Dutch Reformed Wiki

This article was written by David Snapper, PO Box 3050, Silverdale, WA 98383. It first was printed in Calvin Theological Journal (Grand Rapids, Mich.: Calvin Theological Seminary, 1996). Except for formatting changes, the present copy is identical. Reprinted by permission.

I. The Church Growth Paradigm[]

New Church Development (NCDs) ministries in the Christian Reformed Church (CRC) are integral to participation in Christ's Great Commission. Through Christian Reformed Home Missions (CRHM) the denomination invested approximately $9,000,000 during fiscal year 1994-1995 towards planting new congregations, as well as invigorating the gospel outreach among established congregations, funding special-needs ministries, and giving enthusiastic leadership to evangelistic ministries.1 For example, between January, 1988, and December, 1994, CRHM started, or helped start, 133 new congregations in North America.2

A. Membership Growth as Ministry Effectiveness[]

Church membership growth is a goal so vital to the denomination that the CRHM's Guidelines for New Church Development (GNCD), summarizes the "Five Measures of Church Effectiveness" as spiritual maturity, ministry capacity, empowering the poor, world-wide witness, and numerical growth. The fifth, numerical growth is "the simplest measure of church effectiveness,"3 the sine qua non of NCD success. In such clear terms, numerical growth overshadows all other tests for success.

Despite this commitment to growth, actual results are mixed. Only rarely does any NCD surpass 200 individual members (the hurdle widely considered to be the mark of success) with a significant portion coming from evangelistic outreach. Most NCDs stagnate below 150 members for at least a decade.

Mixed results notwithstanding, CRHM's GNCD favorably cites Peter Wagner's declaration that a new church "should expect to pass through the 200 barrier within about twelve months after going public," and his judgment that, "If you are not through it in two years, something is going wrong and your chances of ever doing it are greatly diminished." If Wagner is correct, then something is "going wrong" with the CRC NCD process.

Lyle Schaller adds that once a slow-growth pattern has been established small churches can expect to grow only if (1) they receive new leadership or (2) there are radical changes in their environment.4 When a church grows slowly, not only is something going wrong, there is little optimism for remediation, according to Schaller.

To underscore that discouraging news, Wagner uses this analogy: "It is easier to have a baby than to raise the dead."5 Make no mistake: "the dead" refers to all congregations not in the Church Growth fast lane - most CRC NCDs. In summary: Most CRC NCDs fail to cross the 200-Barrier within the stated time period and have few prospects for ever doing so. From an administrative standpoint they are as good as dead and may better be aborted in favor of a(nother) new ministry with better prospects. Though HM has spent as much as $9-million per year, the results are discouraging. One is impelled to search for understanding of just what has gone wrong.

Recognizing the significant discrepancy between anticipated effectiveness and actual church growth results of NCD ministries, I wanted to explore this phenomenon to determine whether causes of the all-too-common disappointment among church planters might be discovered. Through careful analysis of a wide range of environmental factors which I believed might affect NCD membership growth, a number of conclusions emerged. This article is a summary of that research.

To gain a full appreciation of the importance of the findings of this research, one needs to consider the assumptions that underlay church planting policies of the CRC. These are best represented by an examination of the Church Growth Movement initiated by missiologist Professor Donald McGavran.

B. McGavran: Quantifiability and Accountability[]

In the mid-1950's, McGavran linked conversion to local church membership. He worked from conversion membership backwards to missionary technique from effect to cause. Membership was caused by effective evangelism; effective evangelism caused by effective proclamation; effective proclamation caused by proper technique. Soon, American church planting began to focus on quantifiability (numbers) and accountability (technique.)6 These words ignited the tinder of American pragmatism:

How do peoples become Christian? Here is a question to which not speculation but knowledge must urgently be applied. The question is how, in a manner true to the Bible, can a Christward movement be established in some caste, tribe or clan which will, over a period of years, so bring groups of its related families to Christian faith that the whole people is Christianized in a few decades?7

Historian Thom S. Rainer summarizes the eventual evolution:

Salvation would be "measured" by "responsible church membership." If someone was attending a church and participating actively in the fellowship, then the probability was high that he or she would be a Christian. . . . Hence the Church Growth Movement arose when salvation became quantifiable [emphasis added], and churches became accountable [emphasis added] for their numbers - in terms of membership, attendance, baptisms, and so forth.8

The logic was reduced to this sequence: Increasing church membership indicates effective discipleship. Discipleship measures effective evangelism. Evangelism measures effective leadership technique. Therefore, membership measures pastoral technique because proper pastoral technique results in growth.9 Fortunately, it was commonly believed, proper technique had became available through key persons within the Church Growth Movement, techniques which were often dispensed for as little as $295 per seminar.10 Among those persons who offered these seminars were Carl George, Robert Logan and C. Peter Wagner.

Church Growth Claims[]

Carl F. George, a strong proponent of the Church Growth movement, sounding like a circus barker, proclaimed a patented elixir for church growth, emphasizing both the quantitative and accountability aspects of church growth.

