Style

   Donald Trump got about 46% of the votes in 2016, a little less than what Mitt Romney got in 2012.  But he did considerably better among less educated voters than Romney had (and worse among more educated voters).  What caused this shift?  I think that "style" may have been an important factor.  Most people seem to have pretty definite feelings about what qualities make for a good leader,  and a lot of campaign coverage involves the personal qualities of the candidates.  There haven't been many questions about leadership qualities, and most of those are about candidates (e. g., would X be tough enough....).  However, there was one survey from March 2008 that asked people to choose between different styles of leadership.  I have written about it before, but I will go into more depth here.  The relevant questions:

"would you prefer a president who gets involved in many of the details of most issues, or a president who sets broad policies and then delegates to others the implementation of these policies?"

"would you prefer a president who makes decisions and then sticks with them no matter what, or a president who reconsiders decisions after making them when circumstances change?"

"would you prefer a president who spends a lot of time thinking things through and deliberating before making decisions, or a president who makes decisions more quickly based on his or her gut instincts?"

"Suppose a loyal and long-time employee of the president was not doing a good job. If you had to choose, would you prefer a president who keeps that employee out of loyalty, or a president who would fire the employee?"

"What kind of advisors and employees do you think a president should hire--should he or she hire people with different views who may disagree on important issues, or hire people who mostly share his or her views?"

"would you prefer a president who sets peoples' expectations higher even if he or she may not be able to meet them, or a president who sets peoples' expectations lower but probably will be able to meet them?"

A summary of the relationship to education:  the first column is the direction and the second is the size of the estimated effect (in a logistic regression):

Delegates                .46

Reconsiders            .67

Deliberates            (.18)

Fires                      (.10)

Different                 .23

High                       .27

The two estimates in parentheses aren't statistically significant.  There's a good deal of uncertainty about them, since overwhelming majorities at all educational levels favored deliberation and firing the loyal employee.  The effects for reconsidering decisions and delegating are clearly the strongest.  In terms of percentages, 27% of people with a high school diploma or less and only 6% of college graduates wanted a president who would stick with decisions no matter what.

The answers to the different questions had only small associations with each other, but there was a strong association between reconsidering decisions and another question:  whether people were more offended "negative comments about women in general, or when people make negative comments about African Americans in general, or don't remarks like that offend you?"  Over 40% volunteered that they found them equally offensive.  There was little or no difference between people who were more offended by remarks about one group than the other, so I combined them:

                                                 Sticks       Reconsiders

Not offended                              25%            75%

One more offensive                   16%             84%

Both offensive                           11%             89%

I didn't anticipate this association, but in retrospect the questions seem to fit together into a "never apologize, never explain" style.  

I don't know of any data on how people perceived Trump's leadership style, but my impression is that the image he cultivated was more in line with what less educated people preferred--get involved, make a decision and stick to it.  

[Data from the Roper Center for Public Opinion Research]