This study, designed and completed by the Association of Statisticians of American Religious Bodies (ASARB), represents statistics for 149 religious bodies on the number of congregations within each county of the United States. Where available, also included are actual membership (as defined by the religious body) and total adherents figures. Participants included 149 Christian denominations, associations, or communions (including Latter-day Saints and Unitarian/Universalist groups); two specially defined groups of independent Christian churches; Jewish and Islamic totals; and counts of temples for six Eastern religions.
It is important to understand the methodology producing these data and its limitations. While these data contain membership data for many religious groups in the United States, including most of the larger groups, they do not include every group. It is recommended that users read the notes below. Users may also want to refer to a paper by Roger Finke and Christopher P. Scheitle that explains the "adjusted" adherence rates included in the file.
- Data File
- Cases: 3,142
Weight Variable: None
- Data Collection
- Date Collected: 1999-2001
- Funded By
- The Lilly Endowment Inc., Association of Statisticians of American Religious Bodies, Glenmary Home Missioners, Church of the Nazarene, American Baptist Churches in the USA, National Association of Free Will Baptists, Presbyterian Church (U.S.A.), United Church of Christ, Wisconsin Evangelical Lutheran Synod
- Collection Procedures
- The actual data collection was carried out in the offices of the Church of the Nazarene, Kansas City, Missouri. Richard Houseal, Media Center Coordinator for USA/Canada Mission/Evangelism, oversaw the data collection. In 1999 an invitation to participate in the study was sent to every U.S. religious body listed in the Yearbook of American and Canadian Churches. Additional contacts were suggested by the Advisory Committee and by members of the Operations Committee. The initial written invitation was followed by two additional general mailings and, where needed, by special letters, personal visits and phone calls. As a result of these efforts, which extended over a two-year period, 285 denominations were invited, 149 actually participated, 22 expressed the intention to participate but were prevented from doing so, 12 declined to participate, and 102 did not respond.
Groups agreeing to participate were asked to appoint a contact person. Two forms were then sent to the contact person: instructions for reporting data; and a transmittal sheet to be signed and sent with the data collected. A state-county form for listing the statistics themselves was made available by request. The contact persons were given the option of submitting the data via their own computer print-out, sending the data electronically, or using the state-county listing provided by the Church of the Nazarene Research Center staff. The process put the major burden of work on the religious groups, since they were asked to compile data by county for all their congregations. In some cases, however, groups were able to furnish information only in the form of yearbooks or other sources. Transferring yearbook information into county data then became the responsibility of the Research Center staff. In a few cases the groups instructed staff to estimate congregational membership according to a formula, and approved the result. In all instances, however, the contact person was asked to review the statistics.
The Research Center staff employed standard procedures for checking the accuracy of data submitted. The state and national totals were checked against the county data and discrepancies adjusted. When appropriate, the estimating procedure for adherents was applied. A printout was then made the data, comparing 1990 figures to 2000 figures for each county and state. A series of four maps was also created, comparing 1990 and 2000 presence by county, location of congregations, ratio of adherents to each county's population, and number of adherents by county. These materials were reviewed by the staff and then sent to the contact person. Only after all problems raised by both the staff and the contact person were resolved were the statistics considered ready for publication. The final step was to run a series of computer edit tests to check for errors. Finding incorrect county codes and locating records with no data were the most common corrections at that step. Finally, the Church of the Nazarene staff produced the printouts of tables and maps.
- Sampling Procedures
- The Association of Statisticians of American Religious Bodies (ASARB) invited all religious bodies that could be identified as having congregations in the United States to participate. Final totals include information from 149 Christian and other religious bodies. Participants included 139 Christian denominations, associations, or communions (including Latter-day Saints and Unitarian/Universalist groups); two specially defined groups of independent Christian churches; Jewish and Islamic estimates; and counts of temples for six Eastern religions. The 149 groups reported 268,254 congregations with 141,371,963 adherents, which is 50.2% of the population of 281,421,839. There are 14 non-participating religious bodies that reported more than 100,000 members to the Yearbook of American and Canadian Churches, 2000, including all historically African American denominations. These groups reported a combined membership of 31,040,360 in the Yearbook, which is not reflected in the congregations and membership data. The lack of historically African American denominations should be noted when referencing the number of total adherents or denominations in an area.
