Let me state my bias right up front: I'm a skeptic when it comes to the significance of generational differences. All these "analyses" that purport to distinguish among "boomers" and "generation Xers" are so much froth, in my opinion. They take small variations, blow them out of proportion, and make grand conclusions all the while ignoring the much larger similarities also present.
Data from the Presbyterian Panel on "Worship" (May 2000), an area in which generational differences are often asserted, provide a case study.
We asked a total of 23 questions on worship preferences--8 on the sanctuary, 15 on the service--through sets of paired comparisons. Of these 23, we found 11 where the pattern of response differed by age. Statistically significant differences. Sounds like age is important, doesn't it? But remember that statistical significance tells us only that the pattern found in the sample data (in this case, Panel members) has a good likelihood of existing as well in the larger population from which the sample was drawn (the entire membership of the PC(USA)). It does not tell us anything about substantive significance.
The lack of practical import becomes clearer when we look at the patterns of response for the youngest and olders members (see the figures). Age differences in response are apparent. More older than younger members prefer pews, for example. But look again: large majorities of both age groups favor pews (81 percent among those less than 40 years, and 90 percent among those 70 and older). The age differences pale in substantive significance when compared to the age similarities.
Think of it from the perspective of a worship committee, trying to decide how to attract more young adults to worship. Do they make the service more contemporary or keep it traditional? A contemporary style will appeal more to younger than older adults, but it will appeal to only 30 percent of younger adults. In fact, 50 percent of younger adults prefer a traditional service. On this evidence, a traditional design seems to have greater appeal, even when the target group is young adults.
Apply the same approach to the other comparisons. Including mulitmedia often in worship would appeal to more younger than older members, but only a minority of younger members actually prefer regular multimedia use. Similarly, spontaneous worship would appeal to more younger than older members, but a majority of both groups prefer ordered worship. And so on.
In sum, age differences in worship preferences are real. But they tend to be small and can easily distract us from the more important finding: knowing people are members of Presbyterian churches tells us a lot more about their worship preferences than does year of birth.
That noted, I don't want to discourage experimentation that targets certain age groups--especially if the goal is to reach people who are currently unchurched. The Panel only surveys those who have already joined up. It's possible that the age- related tendencies found among current members offer important clues as to worship features that would attract young adults now outside the fold. But you may want to limit such experiments to new services. Change from the preferences noted in this survey would likely alienate more than a few of those already in the pews.
Read the complete summary of the May 2000 Presbyterian Panel on Public Worship
Email the author: Jack Marcum
Research Services