"the interplay between questions about what respondents thought the Church teaches, and questions about what they believe personally, suggests that the problem among American Catholics is more likely a matter of poor catechesis than infidelity."
The other evidence for this read is that it is very very common practice to pretest a survey wording, asking them to think out loud as they go through it - for exactly this reason, to let the researcher make sure people are interpreting the question the way they meant it. Pew is a well respected research outfit; I'd be very surprised if they had skipped this (and if those detailed tests had shown a large chunk of respondents trying to parse whether an option implied something more like consubstantiation than transubstantiation, they would change the wording)
bingo. As Brendan put it:
"the interplay between questions about what respondents thought the Church teaches, and questions about what they believe personally, suggests that the problem among American Catholics is more likely a matter of poor catechesis than infidelity."
The other evidence for this read is that it is very very common practice to pretest a survey wording, asking them to think out loud as they go through it - for exactly this reason, to let the researcher make sure people are interpreting the question the way they meant it. Pew is a well respected research outfit; I'd be very surprised if they had skipped this (and if those detailed tests had shown a large chunk of respondents trying to parse whether an option implied something more like consubstantiation than transubstantiation, they would change the wording)