- Principal Investigators
- Association of Statisticians of American Religious Bodies
- Related Publications
- Jones, Dale E., Sherry Doty, Clifford Grammich, James E. Horsch, Richard Houseal, Mac Lynn, John P. Marcum, Kenneth M. Sanchagrin and Richard H. Taylor. 2002. Religious Congregations and Membership in the United States 2000: An Enumeration by Region, State and County Based on Data Reported for 149 Religious Bodies. Nashville, TN: Glenmary Research Center.
Finke, Roger and Christopher P. Scheitle. 2005. “Accounting for the Uncounted: Computing Correctives for the 2000 RCMS Data.” Review of Religious Research 47: 5-22.
- Note on Membership Data
- The following is from Jones et al 2002 (see above):
Defining Membership: The most critical methodological problem was that of defining church membership. Since there is no generally acceptable definition of church membership, it was felt that the designation of members rested finally with the denominations themselves. In an effort to achieve comparability of data, however, two major categories were established: "Members” and “Total adherents.” Members was defined as “all individuals with full membership status.” Total adherents was defined as “all members, including full members, their children and the estimated number of other participants who are not considered members; for example, the ‘baptized,’ ‘those not confirmed,’ ‘those not eligible for communion,’ ‘those regularly attending services,’ and the like.”
Of the 149 reporting groups, 49 reported members and adherents; 19 reported adherents only; 67 reported members only; and 14 reported only church locations. For purposes of this data, the staff estimated the total adherents for the 67 groups that reported members only, usually according to the following formula:
The total county population was divided by the total county population less children 13 years and under, and the resulting figure was multiplied by the number of members. The 2000 U.S. Census SF1 files were used to determine for each county the total population and the population 13 years and under.
Note: There are 39 counties reporting more adherents than total population. Reasons for this discrepancy will no doubt differ from county to county, but the most reasonable explanations would include U.S. Census undercount, church membership overcount, and county of residence differing from county of membership. This is especially likely in Virginia where many cities have been separated from their adjoining counties.
- Note on African American Denominations
- Major efforts to enlist the participation of historically African American denominations were made. Generally, membership records are not kept nationally, at least not in a form conducive to participation in a study such as this. The omission of these bodies must be considered when studying religious adherence within areas of the country with significant African American population.
It should not be assumed that no African Americans are included in the reporting denominations. In 89 counties throughout the South, the reported adherents exceed the non-African American population, yet are reasonable ratios of the entire population. Many large cities throughout the South appear to have significant numbers of African-Americans reported in this book: Atlanta, Baltimore, Washington, New Orleans, Richmond, and Jackson, Mississippi; as well as smaller cities such as Lafayette, Louisiana, and Selma, Alabama. Even so, the authors advise caution in assuming that the African American community is well-represented in this study.
- Note on Eastern Christian Groups
- Alexei Krindatch, compiler of data for the Eastern Christian groups, offers these insights into the collection processes he used: A distinctive feature of this millennium edition of the Religious Congregations and Membership in the United States 2000 is a special study, which was done on Eastern Christian (broadly known as “Orthodox”) Churches in United States. The data on various Eastern Christian religious bodies were included in previous publications of Churches and Church Membership in the United States, but they were rather incomplete and sometimes arguable. Therefore, special attention was paid this time to collect—to the extent possible — accurate and comprehensive statistics on parishes and on membership in both Oriental (Armenian, Coptic, etc.) and Byzantine (Greek, Serbian, Romanian, etc.) Orthodox denominations. Whereas virtually all Orthodox Churches included in the study were able to provide updated lists of local parishes, the initial information on membership in individual parishes (which was obtained from the denominational headquarters) has required (in most cases) its further verification and unification through various statistical procedures.
Two major problems should be mentioned in this regard:
1) Different definitions of the term “full members” is used in various Orthodox Churches;
2) The absence of adequate information on the number of “adherents ”— the most inclusive category used in this study.
The second problem was due to the common approach of many Orthodox churches to consider as their members all representatives of corresponding ethnicities living in the country. In other words, according to such an approach all Armenians, Serbians, or Greeks living in the USA would be seen as the members of the Armenian, Serbian or Greek Orthodox Churches.
