by Patrick Murray
As discussed in a prior post, Monmouth University’s Polling Institute assisted with a survey that formed a central part of Authoritarian Nightmare, a new book by John Dean and Bob Altemeyer. In another post, I examined measurements for Right Wing Authoritarianism (RWA) and Social Dominance Orientation (SDO) and how they relate to support for President Donald Trump. I take that analysis a step further to examine variations within the electorate on both sides of the partisan divide and a key question about constitutional norms.
While the analysis in the book, as well as in my last post, show a clear correlation between RWA and SDO inclinations, that is not to say that all Trump supporters are authoritarians. Or indeed that all Trump opponents are not authoritarians. While 32% of strong Trump approvers score in the highest quartile on both the RWA and the SDO scales, 23% actually score below the top quartile on both dimensions (Table 1). Among voters who somewhat approve of the president’s job performance, 51% fall below the top quartile on either scale. Conversely, while the vast majority of Trump opponents have a moderate or low score on these scales, 21% of those who somewhat disapprove of Trump and 5% of those who strongly disapprove of Trump exhibit high RWA and/or SDO tendencies.
| Table 1. | RWA/SDO Inclination by Trump Job Rating | |||
| Strongly disapprove (n=463) | Somewhat disapprove (n=43) | Somewhat approve (n=119) | Strongly approve (n=346) | |
| Double High RWA & SDO | 1% | 0% | 9% | 32% |
| High RWA only | 2% | 14% | 24% | 20% |
| High SDO only | 2% | 7% | 15% | 26% |
| Moderate RWA/SDO | 21% | 49% | 41% | 19% |
| Low on both | 74% | 30% | 10% | 4% |
Some Trump supporters aren’t particularly enamored of his authoritarian tendencies and at the same time some Democrats prefer authoritarian leadership, just not Trump’s. Since opinion of Trump’s job performance correlates highly with partisan identity, we can ask which types of Republicans are more likely to be on the authoritarian train. The data do not show a lot of variation demographically (Table 2). Republicans without a college degree tend to score highest on the RWA scale, while younger Republicans are somewhat less authoritarian than older partisans. But that’s about it.
| Table 2. | Mean RWA/SDO Scores Among Republicans | |
| RWA | SDO | |
| All Republicans + leaners (n=465) | 112 | 86 |
| GOP identifiers only (n=300) | 115 | 86 |
| GOP leaners only (n=165) | 106 | 86 |
| Men (n=292) | 113 | 88 |
| Women (n=169) | 111 | 83 |
| Age 18-44 (n=53) | 95 | 82 |
| Age 45-64 (n=188) | 110 | 90 |
| Age 65+ (n=215) | 118 | 84 |
| High school or less (n=71) | 120 | 89 |
| Some college (n=137) | 120 | 86 |
| 4-year college degree (n=160) | 108 | 86 |
| Post-graduate (n=96) | 102 | 84 |
| White (n=415) | 112 | 87 |
| Latino, Black, Asian, other (n=39) | 109 | 83 |
There’s a similar lack of differentiation among Democrats (Table 3). All demographic groups score significantly lower than their Republican counterparts, but there are only minor variations within the party itself. Democrats who never attended a college class tend to be most authoritarian and white Democrats tend to score lower on the RWA scale than Democrats of other racial or ethnic backgrounds. Otherwise, the scores are fairly similar across the board.
| Table 3. | Mean RWA/SDO Scores Among Democrats | |
| RWA | SDO | |
| All Democrats + leaners (n=408) | 53 | 45 |
| DEM identifiers only (n=284) | 53 | 44 |
| DEM leaners only (n=124) | 52 | 46 |
| Men (n=192) | 53 | 49 |
| Women (n=216) | 52 | 41 |
| Age 18-44 (n=93) | 49 | 45 |
| Age 45-64 (n=147) | 48 | 43 |
| Age 65+ (n=168) | 59 | 47 |
| High school or less (n=36) | 75 | 47 |
| Some college (n=69) | 55 | 45 |
| 4-year college degree (n=137) | 49 | 45 |
| Post-graduate (n=166) | 49 | 44 |
| White (n=323) | 49 | 45 |
| Latino, Black, Asian, other (n=81) | 66 | 45 |
I looked at potential variations among Democrats in another way by using a question on presidential preference as a proxy for different orientations within the party. Remember, the Democratic nomination was still wide-open when this survey was conducted in the fall of 2019. Mean RWA scores ranged from a relative high of x̅=59 among Joe Biden supporters (n=101), to x̅=48 among Bernie Sanders supporters (n=41), x̅=48 among Pete Buttigieg supporters (n=88), and x̅=44 among Elizabeth Warren supporters (n=75). The mean RWA score among Democrats with no candidate preference (n=63) was x̅=59. [Interesting side note: Tulsi Gabbard was named by 32 respondents as a preferred Democratic nominee, but not by anyone who actually identified as a Democrat (the nominee preference question was asked of all survey respondents). Her supporters averaged x̅=106 on the RWA scale, which is much closer to the Republican average than the Democratic one.]
