No they do not. It's all on the individual woman, not the ethnicity.
By the way, what was your sample size? Are we talking, like 500 or 1,000 or each ethnicity? Then I would begin to consider that a fairly accurate (and very fun!) control in your, ahem, experiment.