Correspondence A dispute over attendance rates at Mass

CHAVES, MARK & CELIO, MARY BETH

CORRESPONDENCE To the Editors On counting heads The letter from Mary Beth Celio, director of research in the Archdiocese of Seattle, ["Numbers can mislead," September 12] has recently been...

...The Protestant and Catholic attendance rates yielded by survey always are based on the respondents who self-identify as "Protestant" or "Catholic" whether or not they are formal members of a church, are registered with a parish, have ever been baptized, or have any other formal connection to a church...
...Answering that question required calculating attendance as a proportion of the self-identified Catholic population...
...This is false...
...A study whose proclaimed purpose was to investigate the accuracy of current methods of estimating attendance-a matter of interest to few people other than church researchers-resulted in a flat statement of the number of "Catholics" attending Mass each week...
...This conclusion stands firm...
...One's opinion on this matter, however, does not affect the fact that my colleagues and I always were open and honest at every stage in this research project...
...It's conceivable, I suppose, that Doyle and I and other diocesan researchers affiliated with the Catholic Research Forum all wildly misinterpreted what we were told...
...Part of the research project discussed here was carried out at the University of Notre Dame...
...From the editors: This correspondence is now closed...
...I, for one, did not anticipate that this research would receive the amount of attention it has received...
...Indeed, we went to great lengths to ensure that the counts we used to calculate attendance rates were, if anything, overestimates of actual attendance...
...To say church attendance is "lower than generally believed" is one thing...
...Never did we assure Celio or anyone else that diocesan names would not be published...
...On the question of the quality of data: A major point in my September 12 letter to Commonweal was that the use of the October Mass attendance counts in sociological research is fraught with perils for the unwary...
...This is not a narrow academic matter...
...On the first point, Celio's allegation is simply false...
...I shared that understanding of Chaves's data needs and the conditions of my cooperation with my counterparts in the Rockford, Cincinnati, and New York dioceses...
...Ruth Doyle, director of pastoral planning for New York, reports that she understood herself to have reached the same agreement with Chaves...
...Although Chaves contends that the counts he used were of uniformly high quality, anyone who has ever worked in a parish or a diocesan office knows that the counts are only as good as the individual usher or sacristan or choir director in each parish who takes the count and sends it (or doesn't) to the chancery...
...church attendance...
...Such surveys form the basis of scholarly and popular understanding of U.S...
...In making such a test, it was necessary for us to use the same denominator that surveys use...
...No such assurance was ever given, and it is outrageous and hurtful for Celio to claim the contrary...
...All the counts that we reported, including Seattle's, clearly were of high quality...
...However, after Chaves found that reported attendance ranged from 15.9 percent to 67 (!) percent of self-identified Catholics in his sample dioceses, he reported an overall average and never discussed why the dioceses differed so radically from one another and why the computed average differed so substantially from more than fifty years of data collected by Gallup and others...
...Nowhere does this correspondence mention some prior assurance of anonymity...
...and (b) that our research methodology was flawed...
...Knowing what I now know about the impact of our work, I am sympathetic to the suggestion that it might have been better to have tried to find a way to publish our results sans the names of the dioceses from which the data came...
...For example, we did not report data from any diocese that estimated attendance rather than actually counting heads...
...She suggests that, instead of the population of self-identified Catholics in each diocese, we should have used either the population of registered Catholics or the population of Catholics affiliated with the church-this last estimated by using the number of baptisms and funerals in each diocese...
...We assessed the quality of these data carefully...
...This number is now the coin of the realm (witness recent ads for the New Oxford Review...
...To be honest, I now question the wisdom of that decision...
...I can say, however, that from the time I first spoke with Chaves, it was my clear understanding that the dioceses which agreed to provide him with the data he requested would not be identified by name in any published article...
...It is worth restating our conclusion: Although surveys based on individuals' reports of their own behavior routinely show that approximately 50 percent of self-identified Catholics and approximately 40 percent of self-identified Protestants attend church in any given week, we showed that the actual weekly attendance rates for self-identified Catholics and Protestants are more like 27 percent and 20 percent respectively...