I have every confidence, then, that the astounding growth of Willow Creek Community Church has come not because Bill Hybels said, "I want the biggest church in North America." No, he said, "I want to abide in Jesus Christ," and then he was able to receive the vision God had for the spiritual harvest the Willow Creek church is to reach. That is the kind of vision God longs to give to all who will seek him.11

Quantity and technique are George's themes. "Astounding growth" is the quantifiability of success. As for technique, George divines a cause-and-effect relationship between Hybels= "desire to abide in Christ" and his resultant invigorating vision.

Effect surely follows cause, as George sees it, for God "longs" to give such a vision "to all who seek Him" - and, presumably, read George's book. Although bathed in the protective oil of mystic godliness, George's pious link of effect and cause is unproved in his own writing. He cites no statistical data beyond the Hybels example.

George is not alone in claiming what he does not prove; in his book Beyond Church Growth Robert E. Logan declares:

One of the most important questions to ask pastors and church leaders is: If you knew that you could not fail, what would you do for the glory of God and the growth of his kingdom?

A person's answer to this question might be the cup which either enables or limits his ability to receive God's blessing....

If we take seriously the message of Ephesians, here's the exciting news: Our cup never can be too large to contain the blessing that God is able to pour out upon us. The promise is that God "is able to do immeasurably more than all we ask or imagine."12

Logan, likewise, exudes only confidence that since God can do more than we imagine, He will do more than we imagine, if only one's cup is large enough. Logan leads readers to think that because they have a large cup of expectation, God must fill that cup beyond its capacity. Veiled in layers of hypothetical contrary-to-fact imagery, Logan's use of the Ephesians text (mis)leads readers to believe that right technique (necessarily) triggers God to grow churches. For tens-of-thousands of small churches, the non-growth of a congregation suggests that the pastor does not have faith sufficient to be a church growth pastor.

C. Ambiguous Results[]

With such claims one would expect prominent church growth authors to publish accounts of outreach strategies with quantifiable effectiveness which have been consistently transported to other churches. However, for the few who do attempt such a report, the results are weak.13

One example will illustrate the paucity of evidence for reproducible Church Growth effectiveness based on contemporary methodology. In his book Strategies for Church Growth, C. Peter Wagner attempts to persuade readers that church growth practices (accounting) can yield numerical (quantitative) church growth. There he adduces a handful of examples in which a ministry based on a church growth model (such as crusade evangelism, saturation evangelism, or body-life evangelism) may have produced results. However, under "crusade evangelism," Wagner laments that "It is only very rarely, if at all, that an evangelistic association will provide any information relating to whether the crusade helped the cooperating churches grow."14 and concedes, "My continuing research on this matter shows that a considerable number of crusades cannot claim such results."15 In other words, documented results of crusade evangelism cannot sustain Church Growth claims.

Second, in "saturation evangelism" Wagner "was surprised to discover that the year-long program had not increased the rate of growth of the churches. In fact, the percentage of annual growth of the seven cooperating denominations for which reasonably accurate statistics were available was greater for the year just preceding Evangelism in Depth than it was either during [the year-long program] or during the two following years."16 In short, neither crusade nor saturation evangelism can fulfill the predicted Church Growth expectations.

Wagner finally claims that a third outreach model, "body evangelism," produced encouraging results in the Bible Fellowship Church, "a denomination of some 40 or 50 congregations centered in Eastern Pennsylvania and New Jersey."17 The Bible Fellowship Church grew, adding 1,205 members in ten years, or about thirty persons per congregation in a decade.18 Such a minuscule experiment with modest results is Wagner's best substantiation of Church Growth theory and practice in the USA found in Strategies for Church Growth.

Incredibly, Wagner's failure to present solid evidence of Church Growth effectiveness based on technique is most clearly disclosed in a chapter entitled "Body Evangelism Helps Grow Churches." That chapter begins with these words: "The last two chapters on the meaning of mission and evangelism have been largely theoretical. In contrast, this chapter is mostly empirical."19 McGavran called for knowledge to be applied to missiology; apparently Wagner has neglected empirical data as part of knowledge.20

D. Acknowledged Discrepancies[]

The church planting reality is quite different from the rosy picture promised. Carl George, when he has statistics to report in How to Break Growth Barriers, cites an extensive study based on weekly worship attendance. Of 115,374 surveyed congregations in nine denominations, the average weekly attendance had not managed to grow beyond seventy-five persons in worship. To put it another way: "51% have an attendance of 75 or fewer."21 At the time of that writing, some thirty years after McGavran's Bridges for God, George's chart shows that 85% of the congregations that were surveyed remained below 200.22 Apparently, something has gone wrong not only in CRC NCD, but also apparently in most denominations across the United States and Canada. The Church Growth Movement may not have done anything to change membership growth for most congregations.

Towards a New Church Growth Understanding[]

The research underlying this report rests on the premise that Church Growth principles, as proclaimed by specialists, have been either incorrect or wrongly applied to CRC NCDs. That research constituted the core of my D.Min. dissertation for Northwest Graduate School of the Ministry23 and is currently available from the author for $15 postpaid. From the outset I believed, first, that research into actual CRC NCD growth results would contradict Church Growth claims that right missionary technique consistently causes membership growth; and I believed, second, that it would demonstrate that the strongest predictor, though not a guarantee, of CRC NCD membership growth is found in the local environment. As will be shown, these two hypotheses were confirmed.

