From David Kinnaman at Barna:
The faith leanings of likely voters are related to the perceived importance of these factors. For instance:
• Evangelicals are much more likely than other voters to consider a candidate’s religious faith to be an important clue. Half of evangelicals (50%) listed that as a critical insight compared to just one-quarter of non-evangelical born again voters (23%), and by less than one out of every twenty other voters.
• Evangelicals were less likely than other religious segments to rate political experience as a key factor in their decision. Only one out of every 25 evangelicals (4%) named experience as a key. On average, one out of every six other likely voters considered experience to be an important indicator.
• The religious group that was most distinct in its decision-making process was people associated with non-Christian faiths (e.g., Judaism, Buddhism, Islam, etc.) That small portion of the likely voter population was the most likely of all to rate issue positions as the most significant factor (92%) and they were also much more likely to list party affiliation as a key indicator (32% – double the proportion of all other likely voters). They were also the group least likely to consider character (32%) to be important.
• Knowing about the religious faith of candidates was important to more than one out of ten members of a segment among just three segments: evangelicals (50%), non-evangelical born again Christians (23%), and people who attend non-mainline Protestant churches (33%).
• There were notable distinctions within the Protestant community between likely voters attending mainline churches and those attached to non-mainline churches. Mainline adults were more likely to deem candidates’ issue positions as important (91% versus 74%, respectively) and twice as likely to consider political experience to be critical (19% vs. 10%). Those attending non-mainline congregations were three times more likely to label the religious faith of candidates as important information.