William Cohen’s father was Jewish, but his mother was not.
The young Cohen attended Hebrew school for years, but shortly before his Bar Mitzvah, the rabbi informed him he would have to undergo a conversion since he was not Jewish according to halacha, or Jewish law.
Offended, Cohen walked out of the synagogue, never to return. Now an adult – – and the U.S. secretary of defense — Cohen is married to a Christian woman.
Should William Cohen be counted as Jewish? Should he be counted among the intermarried?
Identity questions raised by people far less prominent than Cohen are challenging the planners of National Jewish Population Study 2000, which was originally scheduled to get under way last month but has been postponed until May.
In an effort to address these challenges, those involved with the study last week made one major decision: to take into account the nuances of Jewish identity, so that in analyzing things like intermarriage statistics, the William Cohens can be separated from Jews whose Jewishness cannot be questioned.
The approximately $5 million study, the first national census taken of American Jews in 10 years, is charged with providing data on everything from intermarriage rates to level of Jewish identity to philanthropic habits to assessing whether the decade’s proliferation of continuity initiatives have had any impact.
It is expected to shape the priorities of Jewish organizations and scholars for the coming decade.
At its core, the debate over NJPS 2000 parallels the ongoing outreach vs. inreach debate in the Jewish community: To what extent should scarce resources be invested in reaching, or at least studying, those who have only tenuous connections to the Jewish community as opposed to focusing on more committed Jews?
In gathering this data, the study’s planners have the sticky task of determining who to count — and what to ask.
NJPS 2000 has been postponed ostensibly so that leaders of its sponsor organization — the newly formed United Jewish Communities — have time to review the process and add input.
A steering committee overseeing the whole study is scheduled to confer next week with representatives of the UJC’s four pillars, committees formed to help set the agenda of the revamped social service and fund-raising umbrella.
But the survey has snagged its share of controversy, mostly stemming from dissatisfaction with what happened in 1990.
Several demographers — some of whom were involved in the 1990 study — have questioned the finding that 52 percent of Jews who married between 1985 and 1990 had married a non-Jew, suggesting that a more accurate count is 10 to 20 percentage points lower.
A number of Orthodox leaders claim their community was undercounted in 1990 due to methodology that may have disproportionately emphasized Jews living in areas where Orthodox Jews are less likely to cluster.
Community studies, generally commissioned by federations in order to gauge what programs are needed and which fund-raising strategies might be effective, tend not to interview those who are on the margins of the community, says University of Miami demographer Ira Sheskin.
Sheskin, who serves on the NJPS technical advisory committee, has conducted 20 such community studies and has authored a forthcoming book comparing the findings.
Hebrew University sociologist Steven Cohen and syndicated columnist J.J. Goldberg have used Sheskin’s community studies — which find much lower intermarriage rates and higher rates of Jewish ritual activity than the national study — as further reason to question the validity of NJPS’ 1990 findings.
But Sheskin says the differences can be attributed less to inaccuracy in the national study and more to the fact that it casts a wide net.
Also, says Sheskin, Jews living outside the large communities that commission the studies are more likely to intermarry.
However, a number of changes are being made in this year’s national study – – which will be surveying by phone 5,000 people, more than double the sample of 1990 — that seem, at least in part, to respond to the criticisms.
In deciding to address the nuances of Jewish identity, Sheskin said, a change has been made from 1990.
In 1990, only “yes” or “no” answers were recorded when asking whether the subject had Jewish parents and was raised Jewish.
But this time, “in addition to yes or no we are going to let people say `maybe, I suppose, half and half’ and record those answers, so when we go back and do analysis, we know who said what and can compare those different people,” explained Sheskin.
In addition, this year’s study will limit the number of interviews with marginal Jews to approximately 500 people, said Sheskin. “There are certain questions that aren’t relevant to those people, and this will save resources, so we can do better job in other places. The whole thing is a series of tradeoffs.”
Responding to concerns that changing the study will make it difficult to compare the findings from 1990, the final study may list several sets of statistics — one set that can be compared to the 1990 rate by casting as wide a net and another that uses a stricter definition of Jews and intermarriage.
After the data is collected, researchers are also considering using a variety of weighting techniques, a statistical way of compensating for segments of the population that are least likely to respond to surveys and thus might not be represented accurately.
In 1990, researchers used U.S. census weights, but a number of demographers – – both those on the advisory board and those critical of it — have questioned the validity of those weights, which assume, for example that people in the South and rural areas respond less.
But some of the demographers involved said they were concerned that no better weighting system existed.
It is not yet clear to what extent Orthodox critiques will be addressed. Rabbi David Eliezrie, a Chabad leader from suburban Los Angeles, said the study’s researchers have been cordial but not responsive to the specific concerns he outlined in a memo sent in the spring.
In particular, Eliezrie believes the study should be more careful to avoid undercounting — taking into account that a policy of locating subjects by randomly dialing phone numbers may not fairly represent densely populated Orthodox communities or account for the significantly larger family sizes in Orthodox households.
“Pizza shops don’t lie,” said Eliezrie, referring to the proliferation of kosher restaurants — and other services catering to Orthodox Jews — in communities throughout the country.
This year’s study will not repeat the highly controversial practice of conducting some of the initial screening calls on Shabbat or Jewish holidays, a cost-cutting measure in which NJPS screening was tacked onto other survey calls conducted by outside firms, said Sheskin.
That may make a small difference in counting, said Sheskin, but the main reason the Orthodox community appears anecdotally to be on the increase while statistically it is stagnant or even declining is because the community has not grown but become more observant.
“The Orthodox that we have now are Orthodox to a much greater extent than 20 years ago,” he said, but also added a caveat: “Every group always thinks there’s more of them than there are.”
For its part, UJC is optimistic it will get an accurate portrait of American Jewry.
“We think that from a technical point of view, we really have the details down well,” said Jim Schwartz, research director of UJC. “Our mandate is to produce the most accurate sample of American Jewry possible.”
The Archive of the Jewish Telegraphic Agency includes articles published from 1923 to 2008. Archive stories reflect the journalistic standards and practices of the time they were published.