouth maintained full control over the selection and management of the sample in line with our standard protocols. And while we agreed to work cooperatively with Altemeyer on the questionnaire content, Monmouth effectively had veto power over any questions, aside from the validated psychological batteries, that did not meet our standards for objective measurement. Dean made it clear on multiple occasions that he was fully prepared for the possibility they might need to rethink their premise if the data led to different conclusions. The numbers would fall where they may. In the end, the findings did support their hypothesis.
Monmouth delivered the final data set to the authors late last year, but played no role in the findings and conclusions of their book. However, part of the agreement was that Monmouth would have the ability to publish its own analysis of the data once the book was released. As such, we provide some initial thoughts on the results Part 1 and Part 2.