To test these hypotheses, I decided to study all 136 CRCUSA NCD24 congregations which formally organized between 1970 and 1990, and which were reported in the 1995 Yearbook the Subject Congregations. Congregations considered successful would be those which surpassed 50 member families in size a number approximating 200 individuals.25 Results of the study are presented below, immediately following an overview of CRCUSA.

A. CRCUSA Membership Overview[]

CRCUSA, at the time of this research (1995), comprised 605 congregations and approximately 200,000 persons, yielding an average size of about 330 persons per congregation.26

Figure 1, which depicts the membership size distribution of CRCUSA, decisively demonstrates that congregational memberships are asymmetrically skewed towards smaller sizes and that approximately 30% of CRCUSA congregations remain below 200 members. Some 158 congregations (26%) report in the narrow range between 20 and 39 member families, and appear as two "spikes." In round figures, two-thirds of CRCUSA successfully crossed the elusive 200-Barrier.


Figure 1. CRCUSA: Distribution of Congregations, by Member Families, 1995

B. Subject Congregations[]

Table 1 compares the success rates of CRCUSA and its components in crossing the 200-Barrier. CRCUSA crossed the 200-Barrier at the 61% success rate, while all congregations organized before 1970 crossed that barrier at the 66% rate. In contrast, only 38% of Subject Congregations crossed that benchmark. In round figures, two-thirds of Subject Congregations failed to cross the 200-Barrier.

Table 1. Success Rate for CRCUSA and Fractions[]




Success rate for all CRCUSA, N = 605



Success rate for CRCUSA excluding Subject Congregations, n = 469



Success rate for Subject Congregations, n = 136



Why two-thirds of Subject Congregations failed to attract 200 worshiping members within five years, and in some cases twenty-five years, will be discussed in the balance of this article. After extensive preparatory research and analysis, seven factors were tested for correlation with successful membership growth. All seven are discussed in the original report; four were selected for discussion here.

C. Factor One: Local CRC Population Density[]

The first environmental factor associated with successfully crossing the 200-Barrier is a Local CRC Population of 2,000 or more CRC persons. The computer program MapLinx measures the CRC population within a 20-mile radius (the Local CRC Population ) of a selected Subject Congregation. The 20-mile radius approximated the CRC population within convenient driving distance of a congregation - the de facto Local CRC Population.

The Local CRC Population Density varies significantly. For isolated Arroyo Grande, California, the total CRC population drops below 100 persons within the 20-mile radius, which yields 0.08 CRC persons per square mile, while in another locale the corresponding number can be as high as 70,000 individuals, which yields a notable 58 CRC persons per square mile - a density more than 700 times greater than that of Arroyo Grande and other locations. How population density affects membership growth is shown below.

Table 2 summarizes the significant advantage experienced by congregations with a local CRC population of at least 2,000 individuals. The "successful" congregations (50 member families) are most frequently found in CRC clusters of 2,000 or more persons. Strikingly, for congregations planted in under-2,000 regions, the failure rate was near 80%.

Table 2. Subject Congregations, Crossed the 200-Barrier, by Local CRC population, 1995[]

Crossed the 200-Barrier

20-Mile Radius
CRC Population

Number Successful

Number Unsuccessful

0-1,999, n=76

17 (22%)

59 (78%)

2,000-70,000, n=60

34 (57%)

26 (43%)

Several additional facts may be added. First, no subject congregations attained 100 member families when isolated. Second, a large local CRC population does not assure successful church growth, for 26 such congregations (43%) remained unsuccessful while sited within a population base of more than 2000 CRC persons. Third, many of the 26 congregations planted within a CRC cluster of 2,000 or more were, in fact, cross-cultural congregations. If those congregations were filtered out, the success rate for the high density cluster would approach 80%. Fourth, though the threshold was set at 2,000 Local CRC persons for these tests, additional testing shows that as few as 1000 Local CRC persons dramatically improve an NCD's prospects for success.

If a large pre-existing local CRC population positively affects growth for Subject Congregations, what growth patterns are found when the local environment has a lower CRC population density? Similarly, what happens to membership growth results for those congregations which target another (a non-CRC) population group? Those questions will be explored in the following two sections: Factor 2: Sponsorship Type and Factor 3: Evangelism Effectiveness.

D. Factor Two: Sponsorship Type[]

As the Local CRC Population decreases, a growing congregation necessarily draws members from its non-CRC community, thus giving rise to the next factor considered, target population heterogeneity - Sponsorship Type.

