New Population Survey Retracts Intermarriage Figure
The long-awaited National Jewish Population Survey 2000-2001 sent shockwaves through the Jewish community this week by retracting the single most hotly debated number in American Jewish life, the 52% intermarriage rate.
The figure, first published in the National Jewish Population Survey 1990, has become conventional wisdom in public discussion of Jewish life during the last decade. In their follow-up study, the sponsors admit the figure included a significant number of marriages in which neither partner was Jewish. The new study cites a 1990 intermarriage rate of 43%.
Publication of the new survey follows three years of debates, missteps and embarrassments, including the reassignment of its lead researcher and the temporary shelving of the $6 million study last fall after certain support data was found to be missing.
In what some saw as an indication of the tensions within the research team, the survey’s sponsor, United Jewish Communities, abruptly disbanded the survey’s technical advisory committee last week, on the eve of the survey’s publication. The co-chair of the advisory committee, demographer Vivian Klaff, responded by decrying the process by which UJC disbanded his committee, saying he was given no prior consultation.
UJC leaders said the results of two reviews of the study as well as a process of matching data with similar studies have proved the survey to be reliable and comprehensive.
“In all cases we examined, it checked out favorably,” said Lorraine Blass, the senior planner of UJC and the survey’s project director.
The study’s upbeat subtitle is “Strength, Challenge and Diversity.” Its numbers paint a grim picture, however, depicting a Jewish community in numerical decline, with a graying population, few children and a fertility rate below the natural replacement level.
At the same time, the figures also point to a renaissance in some areas, particularly in Jewish education. Most notably, the survey found that some three-fourths of all Jewish children under 17 had received some form of Jewish schooling, and more than one-third of Jewish college students had taken a Jewish studies course.
“NJPS reveals the many important and diverse ways that Jews are connected to their faith and their community,” Stephen Hoffman, president and CEO of UJC, said in a press release. “Still, the study underscores that for many, those ties are tenuous. We have a great deal of work to do to make being Jewish more meaningful and relevant for those on the edge.”
The survey found significant variations in Jewish identification and practice by region and age, with Jews in the Northeast far more likely than Jews in the West or South to light Chanukah or Sabbath candles, belong to a synagogue or keep kosher at home. Younger Jews were found less likely than older Jews to report that half or more of their friends were Jewish or to describe themselves as “emotionally attached to Israel.” However, Jews under 45 reported higher rates than their elders over 55 of observance of Jewish dietary laws (22% to 19%), monthly synagogue attendance (32% to 25%) and lighting Sabbath candles (32% to 25%), suggesting the impact of increased Jewish education, higher Orthodox birthrates and perhaps the dropping off of the least identified Jewish youth.
The number most likely to be discussed is the intermarriage rate. The last Jewish population survey, conducted in 1990 by the Council of Jewish Federations, UJC’s predecessor organization, reported an intermarriage rate of 52% for the five years previous to the study. The finding whipped communal organizations into a decade-long frenzy of debate over how to stop a supposedly hemorrhaging community.
But UJC officials admitted this week that the 52% number released in the 1990 NJPS study was in effect misleading, because it included respondents who were “considered non-Jewish” elsewhere in the same study. “In fact they were not defined as Jewish in the 1990 survey,” Blass told the Forward during a briefing.
According to the published highlights of the new study, the authors of the 1990 study had hoped “to throw as wide a net as possible in calculating the intermarriage rate, in contrast to the narrower definition of Jews they employed for other analyses in their report.”
At least one member of the National Technical Advisory Committee for both the 1990 and 2000 surveys, geographer Ira Sheskin of the University of Miami, admitted it had been “a mistake” to publish the 52% figure as the official intermarriage rate.
The recalibrated rate for 1990, based on the raw data gathered in that year’s survey, is 43%. The number was found to have risen to 47% in 2000, a short climb that may be insignificant given the margin of error, UJC officials said. The rate of increase is also described as having slowed from the 1970s and 1980s, when intermarriage rates surged.
Other pitfalls in the survey may not be as easily corrected. UJC staffers acknowledged that the 2000-2001 survey’s questionnaire framed some key questions differently from the 1990 survey, making comparisons between the two studies difficult and at times impossible. In addition, the new survey was plagued by a low 28% response rate, which researchers said reflected a growing difficulty in the survey field.