It should be pointed out that the data on adherents of Orthodox Churches included in this publication represent the estimated number of persons (including children) who are known to the local parish and who visit church at least during the largest religious festivals (Easter, Christmas, etc. ) . Also, in most Orthodox churches the actual requirement, and more importantly “social expectation” (from the side of parishioners), to attend church services on a more or less regular basis is not the norm—especially in comparison with many Protestant denominations. Therefore, in the case of Eastern Christian Churches it would be rather difficult to make a clear separation between categories of “attendees” and “adherents” (what was done for many other Christian denominations). Because of this reason, and in order to avoid possible confusion and misunderstanding, the data on membership in individual parishes of Eastern Christian Churches can often be thought of as equal to that of the categories of “attendees” and “adherents."
- Note on Eastern Religious Groups
- The RCMS staff was unable to locate anyone from the Buddhist, Hindu, Jain, Sikh, Tao, or Zoroastrian religions to supply information about locations of temples or numbers of adherents. However, the Operations Committee members knew of the Pluralism Project, which has been creating a directory of religious centers from diverse groups. That organization was able to supply us with county locations of temples for the above mentioned religious groups. The following is the statement they included with the data, indicating how they were able to supply the locations.
Methodology for the Pluralism Project Directory Data The Pluralism Project has been researching the immigrant religious traditions in the United States for over a decade. We have research affiliates across the United States who have provided lists of centers based on their on-site field research. We have requested participation in our data collection through our Website, and have been including the addresses of centers that have been submitted over the years, as well as changing and deleting addresses. Often our advisors have assisted us in staying abreast of changes in contact information, and newspaper articles have advised us of the opening of new religious centers.
We clean the data by receiving updates from our affiliates, crosschecking with other lists, by phone call verification (often associated with area code updates), by selected mailings, by running duplicates checks and by collaborating with information sources within the religious communities. Where possible, we have emailed contact people for centers to ask them to provide us with updates as necessary.
We verify and expand the data by consulting center lists available on the Web. Wherever possible, we consult the center’s Website itself, and we list these Websites in our online directory. For the Buddhist tradition, center lists for various schools within the tradition have been helpful, as has the Dharma Directory published by Tricycle. The publishers of Hinduism Today have graciously shared their extensive list of Hindu centers, updated through repeated mailings. Our advisor Pravin Shah has provided us with updated information on Jain centers. Sikhnet lists Sikh gurdwaras, and for the Zoroastrian tradition, FEZANA lists associated member centers, as well as small groups. Yellowpages.com has been helpful.
In the fall of 2001, the data were extensively revised in preparation for inclusion in the second edition of On Common Ground: World Religions in America. Ours is a work in progress, always subject to revision. We allow centers to self-define, and a very small regular gathering of a few religious practitioners would qualify as a religious center. The addresses, along with phone numbers and Websites where possible, are on the Web at http://www.pluralism.org/directory/index.php, and they are searchable and sortable by tradition, by state, and by center name, etc. We do our best to keep up with changes to provide accurate contact information, but obviously cannot guarantee the accuracy of all information, which is constantly changing. Also we maintain a working compilation of estimates of statistics for the overall numbers of members in religious traditions in the United States, available online at http://www.pluralism.org/resources/statistics/index.php.
- Note on Independent Church Data
- These are the steps used to obtain the figures for independent churches, groups #498 and #499 in the study. John Vaughan of the Megachurch Research Center (a ministry of Church Growth Today) carried out this project.
Scope: Congregations included in this census have a minimum adherent base of 300 people. While information for churches below 300 is available, the ability to have a comprehensive representation of churches that size becomes increasingly difficult as size decreases. Churches below 300 adherents tend to have a high percentage of bi-vocational pastors, part-time or no secretary, and more disconnected telephones, making it increasingly difficult to obtain data.
Methodology: As possible, the 2,192 churches identified in 1990 were contacted again to see if they still qualified for inclusion in the study. If contact information from the previous decade was out-of-date, information would be sought for the congregation on two separate national phone directories. If not found on either list, the congregation would be dropped from further consideration. Each of these churches was contacted to see if they still qualified for inclusion in the study: Was each still an independent congregation? Did they consider themselves to be ministering to a group of at least 300 people? If both answers were yes, then the congregation’s estimated adherence was recorded along with its county location and its charismatic /non-charismatic self-classification.
