Testing the attendance gap in a conservative church

Sociology of Religion, Summer, 1999 by Penny Long Marler, C. Kirk Hadaway

Neither age nor gender produced a significant relationship with the misreporting of Sunday school attendance. There was a slight tendency for males to misreport to a greater extent than females (a three percentage point difference), but the relationship was not statistically significant at the .05 level. The respondent's usual pattern of church attendance did produce a significant relationship with the misreporting of Sunday school attendance, however.

Persons who claimed to have attended Sunday school in the last seven days were disproportionately represented among members who said they normally attend church every week - whether or not they actually attended Sunday school. In other words, persons who misrepresented their attendance were almost as likely to say they are every week attenders as persons who accurately reported their Sunday school attendance. Persons who said they did not attend Sunday school were much less likely to say they normally attend church every week.

The attendance gap was produced primarily by very active church members, not by persons who attend only occasionally. Occasional attendees freely admitted that they did not attend Sunday school. Among persons who attend church once a month or less, 88.6 percent said that they did not attend Sunday school in the last seven days. Even among persons who said that they attend church 2 or 3 times a month, 64.7 percent said that they did not attend Sunday school. Does this mean that active church members are somehow less honest about their attendance than less active church members? In a sense less active persons are "more honest" because most admit that they did not attend, but if we restrict the analysis to persons who said they attended, a different pattern emerges. Among persons who said they attend church every week, 35 percent misreported their Sunday school attendance; among persons who said they attend church 2 or 3 times a month or less, 50 percent misreported their Sunday school attendance.(11) It should be remembered, however, that the number of less active persons who claim to have attended church is quite small: only 22 persons out of 300 church members interviewed.

A NOTE ABOUT METHODS

This research combined several interlocking research strategies. The key is the consistency of the findings, rather than the force of any single element. The most important component was the Sunday school test, which involved comparing attendance counts to self-reports from specific individuals. Looking just at these 300 church members, do any of them misreport Sunday school attendance and if they do, is the number substantial? The research design was not dependent on the refusal rate, nor on the rate of non-contact. The experiment showed that 35 percent of our subjects who said they attended Sunday school misreported their attendance.

We also suggest that the level of overreporting in our sample can be projected to the entire church. Doing so is dependent on the representativeness of our sample. We lack data on non-respondents, of course, but we do have a validity check on the count-based attendance rate found in our sample. The actual rate of Sunday school attendance among our respondents was virtually identical to the actual rate of attendance among all adults enrolled in Sunday school (38 percent).