A closely related issue concerns ‘‘shotgunning.’’ This involves constructing a correlation matrix of all combinations of variables in a study and then relying on tests of significance to reach substantive conclusions.

Kunitz et al. (1981) studied the determinants of hospital utilization and surgery in 18 communities on the Navajo Indian Reservation during the 1970s. They measured 21 variables in each community, including 17 independent variables (the average education of adults, the percentage of men and women who worked full time, the average age of men and women, the percentage of income from welfare, the percentage of homes that had bathrooms, the percentage of families living in traditional hogans, etc.) and 4 dependent variables (the rate of hospital use and the rates for the three most common types of surgery). Table 21.20 shows the correlation matrix of all 21 variables in this study.

There are n(n — 1)/2 pairs in any list of items, so, for a symmetric matrix of 21 items there are 210 possible correlations. Kunitz et al. point out in the footnote to their matrix that, for n = 18, the .05 level of probability corresponds to r = 0.46 and the .01 level corresponds to r = 0.56. By the Bonferroni correction, they could have expected:

correlations significant at the 0.05 level and

correlations significant at the .01 level by chance. There are 73 correlations significant at the .05 level in table 21.20, and 42 of those correlations are significant at the .01 level.

Kunitz et al. examined these correlations and were struck by the strong association of the hysterectomy rate to all the variables that appear to measure acculturation. I’m struck by it, too. This interesting finding was not the result of deduction and testing; it was the result of shotgunning. The finding is not proof of anything, but it sure seems like a strong clue to me. I’d want to follow this up with research on how acculturation affects the kind of medical care that women receive and whether all those hysterectomies are necessary.