Next, lists of congregations were purchased from American Church Lists (ACL) and American Business Lists (ABL). Over 500 such churches on these lists were identified as “non-denominational” or “interdenominational” and had a reported membership of at least 300. The same qualifying questions were asked of these congregations, and the same information recorded if the churches qualified. Again, if valid phone numbers could not be found for a congregation on either of two national phone directories, that congregation was eliminated from consideration.
As time and finances permitted, attempts were made to contact the approximately 8,500 ABL/ACL churches whose membership was not known. Based on a study by author Richard Houseal of known sizes of congregations in selected Protestant denominations, fewer than 850 of those congregations would have as many as 300 in worship on Sunday morning. But the only way to determine which churches actually qualified would be through individual phone calls to each one. Again, the same qualifying questions were asked, the same information recorded, and the same lack of phone information eliminated the congregation from further consideration.
At the conclusion of the study, 1,705 independent congregations were identified and included in this report. They reported 2,051,937 adherents and a combined weekly attendance of 1,453,056.
- Note on the Jewish Estimate
- Unlike the data for Christian groups, which are estimates of institutional adherents, these data represent the entire Jewish population in the United States. No attempt has been made to categorize the population by Conservative, Reformed, etc. Information on Jewish population in American cities was published in an article in the American Jewish Yearbook 2001 (David Singer and Lawrence Grossman, 2001. New York: American Jewish Committee, pp. 253-280).
Jim Schwartz and Jeff Scheckner, authors of the article “Jewish Population in the United States, 2000,” supplied RCMS with that list. In addition to specific totals for cities (sometimes grouped as metropolitan totals), there were counts for “other places” within most states. Separately, Schwartz and Scheckner also supplied a count of synagogues by state and county. Their comments on the compilation of the synagogue list included:
The process was far more complicated and cumbersome than anticipated. It involved contact with every denomination and synagogue organization. Every Website thought to be helpful was checked. Lists were also acquired from the federations in all the largest Jewish communities. Further, a variety of local telephone directories were used as data sources. The search was exhaustive and we are confident in saying that this is the best possible enumeration of synagogues that we can do, a project which to the best of our knowledge has never previously been undertaken. Due to complex definitional and methodological issues there may be a small degree of error in the list, but again we believe that it is as accurate as we can make it.
With the listing of synagogues by state and county, RCMS staff assigned the Jewish population of the various cities to the counties they were in. When the population could reasonably be expected to be in more than one county, as in metropolitan areas, the population was distributed based upon the number of synagogues in each of the counties. Because the estimates provided by Schwartz and Scheckner were consistently rounded, the adherent distribution followed the same procedure. In 53 cases, adherents were reported for cities and counties with no corresponding synagogues. The “other places” population was then divided evenly among those counties reporting synagogues but with no corresponding city listings. Again, the rounding system of Schwartz and Scheckner was followed. Because each state had an “other places” category, every county reporting synagogues also has an estimate of adherents.
- Note on the Muslim Estimate
- Over one-third of America’s known mosques (416 of 1,209) responded to a telephone survey conducted by Ihsan Bagby for a study called Faith Communities Today. Nearly all of these were able to give both attendance figures at Friday prayer meetings and an estimated number of adherents (“How many people associate in any way with this congregation?”). The simplest method of estimating total Muslim adherents would be to determine an average size of those mosques that reported, then apply that ratio to non-reporting mosques. However, applying that procedure caused difficulties with other known facts about Muslims in the United States. According to Bagby’s own figures, corroborated by a recent survey by the City University of New York (CUNY), 33% of Muslims are of South Asian ancestry; 30% are reported as African American (which would include Blacks from Africa or the Caribbean); and 25% are of Arab descent. This would leave 12% of Muslims from other cultural groups. Further, according to a 2000 Zogby poll, corroborated by a 1990 CUNY study, only 23% of the Arab population in the United States is actually Muslim.
While national percentages cannot be applied equally within smaller geographic units, applying the “simplest method of estimating” to Bagby’s data yielded many cases where total county and metropolitan figures were incompatible with known cultural affinities as explained above. The authors asked Bagby for clarification, and his responses have been incorporated in the following procedures.