Both parties include their share of “authoritarians,” but there is a clear lack of critical mass on the left. Among Democrats, 11% of Biden supporters and 10% of Sanders supporters had a high score on the abbreviated 5-item RWA scale in the poll, while only 3% of Buttigieg and Warren supporters did. There simply aren’t enough authoritarian Democrats in the electorate to rally around a single leader and take control of the party. Of course, this may also be a “chicken and egg” problem. Voters who prefer authoritarian style leadership may be less comfortable with the Democratic Party as much for its lack of centralized leadership as for any specific set of policy issues. But considering the strong link between authoritarianism and religious fundamentalism in general, social issues must certainly play a key role in these wide partisan gaps on the RWA scale.
I attempted to strip the original RWA scale of its obviously religious items (discussed in my prior post). I found that the scales held up without the items that referenced values on gender and sexuality. However, this finding needs to be taken with an important caveat. The full battery of questions in the survey interspersed the value-laden items among the leadership items, so it is possible that respondents answered the full question set in ways that attained a level of internal attitude consistency (what we call context effects in the polling biz). This means that we could possibly get a different set of answers to the 5 questions in the shorter RWA measurement if they had been the only part of the scale included in the survey. We would need to test this in isolation to be certain.
As I mentioned earlier, my research interest lies in exploring the psychosocial underpinnings of political behavior. But I also have a normative interest in this particular study. To what extent does authoritarianism pose a danger to the U.S. Constitution?
The authors added a question to the survey that attempted to directly measure this question. However, the results may seem counterintuitive, at first glance. The question wording was: “The U.S. Constitution gets in the way of things too much nowadays and should just be ignored when it interferes with taking action on some issue.” Overall, only 6% of respondents agreed with this statement or even took a neutral stance on it. In fact, 69% picked the most extreme “disagree” option.
Now, the first thing that comes to my mind in examining these results is social desirability bias. The idea that we should respect the U.S. Constitution is so baked in to the American psyche that you can hardly expect anyone to admit they are willing to disregard it when asked so blatantly: “I’m not undermining the Constitution. You’re undermining the Constitution!”
In fact, the responses to this question tended not to correlate with any of the RWA items in the survey. In the few cases where there was a significant – albeit weak – relationship, the correlation tended to be negative. In other words, willingness to ignore the Constitution to attain a policy end tended to be held – to the extent it was held at all – by people who had lower authoritarian tendencies.
I have not spoken with Dean and Altemeyer about this, but I expect they may have been surprised by the results. Fully 83% of respondents who strongly approve of Trump said they very strongly disagree with this statement – the most negative option in the survey’s 9-point response set. This level of disagreement was less widely shared among voters who were less stalwart Trump supporters, including those who somewhat approve (68%), somewhat disapprove (51%), and strongly disapprove (62%) of the president’s job performance.
If you look at this question by partisanship, just 4% of Republicans either agreed or took a neutral stance on the question of the Constitution getting in the way of things, compared with 8% of Democrats and 10% of independents. This is a small, but still statistically significant difference. There is even further differentiation among Democrats. Sanders supporters (20%) are more likely to agree with this sentiment than backers of Biden (10%), Warren (5%), or Buttigieg (3%).
This may be a bit of a shocker because the charge against Trump has been that he is willing to trample on the Constitution to further his aims and his followers have been willing to go along with it. Does this mean strong Sanders supporters are more likely to ignore the Constitution to achieve their ends than strong Trump supporters are? I don’t think so. In fact, I hypothesize that these Sanders backers are just more likely to admit it – and indeed even recognize that they are doing it.
If you look at the Constitution as an operational framework for government rather than an ideological document, its purpose can basically be boiled down to concerns rooted in its historical origins. One key purpose was to create a functioning central government that would engender enough public trust to prevent anarchy and chaos (think Shay’s Rebellion). The other purpose was to prevent authoritarianism; that is, allowing any individual leader to put themselves above the law (think King George III). Basically, the Constitution works, not when everyone gets what they want from a policy standpoint, but when the public trusts that those who hold the reins of power are observing its rules of engagement.
Now, there are people on the left who want anarchy and those on the right who want an omnipotent ruler. For both, the U.S. Constitution’s purposeful ambiguity is an impediment to their ends. It just seems those on the left are more likely to actually recognize this and admit it. What neither side seems to realize, though, is this: in a nation as vast and diverse as ours, those tedious checks and balances embedded in the Constitution are in fact what keep this country on an even keel during uncertain times.
The bigger danger to our Republic does not, in fact, come from those who admit to seeking a wholesale change in our form of government. You can see that coming. No, the more perilous hazard comes from those who are willing to erode Constitutional norms from within by manipulating the natural fears and psychological dispositions of a segment of the American public. And the biggest danger of all comes from those who do this while duplicitously giving lip service to the core principles of our Republic’s founding document.