Democratizing access to the best data on religion since 1997

Archive Navigation

View Question Bank

Archive Categories

Sort the above Archive Category by:

Browse Alphabetically
Browse All Categories
Browse Newest Additions

File Summaries

Search Data Archive

Salt Lake City and San Francisco Surveys of Mormons, 1967-1969 (Salt Lake City Sample)




Mauss, A. L. (2020, May 1). Salt Lake City and San Francisco Surveys of Mormons, 1967-1969 (Salt Lake City Sample).


These are the first large-scale surveys of Mormons ever conducted, with or without church auspices, based upon probability samples of adult Mormon householders. As of century's end, these are the only such surveys available to the public, although the LDS Church has in recent years conducted many private surveys of its own for various purposes. Large as they are, the Mauss surveys cannot be considered representative of all Mormons everywhere, of course, even in the 1960s, but they are certainly representative of Salt Lake City Mormons then, as well as of the most highly urbanized San Francisco Mormons (and, by extension, perhaps of Mormons in similar sections of other American cities).

The questionnaire and the survey procedures were modeled in large part after those of the Glock and Stark 1964 survey of Northern California churches (which did not include Mormons). Accordingly, the 23-page questionnaire includes many items intended to measure various dimensions of religiosity; the usual demographic and social class information; the conversion experience (for converts); religious defection and reactivation; civil liberties; and attitudes toward blacks and Jews.

The nature and scope of these Mormon surveys, which used identical questionnaires, were intentionally guided by those of the Glock and Stark instrument and were carried out during the principal investigator's doctoral studies under Glock. The survey procedures in Salt Lake City were fairly straightforward and yielded data as representative for Mormons as the Glock and Stark survey was for Catholics and Protestants. However, the rationale for selecting the two Mormon wards in San Francisco, and none of the others, was that the Bay and the Mission Wards consisted disproportionately of the most "urbanized" church members (as opposed to suburban neighborhoods) -- that is, those closest to the inner-city, the apartment dwellers and the temporary residents. These two Mormon wards also included most of the ethnic minorities among Mormons in San Francisco (primarily Hispanic, Polynesian and Asian-Americans). The idea was to get as stark a contrast as possible to the Salt Lake City Mormons.

The ARDA has added five additional variables to the original data set to enhance the users' experience on our site.

Data File

Cases: 958
Variables: 267
Weight Variable: None

Data Collection


Funded By

The Presiding Bishop's Office of the Church of Jesus Christ of Latter-day Saints provided informal sponsorship by "vouching for" the principal investigator in his efforts to obtain the necessary membership rolls from local bishops. No funds and no other forms of sponsorship were provided.

Collection Procedures

Each respondent received in the mail a questionnaire, with appropriate instructions, and a prepaid postcard addressed to the principal investigator. Postcards contained only a respondent's name, address, ward and declaration that s/he had completed and mailed the questionnaire. Respondents were asked to mail the postcards separately from the questionnaires, so that they could be checked off the ward list without compromising the anonymity of the questionnaires themselves.

A month after the original mailing, a follow-up reminder was sent to all non-respondents. After three months, questionnaires were no longer being received. At that point, the principal investigator made appointments to consult with each bishop and/or ward clerk over a list of non-respondents. For each non-respondent, certain information was obtained to permit estimates of non-response bias. This information included the non-respondent's gender, age, occupation and level of church activity.

Sampling Procedures

There were two surveys, each with a different population universe. The first survey universe included all LDS members (Mormons) of the Greater Salt Lake City area, from the state capitol in the north to and including Murray, about 60 blocks to the south; and from the mountain bench on the east to and including Taylorsville, about 20 blocks to the west. In 1967, this population contained 243,572 Mormons in about 50,000 households, comprising about 400 wards with membership sizes ranging from 372 to 1,019 (mean and mode between 625 and 650). The second survey universe included all LDS members (Mormons) in the eastern half of the city of San Francisco, from the Panhandle of the Golden Gate Park on the west to the Bay on the east (comprising Bay Ward and Mission Ward only), about 550 households.

In Survey One (Salt Lake City), a list was obtained from the Presiding Bishop's Office of all the wards (congregations) and their respective membership sizes in the population universe. A block of numbers was then assigned to each ward corresponding to its membership size. That is, the first ward on the list had 610 members, so that ward was assigned numbers 1 through 610; the next ward had 735 members and was thus assigned numbers 611 through 1,345, and so on. After that, a table of random numbers was used to select 10 wards for the survey, depending on which block(s) of numbers contained the 10 numbers chosen from the random table. For example, if one of the random numbers was 1,102, the second ward on the list would have been selected for the sample (since it contained numbers 611 through 1,345). With this system, a ward's chance of being selected for the sample was proportional to its membership size. The 10 wards selected by this method turned out to be distributed geographically. In Survey Two (San Francisco), only two wards were involved (Bay Ward and Mission Ward), so membership lists were obtained from the two bishops and the same procedures were followed as in the 10 Salt Lake City wards. A response rate of 55% was obtained (N=296).

Over-sampling and weighting
No special system was imposed for either oversampling or weighting. However, the sampling system, more or less automatically, over-sampled inactive church members, females, singles, and, in San Francisco, ethnic minorities and the residentially transient.

Principal Investigators

Armand L. Mauss

Related Publications

Mauss, A.L. (1994) The Angel and the Beehive: The Mormon Struggle with Assimilation: University of Illinois Press

Mauss, A.L. (1976) "Shalt the Youth of Zion Falter? Mormon Youth and Sex: A Two-City Comparison," Dialogue: A Journal of Mormon Thought 10(2): 82-84 (1976).

Mauss, A.L. (1972) "Moderation in All Things: Political and Social Outlooks of Modern Urban Mormons," Dialogue: A Journal of Mormon Thought 7(l): 57-69 (1972).

Mauss, A.L. (1972) "Saints, Cities, and Secularism," Dialogue: A Journal of Mormon Thought 7(2): 8-27.

Mauss, A.L. (1968) "Mormon Somitism and Anti-Semitism," Sociological Analysis 29 (Spring): 11-27

Mauss, A.L. (1966):"Mormonism and Minorities" (doctoral dissertation University of California, Berkeley, 1970).

Mauss, A.L. (1966) "Mormonism and Secular Attitudes toward Negroes," Pacific Sociological
9(2): 91-99

Additional notes on the construction of indices

The last ten variables in this data file are indices. These are additive in nature and are constructed by combining conceptually similar questions. Thus, for the index of Orthodoxy, for example, responses to questions about belief in God, belief in Jesus, belief in the Devil, and belief in Joseph Smith's First Vision were added together, with 4 points given for each "definitely true" response, zero for each definitely "not true," and intermediate responses with correspondingly intermediate scores (e.g., 2 for "don't know" or similar responses).

Allowing for obvious differences in the formatting and response categories of the various questions, all of the indices were constructed in this same basic manner. After having been constructed, each index was validated by at least two different "criterion questions," and all the validations proved very strong. The criterion questions were also used to estimate the appropriate scoring on the index for cases where respondents failed to answer one or more of the indices.

Our Sponsors

Our Affiliates

© 2024 The Association of Religion Data Archives. All rights reserved.