I Don't Get It

PublicPolicyPolling recently asked voters nationwide whether they had a favorable or unfavorable impression of each of the 43 men who have been President of the United States (with the option, of course, of "don't know"). Now, it can be a little tricky to compare these ratings since name recognition, which I define as the percentage giving an answer either way, varies from a low of 16% for Franklin Pierce to a high of 95% for Barack Obama. Is a 12%/9% approval rating better or worse than a 42%/39% one? But looking at the numbers can show us some interesting, and at times perplexing, things:

• Washington is the most popular President, at 89%/3%. Then you've got Lincoln at 85%/7%, Jefferson at 74%/6% (incidentally, why does Jefferson have only 80% name recognition while Washington and Lincoln, among others, have >90%?), Teddy Roosevelt at 66%/9%, John Adams (?!) at 59%/5%, and both Truman and Eisenhower at 62%/11%. Kennedy (70/21), FDR (62/22), and Reagan (62/30) all have high favorability but also higher unfavorability than the others.
• Richard Milhous Nixon's 59% unfavorable rating, and his net -32 performance, are the worst by far. In fact, no one else has a double-digits net-negative rating. If you need an historical political name to provoke a negative reaction, there's just no doubt about it: Nixon's the one!
• The performance of some of the old-time, low-name-recognition Presidents surprises me. James Madison only gets 44% recognition (though that breaks 37-7 in his favor); what's up with that? Andrew Johnson is surprisingly popular, at 26%/18%. What in the world, I wonder, makes William Henry Harrison have a 14%/8% split? What in the world is there to like about him? He was in office for a month! And why do people like Zachary Taylor so damn much (15%/9%)?
• Calvin Coolidge's 18%/22% numbers surprised me, given that he had the biggest margin of victory in the popular-vote era, and is remembered (I thought) primarily as witty and having presided over a good economy. Maybe people really get the whole "it wasn't all Hoover's fault, Coolidge should get more blame than he does" thing. 
• Most of the net-unpopular names among the old-timers are obvious, like Millard Fillmore (7%/12%), James Buchanan (11%/13%), and Martin Van Buren (13%/15%). But why is Chester A. Arthur at 10%/13%? Everyone loved him! Mark Twain loved him! Why in the world do people dislike Chester Arthur? What could you possibly know about him that would make you dislike him?
• John Tyler and James Polk both garner even splits in their ratings, at 10% and 9% respectively. Joining them are Franklin Pierce at 8% and Jimmy Carter at 44%. Polk typically gets great ratings from historians, while Tyler and Pierce are two of the worst ever.

Here's a couple of visualizations of this data that I like. First, we have an array of the history of US Presidents, year-by-year, colored by their net favorability, i.e. Favorable - Unfavorable. Bright green is good; muddier and redder colors are bad. Nixon's represented by dark red, just to emphasize how awfully he ranks by this metric.

And now here's an analogous diagram, but using a formula of mine instead of just raw subtraction. The way this works is that you take the percentage of respondents who gave a favorable rating and the percentage of those respondents who gave a rating one way or the other who did so, and average them. This is my rough attempt to integrate the desire for a good net favorability ratio with the desire for high name recognition. Anyway, here's what you get by that metric:

By both methods we see that there have been two general periods when Americans now think we were well-governed: from the Founding through the Jackson Administration in the mid-1800s and from the Depression through the early 1960s. Since the depravity of Nixon, by far the most recent President in the "really bad" category, we've thought our governance kind of mediocre, especially judging by the more sophisticated metric. We basically think we were badly governed from the end of the Jackson Administration through the beginning of the Civil War and the advent of Lincoln (which is true!) and from the end of Reconstruction and the Hayes Administration through the Depression, with only a brief interruption for Teddy Roosevelt.

One other thing I can do is compare the popular rankings of the Presidents, using the formula from that last chart, with the way a recent survey of Presidential historians ranked them. Then I can see which Presidents the public most "overrates" or "underrates." Overratedness will be measured as a President's ranking among historians minus their ranking among the general public, so a positive number means the public likes them better than historians and a negative number the opposite. I find that Andrew Johnson is the most popularly overrated President, at +20, followed by George W. Bush at +19, Ulysses Grant at +14, John Adams at +12, Gerald Ford at +11, and John Q. Adams, Jimmy Carter, and William Henry Harrison at +10. The most popularly underrated President is James Polk at -20, followed by Chester Arthur at -16, Martin Van Buren at -13, Grover Cleveland at -12, and Woodrow Wilson at -11. James Madison and Franklin Roosevelt were at -9, with Nixon and LBJ at -8. Ronald Reagan is overrated, at +7. Only one President's ranking was exactly the same among both groups, who put Andrew Jackson 14th. Both Andrew Johnson as the most overrated and James Polk as the most underrated confirmed my subjective impressions from looking over the favorability numbers.

Anyway, some interesting stuff, I'd say.