NEW YORK (Nov. 19)
The head of the United Jewish Communities appointed a committee this week to investigate what went wrong with the National Jewish Population Survey.
The appointments came as Stephen Hoffman, president and CEO of the UJC, which funded the $6 million study, traded barbs with top advisers to the much-heralded survey over his decision to delay making key parts of the study public.
Members of the National Technical Advisory Committee are criticizing Hoffman for pulling its release from the organization’s General Assembly in Philadelphia this week.
Hoffman is standing by the decision he made last week after learning that the outside research firm conducting the 2000-01 study lost some data.
Hoffman said he had lost faith in the committee’s top two advisers and that the study was too important to risk going forward at this time.
He said the lost data raised concerns that could damage the credibility of what was being billed as the most extensive portrait of American Jewry to date.
“The issue is not that there’s something catastrophic — there’s no smoking gun here,” Hoffman said. “The issue is an accumulation of questions concerning NJPS.”
Hoffman’s decision to delay the NJPS came on the eve of the General Assembly, where many in the organized Jewish community were hoping to learn the latest data from the survey about Jewish identity issues such as affiliation and intermarriage.
“The integrity of the National Jewish Population Study is of the utmost importance,” Hoffman said Monday as he appointed a task force to investigate the matter.
Hoffman named McGill University’s principal and vice chancellor, Bernard Shapiro, to head the UJC task force.
Hoffman also appointed Howard Rieger, president of the United Jewish Federation of Greater Pittsburgh, and Mandell Berman and Edward Kaplan, co-chairs of the NJPS trustees, to the investigative body.
The news that the study was not being released as expected this week stunned the organized Jewish world.
Some advisers to the study criticized the delay for generating controversy and overshadowing what was essentially an accurate survey.
“We think it was a mistake not to release the data,” said Frank Mott, co-chair of the advisory committee and professor at Ohio State University.
“A molehill has been turned into a mountain,” said Mott, who along with co-chair Vivian Klaff, a professor at the University of Delaware, unsuccessfully lobbied Hoffman to go forward with the study last week.
Despite the fact that “we have a very good data set,” Mott said, “to some extent it’s now being trashed. Mr. Hoffman reacted very hastily. He is, I think, misinformed.”
But Hoffman was critical of the leading advisers on the study, saying last week’s revelations were only the latest in a series of problems.
He said he was concerned that the revelations of missing data only reached him last week despite the fact that at least one NJPS researcher know of the glitch for some time.
“I have lost total confidence in the leadership of NTAC,” Hoffman said, referring to the advisory panel.
The missing data was the “straw that broke my camel’s back” regarding the advisory panel leadership, he said.
For some time, Hoffman said, he had built an “accumulation of doubt” about the panel’s leadership, though he declined to say why.
Still, Hoffman said he had no plans to dismiss any members of the volunteer advisory panel, and none of the critics on the panel had plans to step down.
Anticipation of the new data on intermarriage and affiliation has built since the last study in 1990 produced the controversial finding that 52 percent of Jews married non-Jews in the previous five years.
That finding largely split the community into those who urged outreach to marginal Jews and those who advocated strengthening Jewish identity among those already affiliated.
This time around, the NJPS team set out a timetable in releasing the survey results.
UJC released initial population figures last month.
According to the latest study, the population has fallen to 5.2 million, down 300,000 from 1990, as the median age climbed and women waited longer to have fewer children.
Hoffman last week said that, had he known of the missing data before releasing the initial population information in October, “we would not have released it.”
But several members of the advisory panel said they thought the missing data was relatively minor.
The lost data concerned codes that telephone callers from the firm Roper Audits & Surveys Worldwide were supposed to keep when screening households for Jews, advisory committee members said.
These callers failed to keep, or later lost, codes for two-thirds of the first 14 sets of 22 surveys the overall study was based upon, committee members said.
David Marker, a member of the advisory committee and senior statistician at the firm Westat, said that at worst, the glitch caused the study to underestimate the population by 1 percent — well within a typical margin of error for such a large survey sample of 4,500.
Such missing information could also have resulted in a 5 percent overestimate of some 40,000 people “loosely associated” with Jews, Marker added, while the number of non-Jews living with Jews may have been overestimated by 1 percent.
“On a statistical basis, it’s not enough” to withhold the survey, Marker said.
Klaff said that most of his committee colleagues were “very disappointed” in the decision to delay the study.
Ira Sheskin, an advisory committee member and director of the Jewish Demography Project at the University of Miami’s Miller Center for Contemporary Jewish Studies, is among the minority of advisers who supported Hoffman’s decision to delay the study until “we’re 100 percent certain it’s OK.”
Still Sheskin said the missing data was not a big problem and he was not aware of any other major problems with the NJPS.
“As far as I know, there are not skeletons in the closet,” he said.
But Hoffman dismissed the argument that the missing data was not important enough to halt the study.
Klaff and Mott “have had an excuse that says, ‘it’s not significant’ on a number of occasions,” Hoffman said, and they “keep explaining away things.”
“One percent here, 1 percent there — pretty soon it’s significant,” he said.
Still, Hoffman expressed confidence in the ultimate outcome of the study, which he said would be released at some point in the future.
“I have fundamental confidence in the core data,” he said.
Shapiro, who is heading the investigation, said the task force will convene “in the coming weeks” and “examine the full range of issues concerning NJPS.”
One path the task force will follow will be uncovering who knew of the missing information and at what point.
Members of the advisory panel told JTA that on May 31, Roper sent a memo concerning “incomplete” information to committee members and members of the NJPS research department at the UJC, including Jim Schwartz, the NJPS research director, and Lorraine Blass, the NJPS project director.
On June 11, several advisory committee members met with the NJPS researchers and discussed the missing data, committee members said.
“We didn’t realize what the implications were,” Klaff said.
One member of the advisory committee also sent a memo to the panel co-chairs, as well as Schwartz and Blass, on Oct. 31 about “some serious problems” with missing information.
Yet it was only in the first week of November, after reviewing the issue with Roper officials, that Marker said he realized the lost data could have affected the Jewish population figures.
At that point, Marker said he relayed his concerns about the glitch to other advisory committee members and the NJPS team.
Hoffman said he only learned of the missing data when Blass informed him early last week.
That chronology would mean the NJPS team made the initial population figures public in October knowing that there had been a problem with the research.
Hoffman emphasized that he did not blame the NJPS research team for the communication breakdown. Neither Blass nor Schwartz returned phone calls seeking comment.
A Roper spokeswoman, Malkie Bernheim, said the firm was “cooperating vigorously” with the investigation and expects that “the integrity of the study will validated.”
Meanwhile, at the General Assembly, some Jewish communal leaders said those working in the filed are not disappointed at not getting the NJPS, but agreed with Hoffman’s reasons for delaying it.
Alisa Rubin Kurshan, vice president of the UJA-Federation of New York, who was helping lead a pre-assembly gathering on Jewish renaissance, said, “there’s a clear understanding that things need to be cleaned up.”