A study recently released by The Steinhardt Social Research Institute at Brandeis University shows that the Jewish population in America is actually about 20% larger than previously estimated. Whereas the National Jewish Population Survey 2000-01 put the figure for the “core Jewish population” at 5.2 million, the new numbers range from 6 million to 6.4 million. Moreover, it appears that not only did the NJPS lowball the population, but it also significantly undercounted both non-Orthodox families and the younger non-Orthodox adult population, thereby considerably distorting our collective self-understanding.
Many may understandably be suffering from population-estimate fatigue, given the seemingly endless bean-counting fights and methodological wars. After all, this is not the first time that we’ve heard such pronouncements. Since the NJPS came out in 2001, estimates for the number of Jews in the United States have ranged anywhere from 4.3 million to 12.7 million. But this latest study is worth careful consideration, because it addresses a basic weakness of estimating the American Jewish population that has plagued the field since 1970, when the American Jewish community began conducting its own surveys to learn about its population size and socio-demographic character.
All surveys depend on the same basic logic: taking stock of a small group of people in order to ascertain features of the total population. For instance, you can’t interview the whole American adult population to find out how people feel about Barack Obama or Hillary Rodham Clinton, so instead you speak with a smaller cross section of people.
During these interviews you ask people not only about their views, but also some basic background facts about themselves, like age, gender, ethnicity and education. Then, if the sample turns out to be 75% female and only 5% Hispanic, you know that something is off. Luckily you’d be able to adjust the sample by comparing it to an unassailable source that accurately describes the American population — namely, the United States Census, the bedrock of our national statistical portrait.
But if we want to learn about a relatively small population such as American Jews, we face a significant hurdle. The census does not include a question about religion — as do censuses in such countries as Canada, England and Australia — and any national poll in the United States will contain only a handful of Jewish respondents. For this reason, the American Jewish community got into the business of conducting its own studies of the American Jewish population — the NJPS — in 1970, 1990 and most recently in 2001.
The whole enterprise hinges on the drawing of a representative sample, and it is here that the most recent NJPS turns out to have been more severely compromised than we had previously realized. The Steinhardt researchers demonstrate how that survey, employing a methodology of contacting people at home on their landlines, undercounted the more mobile, busy baby boomers, as well as younger people living away from home, while at the same time over-counting young Orthodox adults living at home with their parents.
The second groundbreaking contribution of the Steinhardt study is that the researchers pioneered a new approach for determining the basic socio-demographic facts about the American Jewish population — and they did it more effectively and at a much lower cost than the hugely expensive survey approach.
They realized that although the census does not look at religion, many key research institutions — among them the National Science Foundation, the Department of Education, the Pew Charitable Trusts and the Ford Foundation — have sponsored superbly executed national studies of the American population that include religion among the background questions asked of respondents. The team collected 31 studies conducted between 1998 and 2006 and then combined these into a single file of more than 73,000 Americans. With some statistical finesse, they were able to produce a much more solid picture of the size and shape of the American Jewish population than ever before.
The result? We finally have a smart, cost-effective and, above all, very robust new approach for determining the size and basic shape of the American population that is “Jewish by religion.”
Much more work, of course, remains to be done on estimating the number of people who have a Jewish identity or background but aren’t likely to describe their religious preferences as Jewish — such as secular, ethnic Jews, and children or grandchildren of intermarriage. This group is growing and will continue to do so, because being Jewish, or having a Jewish background, is no longer the strike against you that it once was.
Like many other social categories, including race, ethnicity and even gender, Jewishness has become less of a singular, fixed status over the past couple of decades. As a consequence, people’s attachment to, or identification with, Jewishness may vary over the course of their lives, and we should not assume that all movement is in the direction of down and out. After all, the “credit rating” of Jews and Jewishness in America has risen significantly in recent decades.
How will these new findings affect our deep-seated tendency to see ourselves as a people on the verge of a major disaster? Will we move away from viewing the fate of Jews and Jewishness in America as one of constant erosion? Rather than continuing to mourn ourselves as an ever-dying people, will we learn from the new portrait emerging from the Steinhardt study to think differently about who we are and the kinds of challenges we face?
Bethamie Horowitz, a social psychologist, directs research for the Mandel Foundation.