Separately, the authors were regularly reminded by those collecting data from non-western religious traditions (such as Jewish, Orthodox, and East Asian groups) that membership is not regarded the same in other religions. In western-oriented Christian denominations, members or adherents tend to be claimed exclusively by one congregation. That is, if one congregation claims 1,000 members and a neighboring congregation claims 2,000, it is likely that the denomination actually has 3,000 individuals who belong to it through those congregations. In non-western cultures, this is not always so. According to reports from non-western religious organizations, the concept of exclusive membership is western. Since western membership concepts do not apply to mosques, either, respondents from each mosque were asked, “How many people associate in any way with your mosque?” That is, two mosques that are in close proximity may be claiming many of the same people as associates (adherents in our terminology). That is, a mosque with 5,000 adherents may be claiming most of the same people that a neighboring mosque with 3,000 adherents is claiming. Bagby confirmed that this is a likely practice among reporting mosques. To compensate for such non-exclusive counting methodologies, the Muslim data was carefully analyzed. According to the survey data reported to the RCMS office, 85% of mosques have no more than 8 times the number of adherents as they have in attendance on Fridays. Of the remaining 15%, only two are not in a metro area with multiple mosques. The mosques in this 15% group seemed most likely to be claiming non-exclusive adherents. The Muslim adherent figures used in the tables of this publication were based on the following procedures:
Any mosque reporting an adherent figure no more than 8 times its attendance at Friday prayer meeting was treated as reporting exclusive adherent information. All reporting mosques reporting an adherent figure more than 8 times their attendance were regarded as reporting some non-exclusive adherent figures. If there was at least one other mosque within the metropolitan area (or county for the one non-metro mosque in this category), the adherent figure for the over-reporting mosque was reduced to 8 times the attendance figure. (This affected 61 of the 63 over-reporting mosques; the other two were the only ones in their metro area or county, and the adherent figures were retained. One was actually 12.5 times as large as the reported attendance, the other 10 times.) The mean adherent and attendance sizes of all reporting mosques (after the above adjustment to adherents) was then determined for six metropolitan classifications. This amount was allocated to non-reporting mosques in each metropolitan classification following the allocation of adherent and attendance figures to each congregation, totals were calculated for each metropolitan area. Any area whose total adherents were less than that of the largest pre-adjusted report from a single congregation was then adjusted by proportionately adjusting the adherence to the maximum previously reported for those regarded as over-reporting. That is, if the adjustments yielded 2,000 adherents within the metro area while one mosque had claimed 2,500 associates, the figure for the metro area would be adjusted to 2,500.
This process was required for three areas: Chattanooga, TN-GA; Grand Rapids, MI; and Kansas City, MO-KS. For Chattanooga and Kansas City, the increases were less than 400 each. In Grand Rapids, the above estimating procedure yielded 2,320 adherents while one congregation had reported 7,000 adherents. After consultation with Bagby and independent analysis of secular and other religious sources, the 7,000 figure was accepted. After the above procedures were run, each metropolitan area was analyzed to see if its known South Asian, Arab, and African American populations were likely to justify the reported adherents. (Metropolitan areas were used because Bagby stressed that mosques drew from more than just a single county.) South Asians include those identified as Asian Indian, Bangladeshi, Pakistani, or Sri Lankan as the single racial category in the 2000 census. African Americans included all non-Spanish Blacks reported in the 2000 census. Arab ancestry was based on 1990 data, inasmuch as complete county figures were not yet available from the 2000 census. Based on Bagby’s estimates, confirmed by the 2000 CUNY study and the Zogby poll results, we estimated that up to 38% of South Asians may be Muslim; 2% of African Americans may be Muslim; and 25% of Arabs may be Muslim. We then calculated that people within those cultural groups should account for 88% of Muslim adherents within a given metropolitan area. Of course, any particular metro area might have a far larger concentration of Muslims in a given cultural group, or fewer in any of the three (drawing perhaps from the East Asian or White groups). But this procedure would be indicative of a possible overestimation. Fifty-five of the nation’s 276 metropolitan areas had Muslim estimates exceeding the “normal” national distribution patterns. Of the five metro areas with estimates more than 10,000 larger than the anticipated ratios, Washington, DC-MD-VA-WV, had the largest absolute difference and the largest ratio, with 1.6 times the national average. When specifically asked about this metro area, Bagby defended the larger figure, although acknowledging the propriety of reducing the estimates somewhat as was accomplished in the earlier procedures. The total Muslim estimate as published includes 353,738 attendees and