Many CRC NCDs target a well-defined population, sometimes an ethnic group. CRHM maintains records of target population types for each of its sponsored congregations, as listed below. Types A, B, C and H designated congregations under the supervision of CRHM. All Type N congregations were planted under the supervision of a local congregation, a group of congregations, or a classis. Type A congregations serve the Native American population, primarily in New Mexico. Type B serves other ethnic groups, especially in larger urban areas. Type C congregations serve the "average" American communities, typically new suburbs. Type H is an older HM classification and may be considered as identical to Type C. All other congregations were designated as Type N, indicating no known specified target population. While Type N congregation commonly serve the dominant culture, some serve one or more niche target populations.

Did the target population type affect church growth success? Yes, Type N congregations attained 50 or more families twice as often as any other type, as shown in Table 3. The average of Types A, B, C and H is 15% success 85% failure.

Table 3. Subject Congregations: Membership Distribution, by Target Population[]

Ministry Types



Type A, n=5

0 (0%)

5 (100%)

Type B, n=9

2 (22%)

7 (78%)

Type C, n= 23

3 (13%)

20 (87%)

Type H, n=9

2 (22%)

7 (78%)

Type N, n=90

44 (48%)

46 (52%)

If the images used by Logan and George were valid, one would have to conclude that about 85% of pastors individually screened and selected for CRC NCD outreach had inadequate vision and faith.

It is well known that pastors of Types A, B, C and H congregations received significant training in Church Growth technique, including regional HM workshops, denominational conferences in Colorado Springs, Colorado, scholarships to non-CRC workshops, and more. These pastors are supervised and coached by regional directors and frequently by Classis Home Missions committees. They are, in summary, thoroughly trained in Church Growth theory and practice. Overwhelmingly, those same pastors failed.

E. Factor Three: Evangelism Effectiveness[]

For at least two decades it has been commonplace for some CRC people to discredit the denomination's traditional communities for (among other reasons) their failure to evangelize their secular communities. Since Types A, B, C and H congregations depend on evangelism for growth success, and their pastors have been heavily trained in Church Growth theory, one might ask if such congregations are more effective in evangelism. A detailed study of conversion growth rates showed virtually no difference between successful and unsuccessful congregations.

Table 4 shows evangelism's relatively minor role in CRC NCD growth effectiveness. Some 95 subject congregations (70%) evangelized between zero and 36 total persons per congregation in the seven years 1987 through 1993. Reading the table by columns, the percentage of small congregations which evangelized 36 or more persons is 19 of 79, or 24%. For the larger congregations the ratio is a comparable 9 to 44, or 20%. Clearly, for most congregations, higher evangelism rates are not associated with successful growth; nor are lower evangelism rates associated with a growth failure.

Table 4. Crossed the 200-Barrier, by Evangelism Effectiveness, Summary[]

Crossed the 200-Barrier

Evangelism Effectiveness (Number of Persons Evangelized, 1987 to 1993)

Number Successful

Number Unsuccessful

0-36, n=95

35 (36%)

60 (64%)

37-210, n=28

9 (32%)

19 (68%)

This represents a clear and direct contradiction to Church Growth theory's most fundamental claim - that effective evangelism is measured by church membership growth from community outreach. Membership growth in the Subject Congregations is rarely linked to evangelistic effectiveness. In one case, a congregation reported 100 converts when it disbanded for lack of members.

It therefore becomes a working hypothesis that membership streams flow into the new congregation from two or more sources the local CRC Population pool and the local non-CRC population. In the historic CRC population centers (such as Grand Rapids) the huge emigration from the city causes most suburban church growth. In medium-sized rural regions (local CRC Population of about 5,000 in Sioux Center, Iowa) the stable population's biological growth demands that new buildings be constructed. When they are finished, they are filled. Alternatively, in areas where few CRC people live, the transfer stream is virtually nonexistent, and most growth comes from evangelistic outreach. Except for the rare congregation, the evangelized inflow is insufficient to grow a large congregation within ten years. If this is the case, then the claimed linkage between Church Growth technique and evangelism effectiveness has been broken for Subject Congregations.

While this crucial component of Christ's Great Commission merits considerably more space than is available here, the research has verified that evangelism effectiveness cannot be used to indicate successful membership growth.

F. Factor Four: Source Proximity[]

Another predictor of church growth and ultimate size emerges from the Book of Acts. In Acts 2, Jerusalem's Pentecost conversion numbered 3,000 men. Far away, Paul, the God-appointed apostle to the Gentiles, delivered his sermon to the Mars Hill philosophers but netted only a few believers. Was Peter a better preacher than Paul? Should the Lord have called Peter instead of Paul? No, the first-century church, as recorded in Acts, experienced decreasing growth effectiveness at increasing radii from Jerusalem.

At increasing radii from Jerusalem, Luke recorded more complex ethnic difficulties (Antioch in Syria), fewer Jewish converts (Philippi) and slower church growth (Corinth). Jerusalem's Pentecost conversion is, therefore, the paradigm for Jerusalem, and its results should not be used to judge results in Syrian Antioch, Philippi or Corinth. One might say that between the thousands converted at Pentecost and the handful of Mars Hill converts lay 500 miles. It will be shown that a similar phenomenon occurs in the CRCUSA in relation to distance from Western Michigan.

