School-effects research in sociology cannot be separated from concerns abou
t causality. Purely descriptive modeling justifications are untenable. Focu
sing on the Catholic school effect on learning, this article demonstrates a
n approach that places regression modeling strategies within a specific and
well-developed framework for thinking about causality. While regression mo
dels should properly remain the workhorse methodology for school-effects re
search, regression estimates should more often be subject to exacting inter
pretations and presented alongside alternative estimates of more specific p
arameters of interest. In this demonstration, propensity-score matching est
imates of the Catholic school effect for the Catholic schooled are provided
to supplement the estimates obtained by regression models. Although subjec
t to their own set of weaknesses, the matching estimates suggest that the C
atholic school effect is the strongest among those Catholic school students
who, according to their observed characteristics, are least likely to atte
nd Catholic schools. Four alternative explanations are offered for this fin
ding, each of which should be pursued in further research.