The same poll found 52 percent of 270 respondents disapprove of Summers’ leadership of the University, compared to 40 percent who approve of his leadership. Eight percent said they did not know.
Reporters from The Harvard Crimson on Friday e-mailed nearly all of the 683 Faculty members listed in the school’s online course catalog with five questions regarding Summers, who has faced attacks in recent weeks from professors at FAS, the largest of Harvard’s nine schools. Respondents to the poll were guaranteed anonymity.
Four experts in public opinion polling said yesterday they thought The Crimson’s methodology was sound, though they cautioned that the sample collected was unlikely to be random and could not be extrapolated to represent the entire Faculty. They said it was unclear whether Summers’ critics or proponents would have been more likely to respond to the poll.
Asked how they would side if a confidence vote in Summers were held today, 50 percent of 273 respondents said they would vote “confidence,” while 38 percent said they would vote “no confidence.” Twelve percent said they did not know.
Summers is unlikely to face a confidence vote at today’s special meeting of the Faculty, but he could very well face one at the March 15 meeting.
Summers’ office was informed of the poll’s results yesterday afternoon. Asked for comment, Summers wrote in an e-mail, “I recognize the strength of the concerns expressed at last week’s faculty meeting. I look forward to further candid and constructive conversation at tomorrow’s meeting and beyond as we work them through.”
Fifty-six percent of 268 respondents to the Crimson poll said they thought Summers had diminished Harvard’s image, while 18 percent said they thought he had improved it. Ten percent said he had no effect on the image of the University, and 17 percent said they did not know.
The results of a fifth question posed to Faculty members were discarded on the advice of outside experts, who said the question—”Do you think Summers stifles dissent?”—was too ambiguous to produce a meaningful result.
In all, 283 Faculty members responded before the poll closed at noon yesterday, though not all answered every question. Over the weekend, non-respondents were e-mailed and, whenever possible, phoned at their homes, to encourage participation in the poll. A final e-mail, which stressed the poll’s anonymity, was sent to non-respondents late Sunday night.
The Crimson accepted responses to the poll by e-mail and telephone. Reporters independently verified the identities of professors who phoned in their responses.
The proportion of total responses from each department closely resembled the distribution of professors in departments across the entire Faculty.
Thirty members of the physics department were inadvertently omitted from the initial e-mail but were e-mailed, phoned, and e-mailed again after the error was discovered on Sunday afternoon.
Robert Y. Shapiro, a professor of political science at Columbia University who specializes in public opinion and survey research, said The Crimson’s poll was imperfect but not invalid. The discarded question aside, Shapiro said the poll was not written in a way which would have likely swayed respondents one way or another.
“These are very educated people you’re interviewing,” Shapiro said. “They know exactly what you’re trying to find out.”
All four polling experts interviewed yesterday raised concerns about potential bias in the response pool.