29
the mainline population, which may heighten the attention of congregations to the
growing health needs of their members.
The results by religious tradition, then, suggest that of those services Bush is
targeting (the two covered here include drug treatment and job-training services) one
(job-training) is more likely to be provided by Catholic congregations (an important
political target for Bush) and at least as likely to be provided by evangelical Protestant
congregations (another key political target) as mainline Protestant congregations while
the other (drug-treatment) is slightly more likely to be provided by mainline
congregations than evangelical ones and Catholic congregations. However, the results in
these models measure the effects of religious tradition after accounting for the impacts of
all of the other measures. It is the estimated impact of religious tradition independent of
the effects of the other variables in the model. This view of religious tradition, while
important to those teasing out competing explanations for congregational programs,
matters little to those interested in the political implications of government efforts to
mobilize congregations. Consequently, it is worth deviating here to report the religious
tradition breakdowns in terms of the percentages of congregations in each tradition who
report sponsoring each kind of service in 2000 (see Table 3). The results in Table 3
reveal a consistent pattern with Catholics being most active in all of the areas covered by
Table 1 and very small (1-3%) differences between evangelical and mainline
denominations in all of those areas. Thus, efforts to target congregations already active
in these areas would target Catholic congregations heavily and could target evangelical
congregations nearly as strongly as mainline congregations, setting up a strategy that
should be politically advantageous to Bush. One of Bush's target policy areas not
Convention