How Movements Count
October 15, 2019
Over 1,020 church planting movements (rapidly multiplying groups that have surpassed four generations of church planting in multiple streams) have been documented. Together, they comprise over 73 million believers in over 4.3 million churches.
When people hear this fact, they often ask: how are they counted? One implication of this question is, are they counted in a way that others can accept as credible? As a basis for an answer, let’s begin with a broader question: how do Christian denominations, in general, count their members? How, for example, do denominations in America count?
Denominations, or groups of churches, in the USA use various means to gather these statistics. These methods vary significantly with the size of each denomination.
Most denominations count one or both of two different types of numbers. Attendees is usually a broader and more complex number encompassing seekers, children, and new believers who have not yet met the requirements for membership. This is usually counted as the number of people regularly in a worship service. Members is usually a smaller number of people who have reached some formal stage (such as baptism).
For example, the Albanian Orthodox Diocese of America counts attendees as the average number of people (including children) who attend liturgy (the main weekly worship service) on a non-festival Sunday – that is, people who come to the main service on a day other than Christmas or Easter. The Allegheny Wesleyan Methodist Conference measures attendees as “average Sunday morning attendance,” and members as “those whose name are on the attendance roll.” Not every denomination counts both attendees and members.
Denominational statistics are usually gathered by means of some form of survey instrument – paper or electronic – which each church self-completes and returns to the denominational headquarters. Here are four examples ranging over various sizes and denominational flavors in the USA.
The Assemblies of God (3.2 million members) asks USA churches to report the total each church considers members, regardless of age, as of December 31. As their researchers told me, “This definition provides a lot of leeway for the local church.” Adherents includes all who “consider the church their home church, whether or not they are enrolled as members.” Surveys are collected via both hardcopy and online options. Responses are checked if there appear to be significant discrepancies, usually by a phone call or by checking with district staff who have a closer working relationship with pastors.
Church of the Nazarene (0.8 million members) reports are self-filed by churches. No one attempts to audit; researchers make sure the numbers add up, starting with the membership number of each church from the previous year and adding the gains and subtracting the losses to make up the new total. If numbers don’t add up, an email is sent or a phone call is made to clarify.
The Southern Baptist Convention (14 million members) uses the Annual Church Profile form to collect statistical data on all member churches. The form is returned via paper or online options. As with all denominations, not all churches fill it out every year. Returned data are compared against previous years to check for outliers; unclear data are usually referred back to state conventions for clarification.
The United Methodist Church (6 million members) groups churches into districts and annual conferences. Each church self-reports, typically using an online form. They submit their data to their district, who aggregates it for the conference, where it is aggregated for the national headquarters. A statistical team reviews the data, and if any major variances are identified, they ask the annual conference to clarify. This usually involves a phone call to the district or individual church.
In nearly every US denomination, either the church is small enough to have a specific list of all members (a “membership roll”), or it is large enough that churches report using the “honor system” – “we trust you to turn in accurate (if not necessarily precise) statistics using a fairly broad definition.” Unclear data are clarified via phone or email. “We are not the IRS [Internal Revenue Service],” one denominational researcher told me. “We don’t randomly select churches for an audit and send teams out to verify numbers. Besides, checking Sunday attendance isn’t really enough [to determine total members]: you’d have to call every member to verify.”
This highlights a complexity of denominational statistics. Attendance is a fairly easy number to estimate, even if it is not necessarily precise: just get a rough count of the number of people in a Sunday morning service. Membership, on the other hand, implies a commitment, and can introduce nuance. When does membership begin, and when does it end? If someone stops attending a church, and switches to a different church, they don’t always announce this fact. How many absences should be allowed before they are “struck from the rolls”? Are people ever struck from the rolls? How long does it take after a death? What if people go to one church on Sunday morning and another church on Saturday night? (This happens when children, for example, attend another church’s youth group.) These kinds of situations make statistical boxes difficult.
Moreover, membership usually introduces significant debates over who should be counted. One example of this is found in the article “Meaningless Membership” . The author compares attendance to membership and asks, “Convention-wide [in the Southern Baptist Convention], there are 16 million members. But only 6 million people show up on a typical Sunday. Where are the other 10 million Southern Baptists? Some are providentially hindered, but surely not 10 million.”
Movements, like US denominations, wish to count their members. There are several reasons for counting, but four seem to be common to most movements. First, movements emphasize growth, and they want to see if they are growing. Second, by counting members in various streams, problems (which can be identified in part by a correlation in lack of multi-generational growth) can be identified and addressed. Third, movements generally don’t count to measure themselves in terms of their own growth, but rather to measure themselves against the surrounding non-Christian populations. The question they are trying to answer is, are we making progress in reaching the lost? Fourth, some movements use this counting for reports to their partners in areas such as prayer, projects, and funding.
