December 30, 1999
Bad News: Things Are Fine
Simply put, the study shows that intermarriage isn't the problem everybody thinks it is. Firstly, Jews aren't marrying non-Jews at a annual rate of 52 percent. That was a statistical error in a 1990 national Jewish population survey. The true figure is lower, perhaps much lower. Moreover, surprising numbers of intermarried couples raise their children as Jews. The 1990 survey said 28 percent do so. The new study shows it's as high as two-thirds in some major communities.
The study doesn't draw big conclusions, but they're obvious if you do the math. The American Jewish community is growing, not dying.
Don't pop those corks yet, though. The study's sponsor, the newly-formed United Jewish Communities of North America, is sitting on the document.
They're "reviewing" it. They can't predict the publication date. Meanwhile it's under wraps. Only a handful of copies have leaked -- "illegally," gripe UJC officials. They won't discuss the contents.
For good reason. These are the folks who, in their previous guise as the Council of Jewish Federations, brought you that 1990 survey. They've touted it ever since as the biggest and best Jewish demographic study ever done.
Its 52 percent intermarriage figure sparked a nationwide panic over impending Jewish disappearance that continues unabated. They're planning a Year 2000 update at 10 times the expense, using the same methods.
Now they're sweating bullets. The Miami study raises big questions about their methods. Partly as a result, UJC recently put Survey 2000 on indefinite hold, weeks before polling was to begin, over the research department's bitter objections. Officially the delay is to let UJC's new committees study the questionnaire. In fact it reflects new doubts about Survey 1990.
This is serious stuff. The 1990 intermarriage figure utterly transformed American Judaism. It moved Jewish spiritual survival to the very top of the Jewish community agenda. It put liberals on the defensive. It inflamed communal tensions, as Jewish movements blamed each other for the looming disaster.
Now it appears there's no disaster. Whoops.
The news puts the UJC and its researchers on the spot. They weren't just wrong. They fought bitterly to defend their blunder. A few respected Jewish population specialists (plus, ahem, one stubborn reporter) have challenged the data for years. The CJF-UJC researchers responded by vilifying the critics. Everyone else kept quiet, convinced it was too complicated to follow, yet ready to believe the worst.
The Miami study is different. It isn't an outside attack. Its author, geographer Ira Sheskin, is a member of the survey's advisory board. He's a key architect of Survey 2000.
Sheskin's study isn't meant to debunk Survey 1990. It merely summarizes local Jewish population surveys conducted in various cities in recent decades. His tables compare individual findings from 40 cities, with the 1990 national findings alongside for comparison. Only in passing, in a footnote, does he note the intermarriage error.
What's the problem? "The much cited 52% figure for intermarriages," Sheskin writes, "would be 43% if calculated only for Core Jewish households." "Core Jewish households" is survey-speak for homes that contain an actual Jew.
Besides Jews, the survey interviewed hundreds of others who had some Jewish ancestry but never considered themselves Jewish. Inexplicably, the survey included those gentiles' marriages in the intermarriage rate.
True, 43 percent is still high. But that's only the tip of the iceberg.
Critics have found other flaws that exaggerate intermarriage in the survey.
Sheskin's comparative charts seem to strengthen some of those claims.
In fact, Sheskin's charts make it clear how assimilated American Jews were made to look in the 1990 national survey. Nearly every table, from intermarriage to Sabbath candlelighting, shows a broad range of religiosity among Jewish communities, from old-fashioned, deep-rooted communities like Cleveland to newer, more transient ones like Orlando. Somehow, the national numbers always land near Orlando.
That can't be right. Older Jewish communities in the Northeast still outnumber Sunbelt transplants by two to one. The national averages shouldn't resemble Orlando.
Sheskin claims the national survey was simply more thorough than local studies. But the numbers don't compute. Critics, by contrast, argue that Survey 1990 used mistaken methods that exaggerated signs of assimilation.
The most important of these was data "weighting." All surveys "weight" or over count responses from blacks, Southerners and rural folks, to compensate for their tendency not to cooperate with pollsters. But black, Southern and rural Jews are more educated and probably more likely to cooperate, not less. On the other hand, all three groups are less likely to eat kosher food or marry Jews.
According to Hebrew University sociologist Steven M. Cohen, one of the survey's critics, removing the weights puts intermarriage at 38 percent, a figure now gaining acceptance.
But that ignores a critical question. What kinds of Jews avoid pollsters? Nobody's ever checked. Still, certain groups come to mind: the Orthodox, immigrants, Holocaust survivors. Weight those groups, and intermarriage might be as low as one-third.
The difference is critical. If half of all Jews marry non-Jews, and only 28 percent of them raise Jewish children, the prognosis is demographic disaster. That's what Survey 1990 reported, and what most Jews believe. But if intermarriage is one-third -- and if half the interfaith couples raise Jewish children -- then the community is growing. That's what the Miami study seems to show.
What made the surveyors choose the gloomier path at every turn? One reason is personnel. Some of America's leading demographers were involved, but few specialized in Jewish population. They followed standard procedure, even when logic said otherwise. Most leading Jewish demography specialists became critics.
Some critics suspect it wasn't coincidence. They say the 1990 survey was assembled with an eye toward raising Jews' consciousness, not finding the truth.
The issue isn't just intermarriage. Survey 1990 initially called 125,000 households and asked their religion. About 5,000 said "Jewish." After eliminating false positives -- pranksters, schizophrenics, Bible-thumpers calling themselves the children of Israel -- they were left with 2,441 interviewees. That's how they calculated 5.5 million Jews in America, another sign of stagnation.
But they never called back the other 120,000 to weed out the false negatives. How many Jews heard the religion question and simply hung up? A hint came in 1991, when New York's Jewish federation ran a local population survey. After the polling began, the federation started receiving calls from area police. The cops were hearing from frantic Jews who thought the P.L.O. was out to get the Jews by pretending to be the UJA.
They were wrong. It was the demographers.
J.J. Goldberg writes a weekly column for The Jewish Journal