Much hair-raising panic has gripped the Jewish community in recent years concerning intermarriage — and it may be displaced. As the Cohen Center in Brandeis has recently shown with the fine work of statistical analysis, “when one considers the Jewish background of the Jewish partner in an intermarriage, the difference in the Jewish beliefs and practices of inmarried and intermarried families becomes much less glaring.”
I argued this same point about intermarriage, qualitatively, in the last issue of PresenTense — writing,
the real inconvenient truth is that intermarriage is not the cause of the downturn in communal affiliation. In the science of statistics one learns that sometimes, when two things move in union, there is actually third, hidden variable that is pulling the strings on both. This is known as a hidden variable bias, an affliction of many who try and proffer causal explanations for real-world events. In the case of intermarriage and lack of affiliation, such a not-so-hidden variable is one that few are willing to talk about, and some even dismiss out of hand as unimportant. That variable is the indifference felt by marginal members of the Jewish community to the Jewish People primarily, and the Jewish tradition, as a byproduct. To put it bluntly, most people don’t know why they should give a damn.
And it seems I was right: previous studies didn’t subject their data to tests of statistical significance. They were, therefore, statistically insignificant.
Unfortunately, the lack of statistical reasoning is common to the work done by Dr. Steven M. Cohen, because, as he said, he does not believe in tests of statistical significance. As he told me following my response to his remarks at the American Zionist Movement’s Biennial, “statistical significance denies your ability as a thinker to think.” I begged to differ then, and beg to differ now. Tests of statistical significance simply deny one’s reliance on self-justifying loops of conclusions — and it’d be a good idea for the Jewish community to subject more of our highly-held assumptions to quantitative tests that will separate fact from assumption.