...The writer is a visiting associate professor at the University of Illinois at Chicago...
...We, however, wanted to know whether standard surveys overstate weekly church attendance...
...This is testimony to the good work done by diocesan researchers across the country...
...I was shocked to read in Commonweal that a disagreement over whether or not to publish diocesan names has been transformed into a promise that we would not do so...
...Weekly church attendance in the U.S...
...Celio claims (a) that I assured her and researchers in other dioceses that no diocese would be identified by name in our published articles...
...is much lower than it is generally believed to be...
...Our research was designed to test the accuracy of attendance rates generated by surveys...
...CORRESPONDENCE To the Editors On counting heads The letter from Mary Beth Celio, director of research in the Archdiocese of Seattle, ["Numbers can mislead," September 12] has recently been brought to my attention...
...Conceivable, but not likely enough to warrant Chaves's flat statement that my allegation was "simply false...
...Additional research needs to be done on Mass attendance, but it must use accurate, comparable data and appropriate reference points so that both sociologists and parish/diocesan planners can use the results with confidence...
...Hence, Celio's point that there are more self-identifed Catholics than appear on official lists of registered Catholics (and her arithmetic truism that using registered Catholics in the denominator would generate a higher attendance rate than we calculated using self-identified Catholics) is irrelevant...
...It shouldn't be...
...On the one hand, she suggests that the diocesan attendance counts are not of sufficiently good quality to warrant using them as we did...
...MARK CHAVES Chicago, III...
...Finally, Chaves seems uncertain about what he was trying to do through his research, and thus confuses many of the issues around Mass counts and Mass attendance rates...
...On the contrary, when the subject arose-which it rarely did-I always was completely honest and forthright about our intention to publish the names of the dioceses from which the data came...
...As for methodology, Celio offers two criticisms...
...to say it is 27 percent is another...
...I can certainly see why someone might want to know the (Continued on page 29) CORRESPONDENCE (Continued from page 4) percentage of registered Catholics attending services weekly...
...dioceses is admirable but misplaced, and was the major focus of my letter to him requesting that he omit the Seattle data from his report...
...Hence, to compare self-reported attendance from surveys to actual attendance based on head counts, it was essential to use the number of self-identified Catholics in the denominator of the rates we calculated...
...It is difficult, if not impossible, for scholars to foresee all of the potential political ramifications of their research...
...This was a conclusion much more interesting to the larger world and naturally garnered some media attention...
...On the other hand, Celio argues that we used the wrong denominator in calculating Catholic attendance rates...
...Using the number of registered Catholics-or any other number-would have been analogous to comparing apples and oranges...
...This stated purpose was a limited one and could arguably be addressed using the data he had available to him...
...Chaves's faith in the reliability of one year of Mass count data from a nonrandom sample of just over one-fourth of U.S...
...I do not have any smoking gun in the form of written assurances or statements of intent...
...Quite simply, I felt (and continue to feel) that he misunderstood the limitations of the data we supplied, and I did not want the inclusion of Seattle's data to imply that we agreed with his methodology or interpretation...
...The writer replies: It's a classic case of "He said, she said...
...Celio, in particular, asked us to suppress the Seattle data only after it had already been published in the American Sociological Review...
...At the time, we believed that publishing the diocesan names was necessary for readers to assess the accuracy, breadth, and validity of our research...
...This criticism is misplaced...
...I have correspondence showing that the Catholic Research Forum asked us to consider suppressing the names of dioceses only after they saw what the results were...
...Either Mark Chaves gave the assurances I described, or he did not...
...MARY BETH CELIO Seattle, Wash...
...The aim of his research, he writes, was to test the accuracy of attendance rates generated by surveys...
...we evaluated the counting procedures used in each diocese and declined to report count data gathered from more than a dozen dioceses because they were of questionable quality...

Vol. 124 • November 1997 • No. 20


 
Developed by
Kanda Sofware
  Kanda Software, Inc.