Three forms of “counting” are generally found.
Method 1 – We know everyone in the movement, whether we document them on a membership roll or not.
Some movements or pre-movements are small enough (under 1,000 members, for example) that all the groups, leaders, and even members can be known. Perhaps the stories of the individual leaders can be recounted. (For example, “This man came to faith because that grandmother prayed for his healing and he was healed. Then he shared with his brother, and their whole family came to faith.”) In their small numbers, they can easily be counted on a spreadsheet or a series of diagrams on papers. This is similar in practice to the “membership rolls” of smaller US denominations.
Method 2 – Each of the various streams within a movement know their members very well, and their numbers are aggregated to count the whole.
Some movements or pre-movements are too large to easily have everyone listed on a spreadsheet. (This “too large” threshold is often reached when a movement grows to the size of thousands of members, and definitely reached at the 10,000 member level.) Particular streams or portions of the movement, however, can be small enough individually to be similar to small movements above. They can aggregate their own numbers, and then each stream’s total can be counted together to come up with totals for the movement as a whole.
This process is similar to large US denominations that divide their churches into districts. Some streams might need to break their counts down further as they in turn get too large to count individually. However, when movements have thousands or tens of thousands of adherents, their individual streams are mostly “small-ish” and can be easily counted.
As movements become larger, they can encounter issues of security and technical logistics that make data collection risky or difficult. In a restricted-access area, a large data set of several thousand people can be very risky indeed. In places with very little technology or even very little literacy, the idea of gathering even sheets of paper might be challenging.
Because of these factors, a movement might decide to estimate their numbers based on data points like “the average number of people discipled by a leader” or “the average number of people in a group.” These sorts of estimates are just as accurate as any American denominational count (such as, “We have 10 churches, and each church has about 200 people”), although they might be less precise (see discussion of accuracy and precision below).
For example, I helped one movement estimate its total membership at between 8 and 12 million people. The estimate was made on the basis of the number of leaders, the number each discipled on average, a survey of the number of “generations” of leaders in each stream, and the geographic spread of the movement, with an estimate of its saturation of individual districts. The estimate, with a range of millions, was a truthful and accurate statement, but obviously very imprecise.
Method 3 – We are large enough to have the resources to invest in complex and regular counts.
Some movements are very large: organized in the millions, they are the equivalent of any national denomination in the United States or elsewhere (Southern Baptists, Assemblies of God, etc.). Because of their size, they have the resources to make a heavy investment in counting and do a regular census of their members (which is something very few American denominations actually do).
To accomplish this, a research team physically visits most leaders and completes a survey to gather both quantitative and qualitative data. This can result in numbers that are both accurate and very precise and that are frequently updated. Such numbers are also, for obvious reasons, highly sensitive. Very large censuses are also complex processes that are difficult for smaller groups to implement.
We know movements count their people in ways similar to how counts are made in other parts of the world. This similarity is natural: when adding up the number of people in a set and recording them, similar problems are encountered around the world and solved in similar ways. Are the counts reliable and credible? To answer that, we need to consider the various reasons why someone might look at a number and respond, “That’s just got to be wrong.”
Misunderstandings can happen when someone gives a number without explaining what that number is. Is it attendees or adherents?
This can be especially true of movements that have both “churches” and “seeker groups.” Such movements often bring pre-believers who are spiritually hungry together in groups to explore Scripture stories. Eventually these “seeker groups” (often named different things in different movements) will either disintegrate due to lack of interest, or their members will become believers and form into a church.
“Seeker groups” are therefore closer to “attendees” in a Western church. Movements don’t typically report those numbers. They are in constant flux.
Movements, when reporting, usually provide “churches” and “adherents” – but the exact definition of “adherent” will vary from place to place. Generally, the majority of adherents are baptized believers. In some movements, however, believers might take a long time to be baptized, for a variety of reasons. Some movements report children, and some don’t (as with some American denominations). Some count “adult” at a much lower age than the typical American denomination would.
As with all research, when examining or comparing numbers, it’s important to know the definitions.
In the World Christian Encyclopedia, some denominations report their membership to the last digit; others round the number (usually to the nearest thousand). The difference between exact and rounded numbers is not accuracy, but rather precision. To say a denomination has 952 or 950 or 1,000 adherents is to make a true, accurate statement within the same order of magnitude, with varying levels of precision.
