Let's start with "online project", with a "29.3%" self-selecting confirmation bias, including people who apparently attempted to jam the ballot box by voting more than once, and also including undervotes. Sound familiar?
"The 2024 Presidential Greatness Project Expert Survey was conducted online via Qualtrics from November 15 to December 31, 2023. Respondents included current and recent members of the Presidents & Executive Politics Section of the American Political Science Association, which is the foremost organization of social science experts in presidential politics, as well as scholars who had recently published peer-reviewed academic research in key related scholarly journals or academic presses. 525 respondents were invited to participate, and 154 usable responses were received, yielding a 29.3% response rate."¹¹ Responses where respondent did not confirm they officially agreed to participate were dropped and are not included in the submission figure. Similarly, in the event a respondent completed the survey twice, the second submission was dropped, as was any submission where the respondent did complete the questionnaire through the initial presidential greatness rating sections. Incomplete submissions were otherwise included, with responses being analyzed for those questions where answers were provided.
Missing Data"There is variation in whether respondents answer each question or not. In Table 7 we include details about the number of missing observations missing a response, unique values, and the minimum and maximum value. Most respondents completed the full rating across presidents but responses for some presidents (Harrison, Tyler, Polk, Arthur) had more missing observations than later presidents likely due to more information about those presidents."