To measure effects of Source Proximity on membership growth among Subject Congregations, it was necessary to make two determinations. First, it was necessary to assess the effects of source proximity on the congregations of the entire CRCUSA - as a standard of comparison for the Subject Congregations. Second, it was necessary to determine several scales of distance from Grand Rapid, Michigan.

Several "rings" around the city of Grand Rapids were defined. A "ring" is the region between two concentric circles of different radii. Rings were chosen based on the following radii: "zero" miles (represented by the congregations of Classis Grand Rapids East), 30 miles, 200 miles, 500 miles, 1000 miles and beyond. Admittedly, the chosen radii are somewhat arbitrary. The reader may repeat the calculations with different radii - but the results will be similar.

Table 5 summarizes results for the entire CRCUSA, showing that geographic distance from Classis Grand Rapids East affects congregational size.

Table 5. CRCUSA Membership Size, by Source Proximity, 1995[]

Radius from Grand Rapids, Michigan

Number of Congregations

Congregational Average Member Families

Congregational Average Individual Members

"Zero" (Grand Rapids East)




0 to 30 Miles (Excluding Classis Grand Rapid East)




31 to 200 Miles




201 to 500 Miles




501 to 1000 Miles




Beyond 1000 Miles




  • Zero-Mile Radius Classis Grand Rapids East is the region of zero-miles radius with 13 of 15 congregations (86%) reporting over 100 member families. The 15 congregations of Classis Grand Rapids East include 2346 member families and 9709 individuals, thus averaging over 156 member families, about 650 persons per congregation. Within its boundaries are the denominational offices, the denominational seminary, and the denominational college. For those reasons Classis Grand Rapids East probably represents the most successful (historic) portion of the Grand Rapids CRC community. The smallest congregation in Classis Grand Rapids East is Sherman Street, with 91 member families and 401 individual members. By any known standard, such a concentration is noteworthy, if not unique.
  • Thirty-Mile Radius Similar results are obtained for the additional classes of Grand Rapids and its surrounding communities. Included within thirty miles are historic CR communities, such as Grandville, Jenison, Kellogsville, Holland, Hudsonville, and Wyoming, Michigan. In addition to the number reported for Classis Grand Rapids East, the 30-mile radius membership includes a phenomenal 17,710 member families, or 62,294 individual members, just over 30% of CRCUSA, worshiping in 125 congregations. The average membership size is approximately 143 families, or 445 persons per congregation.
  • 200-Mile Radius Just beyond the Grand Rapids-Holland population concentration, congregational membership sizes drop dramatically. Beyond 30 miles but less than 200 miles from Grand Rapids, Michigan are 135 congregations, 12,574 member families and 41,354 individual members. The congregations average 93 member families per congregation and just 306 individual members per congregation. These congregations average less than half as many individuals as does Classis Grand Rapids East.27 A probable factor in this decline is geographic distance from Grand Rapids.
  • 500-Mile Radius Between 200 miles and 500 miles from Grand Rapids are 68 congregations, reporting 6,127 member families and 16,275 individual members. These congregations average 81 member families per congregation and 239 individual member families per congregation, and are a nominal 40% as large as Classis Grand Rapid East.
  • 1000-Mile Radius Between 500 and 1000 miles from Grand Rapids the ring encloses 102 congregations, 7,460 member families, and 29,740 individual members. Membership data yields an average of 73 member families per congregation and 291 individual members per congregation. These congregations report a size increase compared to the preceding ring because of the effect of the historical Iowa and Minnesota communities. Nonetheless the average is merely 50% of the member family size of congregation in the 30-Mile Radius.
  • Beyond 1000 Miles Beyond 1000 miles from Grand Rapids are 180 congregations reporting 14,124 member families and 42,942 individual members, thus averaging 79 member families and approximately 241 persons per congregation. Some people have found this statistic difficult to believe. Perhaps the easiest way to confirm these figures is with a spot-check in the Yearbook. One might calculate these figures for every church in a state (e.g. Florida where the total number of families (750) divided by the number of congregations (23) yields an average church size of 33 families per congregation).

Subject Congregations and Source Proximity With the preceding geographic analysis of CRCUSA membership completed, it now is possible to reproduce the same tests for the Subject Congregations. What was true for the entire CRCUSA is more dramatically true for Subject Congregations: The closer a congregation is located to Grand Rapids, the more likely the congregation is to attain 200 or more members.

Table 6 summarizes the relationship between membership size and radius from Grand Rapids for the Subject Congregations. (The primary difference between Table 5 and Table 6 is that the former table includes the 605 congregations of CRCUSA, while Table 6 includes only the 136 subject congregations, one of which was excluded because of its exceptional size.)28

Table 6. Subject Congregation Member Family Size, by Source Proximity, 1995[]

Radius from Grand Rapids, Michigan

Number of Congregations

Average Member Families

"Zero" (Grand Rapids East)



0 to 30 Miles (Excluding Grand Rapids East)



31 to 200 Miles



201 to 500 Miles



501 to 1000 Miles



Beyond 1000 Miles



Rather than repeat the entire chart, I call your attention to the membership size gap between congregations sited within 30 miles of Grand Rapids average and those beyond that radius. Inside the ring congregations average over 135 member families per congregation while those beyond that radius average about one-third as large. This huge difference can be attributed to geographical isolation from large CRC ethnic populations. Perhaps following the lead of McGavran's Pentecost - People Movement paradigm has caused contemporary missiology to overlook the fact that first-century congregations and contemporary congregations grow at locally-conditioned rates.

