Such efforts are sometimes interesting, but in the end, they are meaningless.
These processes are hopelessly subjective, and reveal, at most, the personal preferences and partialities of the researchers involved. Because these types of surveys have been going on for years, one can trace their contradictory results which expose their sheer non-confirmability and un-verifiability.
Writing in 2012, Robert Merry traced the flip-flops and reversals of such surveys:
Consider Dwight Eisenhower, initially pegged by historians as a mediocre two-termer. In 1962, a year after Ike relinquished the presidency, a poll by Harvard’s Arthur Schlesinger Sr. ranked him 22nd — between Chester A. Arthur, largely a presidential nonentity, and Andrew Johnson, impeached by the House and nearly convicted by the Senate. Republicans were outraged; Democrats laughed. By the time a 1981 poll was taken, however, Eisenhower had moved up to 12th. The following year he was ninth. In three subsequent polls he came in 11th, 10th and eighth.
The academics did a similar turnaround, and did an about-face on another famous president:
Academics initially slammed Reagan, as they had Eisenhower. One survey of 750 historians taken between 1988 and 1990 ranked him as “below average.” A 1996 poll ranked him at 25th, between George H.W. Bush, the one-termer who succeeded him, and that selfsame Chester Arthur. Reagan's standing is now on the rise.
If the search for the “best ever” president, or even the “top ten” presidents, is an empty pursuit, can scholars give more meaningful results? Perhaps: while it is meaningless to say that Calvin Coolidge is a “good” or “bad” president, it is meaningful to say that he lowered taxes, lowered the national debt, and reduced the federal government’s spending. Such statements are verifiable and quantifiable.
Historians can give us meaningful data when they research specific and measurable details about a president, instead of merely trying to assign him a relative rank as “better than” or “worse than” some other president.
It is observable, and therefore material, the President Polk’s management of the Mexican-American war impacted presidential elections after the war’s end in 1848.
Such observations are not only more reliable and objective, but also protect scholars from ending up with the proverbial “egg on the face” of declaring some president to be “good” or “bad” and then find themselves facing stiff opposition to such judgments. One example of academics hastily praising a president, only to find themselves slowly retracting such glowing evaluations, is the case of Woodrow Wilson.
Wilson’s high marks from historians belie the fact that voters in 1920 delivered to his party one of the starkest repudiations ever visited upon an incumbent party. Similarly, historians consistently mark Harding as a failure, though he presided over remarkably good times and was very popular.
Exactly as scholars revised their estimates of Eisenhower and Reagan upward, so now they are reconsidering Harding in a more favorable light. Wilson’s reputation, meanwhile, has descended.
In sum, it is more important to gather data about a president than to evaluate him.
Writing about a president should emphasize, not general impressions, but rather observable, measurable, verifiable, and quantifiable data. That’s how serious historians work. Reports about presidents should be full of dates, places, specific actions, and the names of other individuals with whom that president interacted.
Such a method would lead to the “best ever” texts about presidents!