Moreover, researchers admitted that their overall Jewish population figure, 5.2 million, may be too low due to a variety of methodological factors, including an opening or “screening” question on religion that may have led some respondents to deny being Jewish.
The 1990 survey found a Jewish population of 5.5 million, seemingly higher than the current figure, but researchers cautioned that the figures were not precise enough — nor were the methods sufficiently comparable — to indicate a population decline.
Potentially more damaging to the new survey was a decision by researchers to separate 303 Jewish respondents — representing a population of some 900,000 — as having a “weaker” Jewish identity, and not to pose many specific questions to them about Jewish belief and practice. As a result, many of the survey’s findings about Jewish identity — from fasting on Yom Kippur to observance of dietary laws, visiting Israel and even “regarding being Jewish as very important” — refer only to a subgroup of 4.3 million persons considered to have a “stronger” Jewish connection. Comparisons with other studies may very well be impossible.
NJPS leaders defended their methods. Klaff said that asking questions about Jewish affiliation “doesn’t seem to make sense when someone says not only do they not consider themselves Jewish, they say to us ‘I’m not currently Jewish.’”
Asked about concerns that the screener question may have caused an undercount, Klaff said, “we don’t have evidence” of that.
But Sheskin said: “We all agree not having interviewed [the 303 respondents] with the full interview was a mistake. I could imagine a 23-year-old law student born Jewish saying ‘nah, I’m not Jewish.’ When he’s 28 and married he is suddenly Jewish.”
Sheskin also affirmed that the 5.2 million population figure is “probably somewhat of an undercount.”
Some longtime critics of the NJPS railed against the study’s discrepancies and derided the efforts to improve them as unsuccessful. “You cannot make chicken salad out of chicken s—t” said Gary Tobin, president of the Institute for Jewish and Community Research in San Francisco. “The study is based on a non-representative sample of Jews.”
According to Tobin, a longtime critic of the UJC’s research efforts, “The NJPS divides the Jewish population into three categories. The first category are all Jews they couldn’t manage to find or the ones they offended through screening questions. The second class is not in the 4.3 million category and are not worth talking about, and the third class are the easiest to find and the most worthy of discussion.”
Sociologists are already debating the study’s findings on Jewish affiliation. According to Steven M. Cohen, a senior NJPS consultant and Hebrew University professor, the survey reveals a sharp division between affiliated and unaffiliated Jews. He said that the survey suggests that reaching the unaffiliated remains a big challenge for Jewish organizations such as Jewish community centers and synagogues.
Tobin countered that reality is far more fluid than the study shows and cannot be measured by a snapshot of a person’s life.
According to the NJPS, among the more connected 4.3 million Jews, 44% percent did not belong to any synagogue or Jewish organization; 28% were “moderately affiliated” with one group, and 28% were “highly affiliated” with two or more memberships. Jewish religious and communal ties appeared to be high among those persons belonging to one or more Jewish organizations, and were sharply lower among the unaffiliated.
“Among the current unaffiliated we have some Jews who have been affiliated and will be affiliated,” Cohen said. “But my sense is the majority are Jews who desire little or no institutional contact with Jewish life.”
But Tobin countered: “One may be 75 years old and not necessarily belong to Jewish organizations anymore, but have been a very active member of the Jewish community over his life.” Tobin also argued that Jewish affiliation should be measured more broadly and not only through membership in Jewish institutions.
An external review of the technical aspects of the NJPS conducted by Mark Schulman, immediate past president of the American Association for Public Opinion Research, is expected to be released as early as September 12.
However, Klaff, of the now-disbanded advisory committee, said that the integrity of the review may have been compromised, accusing UJC of having access to it before it has been released to the public.
While UJC denies having Schulman’s review, Klaff points to the preamble of the NJPS report as evidence that UJC had some access to the report. The NJPS preamble refers to Schulman’s review, stating it “was very useful” and “validated both our assessment of the integrity of the NJPS data and its limitations.”
“Somebody has seen it,” Klaff said.
“We do not have the report,” insisted Blass of the UJC. “We’ve had preliminary discussions [with Schulman] but we do not have anything else at this point.”
The review was expected to be published along with the release of the NJPS earlier in the week but was unexpectedly delayed.
About the decision to disband the advisory committee, Blass said that the group had been put in place “to help with the design and planning and implementation of the NJPS 2000. That task has been completed.” While no such action was taken at the conclusion of the 1990 survey, Blass noted that the committee had been convened by UJC’s predecessor.