In summary, an increasing geographic distance from Grand Rapids corresponds with a decreasing average membership size, except in those isolated regions (the suburbs of Chicago, the northwest corner of Iowa, and a few others) where historic, large (2,000 or more individuals) CRC communities continue. This is no mere statistical oddity, but a real phenomenon which was evaluated through several techniques, one of which will be presented in the following paragraphs.

G. Reality Check: Congregations of 1970[]

Congregations are not merely statistics. One may rightly become impatient with charts and tables and graphs and question whether they correspond with reality. The present section helps to establish links between environmental factors and real congregations. Below is a chart comprising the vital statistics for the eight congregations organized in 1970 and still reporting for the 1995 Yearbook.

Table 7 reports the membership status of the 8 congregations that were organized in 1970 and reported in the autumn of 1994. These congregations, selected only because they are the oldest of the Subject Congregations, illustrate the influence of environmental factors on membership growth.

Table 7. CRCUSA Congregations Organized in 1970, 1995[]

Failed to Cross 200[]
State City Name Member Families (Actual Persons) Target Type CSI Local CRC Pop. Miles to Grand Rapids
CA Bakersfield White Lane 16 (65) H No 65 >1000
IA Davenport Kimberly Village 24 (58) C No 58 >200
CO Greeley Fellowship 30 (120) H No 120 >1000
MT Bozeman Gallatin Gateway 46 (191) N No 468 >1000
Successfully Crossed 200[]
State City Name Member Families (Actual Persons) Target Type CSI Local CRC Pop. Miles to Grand Rapids
IA Rock Valley Trinity 125 (568) N Yes 6,746 >500
MI Jenison Cottonwood Heights 185 (759) N Yes 67,290 <30
MI Byron Center Heritage 188 (767) N Yes 56,481 <30
MI Grand Rapids Madison Square 243 (1,052) N Yes 56,592 <30

The "Class of 1970" illustrates the case for environmental factors; the four successful congregations planted in a major CRC population grew uniformly large enough so that the smallest of the four numbers 125 member families. This stands in significant contrast to the four congregations planted outside of the CRC population clusters, where the largest congregation reported 46 families.30 Combined, the four smaller congregations have fewer members than the smallest of the larger congregations. No congregations reported in the range between 50 and 125 families, suggesting that the congregations fall into two categories created, or conditioned, by their environment. In this selected sample, the difference between the largest small congregation and the smallest large congregation is a remarkable 79 families, a gap of approximately 320 individuals is strong evidence that environmental factors influence church growth.

Other features of this table include the following.

  1. The smallest three congregations are CRHM-sponsored congregations. The largest five congregations were started under local initiative.
  2. The four smallest congregations are not located within 20-miles of a school affiliated with Christian Schools International. The four largest congregations are located near such schools.
  3. None of the four small congregations is located within 200 miles of Grand Rapids. All but one of the larger congregations are within the bounds of the Grand Rapids area.

Does this prove that environmental factors are more important than other factors in church growth? No, this table demonstrates that when the environmental factors are unfavorable it is unlikely that any available strategy will succeed on a regular basis. When the factors are favorable, it is likely that many available Church Growth strategies will succeed.

H. Reality Check: Recent Congregations[]

Several people, inquiring into the progress of this research, requested that special research attention be given to NCDs organized in the late 1980's and early 1990's on the assumption that newer congregations might report stronger growth patterns due to increased availability of skills and technology needed to plant a church. Some thought that an increasing reliance upon church growth models aggressively incorporating prayer and "vision" would provide a spiritual base for growth.

If Church Growth theory is effective, then it must produce results. The most recent congregations were planted with the greatest Church Growth training and CRHM supervision of any congregations and, therefore, should cross the 200-Barrier briskly. With these tests in mind, the CRCUSA database was queried to report the membership growth data for congregations organized after 1987.

Table 8 indicates that for the organization years 1987 through 1994, ministry Types A, B, and C successfully achieved 50 or more families in only 2 congregations of 43 a disheartening 6%. In polar contrast, Type N congregations, sponsored by local initiative, were able to report 31 of 43 congregations (72%) above 50 member families. In other words, CRHM-sponsored congregations, which Wagner claims should reach 200 in one or two years, failed at the 94% rate. Locally-sponsored congregations performed at a remarkable 72% success rate.

With these results it is virtually impossible to maintain that Church Growth technology produces church growth. In each factor tested, those congregations with the greatest Church Growth training performed less-well than the Type N congregations. Does this imply that technique is harmful? No. Rather, one should think that good technique cannot, by itself, overcome the enormous difficulties imposed by isolation, lack of support, diminished name-recognition, and similar realities of many NCDs.