To use a different example: if my daughter asks me what time it is, and I reply “It’s a quarter to ten” when the time is 9:43, I am not lying – I am being imprecise but “close enough.”
Variances in precision appear in all sorts of counts. The difference between 21 million and 20 million is less important than the difference between 20 million and 200,000. Similarly, if a given number is thought to be in the tens of millions, but precision is difficult, it might be enough to know whether it is on the low end (10 to 20 million) or on the high end (70 to 80 million).
Regardless of how denominations report their information, we need to keep in mind our own biases: a very precise number can give a false impression of precision. For many denominations – especially movements – the number of members is constantly changing. New people are joining, others are defecting; some are being born, some are dying. We need therefore to hold any single number loosely and preferably report in a rounded form (as I do, when I say there are over 73 million members of movements around the world).
Occasionally, some have told me they believe the numbers in a movement are exaggerated. The primary motivation for movements to exaggerate their numbers would be financial: high numbers could be used in fundraising appeals. We have not seen any evidence for this in the movements we have documented. In fact, we have often seen movements intentionally undercount. Sometimes this means setting aside from the count portions of the movement which they feel aren’t adequately researched, or for which the numbers aren’t really certain. In some movements, counts are reduced by a percentage out of concern for error rates in the count method.
Further, our research has shown most movements fund the vast majority of their ministries internally. The percentage of outside money is minimal, especially when considered proportionally to the size of the larger movements. In other words, if their goal were to raise money by exaggerating their numbers, they would be doing a poor job.
For most movements, exaggeration isn’t an issue due to their small size. The vast majority of individual movements are around the 1,000-member level, and the members can be known, as we have highlighted above.
Finally, we have documented movements in 5-year increments as they grew from 1990 to 2020. Movements have followed a variety of patterns of growth, plateauing, and ending over those periods of time. Movements do not follow any lockstep patterns of growth that would indicate engineered numbers.
A final claim occasionally leveled at movements is that they are outright deceptions. Either the accuser, or someone the accuser knows, “has been in the area” and “there is nothing happening there.”
When I have dug into such accusations, I have never found deception to be the case. In a few instances when deception has been found in part of a movement, the movement leaders have publicly admitted it and corrected their reports. In our experience, movement leaders are highly motivated to find any deception.
Frequently, outside accusations of deception seem not to be based on any evidence other than that the accuser or their colleagues have been in the area without seeing similar results or seeing evidence of the movement. They typically ignore that these movements are usually in extremely high-risk areas. If they are to survive, they have to become very well-versed in hiding their existence from governments and religious leaders. Many movements have had leaders “stolen” by mainstream public churches, often through offering salaries. Some have had their groups labeled as “heretical” and reported to the government by other believers. Westerners have gotten “in the know” and then without discretion have shared what they know, sometimes with very detrimental consequences. And most of all, many of these movements are so contextual that outsiders often don’t recognize them as Christian. Communities of people who dress in local fashions, gather and eat in local ways, and use local music do not look like what outsiders think of as “church.” For all these reasons, movements are often invisible to outsiders.
The 1,000+ movements we have documented have each had multiple contacts with selected groups of trusted friends. This web of trust includes people from many different nations, mission organizations, denominations, and backgrounds. Our team has usually discovered them by being within reach of such a trusted relationship (otherwise we, too, would likely not know about them). In most of the larger movements, we have personally met with leaders at various levels, who are working in very difficult situations, with significant security risks and very little money involved. We have shared meals with earnest church planters who have shown us the scars from persecution. They have told us many stories, including their mistakes, failures, and details too bizarre to make up. The similar patterns and details across unconnected movements add to the ring of authenticity.
Over 1,000 movements have been identified in the world. Each of these falls into general size categories of “small” (around 1,000 members), “medium” (some thousands to tens of thousands), or “large” (over 100,000 to some millions).
All movements, in some way or another, with some regularity, attempt a count of their membership, for a variety of reasons. They use methods similar to Western denominations, with similar levels of accuracy. Precision falls off with increases in size, which is to be expected.
Movements are loath to share this kind of information with outsiders, because it can be misused and represents a significant security risk. Movements are often “hidden” from outsiders, and the security risks often make third-party vetting of the information challenging, if not impossible. Yet at the same time, note that outsiders do not usually see the need to vet or audit the information of Western denominations.
In general, the same methods applied to Western denominations are applied to movements and should be accorded similar assessments of their accuracy.
Original article mine. First published in Accel, Vol 1 Issue 2, Nov 2019, pp. 16-20 www.accelmag.org.