Table 8. Membership Size, by Ministry Type, Organized After 1987[]

Number of Congregations

Congregations Organized Between 1987 and 1994

Successful (percent)

Unsuccessful (percent)

Type A, n=2

0 (0%)

2 (100%)

Type B, n=20

0 (0%)

20 (100%)

Type C, n=14

2 (16%)

12 (84%)

Type N, n=43

31 (72%)

12 (28%)

Exactly one known HM-sponsored CRCUSA congregation has reached 100 member families. This congregation, sponsored as a Type C ministry, is located in Lockport, Illinois, a south-side suburb of Chicago, Illinois. Interestingly, that congregation shares ZIP Code 60462 with the Orland Park, Illinois, CRC congregation, with about 12,000 CRC members (20-mile radius) and with at least two CSI-affiliated schools. This only known CRCHM congregation ever to have reached 100 member families did so in this "favorable" environment, as would be predicted by the environmental factors discussed in this report. Again, the favorable environmental factors of church siting are not mere statistical curiosities, but prerequisites to success.31

The Type N church plantings, sponsored by local congregations or classes, reveal a substantial rate of success, again confirming the value of local support.

Table 9 reduces and simplifies the preceding data. Table 9 indicates that any possible recent gains in church planting technology have not overcome the handicaps imposed by a Type A, B, or C environment.

Table 9. Recent Congregations: Membership by Type of Start, After 1987[]

Crossed the 200-Barrier

Ministry Types

Number Successful (Percent)

Number Unsuccessful (Percent)

Types A, B, C. n=36

2 (6%)

34 (94%)

Type N. n=43

31 (72%)

12 (28%)

This should confirm, by way of a sample, that the Type N environment32 has been more conducive to membership growth than any other environment for Subject Congregations.33

III. Constructing a New Understanding of the Church Growth Environment[]

It now is possible to reconstruct the experiences of many Subject Congregations. A central research concern was to determine the existence of growth patterns based on environmental considerations. Instead of one common church growth pattern for all environments, the data reveals multiple patterns based on environmental considerations. This closing section will sketch the growth patterns for the two main types of environment, the favorable and the unfavorable environments.

Table 10 compares growth results for favorable and for unfavorable environments. The favorable environment is defined by a local CRC population of 2000 or more CRC persons and a local sponsorship. The unfavorable environment is defined by a local CRC population of fewer than 2000 persons and a Type A, B, C, or H sponsorship. The majority (87 of 136, or 64%) of the all Subject Congregations belong to one of these two environments. As the table amply demonstrates, the prospects for success are multiplied by a factor of five in the favorable environment. This is in direct contradiction to current Church Growth theory and the CRC NCD expectations.

Table 10. Combinations of Environmental Factors[]

Selected Subject Congregations



Favorable EnvironmentSuccess rate for Subject Congregations with more than 2000 Local CRC persons and Type N sponsorship. n = 51



Unfavorable EnvironmentSuccess rate for Subject Congregations with fewer than 2000 Local CRC persons and either Type A, B, C or H sponsorship. n = 36



Though space does not permit a detailed discussion, the research disclosed high correlation between the success for the NCD and the proximity of a CSI-affiliate school, as also indicated in Table 7. While we cannot conclude that a CSI-affiliate school will cause church growth, its symbiotic presence in the environment may be a measure of a community's "readiness" to plant a new congregation, just as physical development is a measure of a child's "readiness" for reading. Whatever the precise relationship between church and school may be, a CSI-affiliate school was found near 20 of 21 Subject Congregations which crossed the 100-family mark. Alternatively, 83% of congregations which had no access to a CSI-affiliate school failed to cross the 200-Barrier. Such schools seem either to signal a fertile NCD environment or to contribute indirectly to NCD success.

A. Failed Components of the Church Growth Paradigm[]

Based on these results, several changes are appropriate, including the following four. First, it is proper to jettison the unsubstantiated claims of the Church Growth theory; many are simply untrue in CRC NCD. The data underlying this report are the official CRCUSA and CRHM data; this research was not based on surveys, popular sentiment or perception. The fact is that growth prospects for most CRC NCD congregations vary with the factors in the local environment.

Second, much can be said that links success to certain charismatic pastors. While that fact is not denied or minimized, it may not be helpful as a paradigm for other pastors who are not so charismatic. Until the majority of NCD pastors are truly charismatic, it may continue to be true that the NCDs in unfavorable environments will infrequently attain 200 persons. The sooner that such a reality replaces the current church growth myths, the better.

Third, probably the error of the experts has been to find the ministry features common to the largest congregations (such as cell groups, drama, contemporary music styles, etc.) and to assume that those features were the cause of growth, and, by extension, to assume that if the small congregation would imitate the larger congregation, then growth would develop. The logical and methodological flaws are left to the reader to explore. Sadly, more than a few NCD pastors have experienced the personal and professional consequences of being commissioned under false expectations.

Fourth, it is time for a sweeping review of policy among administrators with whom it has been de rigueur to demean the traditional church and its church-planting efforts. Such attitudes presupposed that newer NCDs had achieved greater effectiveness in neighborhood evangelism. In reality, neither the historic communities nor the isolated NCDs have been regularly effective at evangelism.

B. Components of a Reality-Based Church Growth Model[]

At the time of this writing the denomination is in the pit of a significant membership slump. Rather than the projected "400,000 by 2000," the denomination may be languishing nearer to the 300,000 mark at the millennium. Possibly, the most urgent ministry strategy in the immediate future will include a renewed respect for the established congregation.

It may well be that a new, viable church planting paradigm will include components such as a profound and dynamic interaction between the NCD and a local established congregation, cooperation with CSI in siting schools and congregations, reconsideration of the denominational role in supervision of new congregation, and other similar alternatives.

As one application of these ideas, it is my hope that established congregations will "adopt" smaller NCD congregations. In the case of the congregation pastored by the author, Anchor of Hope CRC, it can be fairly said that the generous affections of the members of Second Lynden CRC and San Diego CRC did more to help Anchor of Hope than any other cause except CRHM. Located in a secular, sometimes spiritually resistant community, and plagued with military transfers, Anchor of Hope would not have endured without the nurture of those two congregations. Renewed respect for such relationships may be vital to other NCDs.

In this respect Anchor of Hope is like many other small churches in the CRC, including NCDs. Though they may be isolated from the denomination's core communities and often struggle for survival, with nurturing support they can be faithful witnesses of the Gospel and their ministry, though not yielding the externally quantifiable results respected by church growth experts, may yet impact hundreds of lives for Christ.

In consequence of this research it may be said that what has "gone wrong" has not been the performance of NCDs alone; the adoption of an untested, reductionistic theology of Church Growth based on Arminian soteriology and an incomplete sociology failed to produce results. McGavran was exactly correct when he said, "Here is a question to which not speculation but knowledge must urgently be applied."

As partial answer to McGavran, we must accept that the traditional Christian community can establish new congregations with the efficiency and with as much evangelistic success as CRHM NCDs. Further, it is only rarely that a CRC NCD can be established successfully without the nurture of a nearby CRC community. And, finally, wherever we may plant a new church its survival and growth remains uniquely a work of God's sovereign grace.

Read the response by Al Mulder, Christian Reformed Home Missions


  1. Christian Reformed Church in North America, Agenda for Synod, 1994 (Grand Rapids, Mich.: Christian Reformed Church in North America, 1994), 92. Raised from approximately 290,000 members, the Home Missions= budget exceeded $30 per person in 1995.
  2. Christian Reformed Home Missions, Gathering, (Grand Rapids, Mich.: Christian Reformed Home Missions, nd), 2.
  3. Alfred E. Mulder, "Five measures of Church Effectiveness," in Guidelines for New Church Development (Grand Rapids: Christian Reformed Home Missions, 1993), B43.
  4. Mulder, "CRHM NCD Strategy and Guidelines, in GNCD, B41. 
  5. Mulder, "Why Start New Churches?" in GNCD, B12. Mulder attributes this saying to Peter Wagner and approvingly incorporates it as one reason for starting additional CRCs.
  6. Thom S. Rainer, The Book of Church Growth (Nashville, Tenn.: Broadman Press, 1993), 35. Rainer summarizes the developments through McGavran's best-known follower: "C. Peter Wagner would look to active church membership as evidence of responsible discipleship. In other words, effective evangelism could be measured by numerical church growth." 
  7. Donald A. McGavran, The Bridges of God (New York: Friendship Press, 1955), 7. 
  8. Rainer, 142.
  9. Presumably the church has been counting converts since Pentecost, as recorded in Acts 2. One must assume that Rainer is claiming that McGavran and others introduced a new intensity of ministry evaluation through accounting.
  10. The typical price of a church growth seminar led by the Charles E. Fuller Institute of Evangelism and Church Growth. P. O. Box 91990, Pasadena, CA 91109-1990.
  11. Carl F. George and W. Bird, How to Break Growth Barriers (Grand Rapids: Baker Book House, 1993), 42.
  12. Robert E. Logan, Beyond Church Growth (Old Tappan, N.J.: Fleming H. Revell, 1989), 23.
  13. The readers is encouraged to consult any of the George Barna publications for a more scientific analysis of American Church Growth.
  14. C. Peter Wagner, Strategies for Church Growth (Ventura, Calif.: Regal Books, 1987), 136. 
  15. Ibid., 138.
  16. Ibid., 141.
  17. Ibid., 144.
  18. Ibid., 145.
  19. Ibid., 133.
  20. Presumably the lack of evidence led Wagner and others to focus on successful individuals (Maxwell, Hybels and others). 
  21. George and Bird, 130. 
  22. Ibid. 
  23. David Snapper, Examination and Evaluation of Environmental Factors Affecting Membership Growth of New Christian Reformed Churches 1970 - 1995. (D.Min. diss., Northwest Graduate School of Ministry, Kirkland, WA., 1966.)
  24. CRCUSA was chosen because a key computer component of the research, Map Linx, was not available in a version for Canada. 
  25. Since the average denominational family comprises slightly fewer than four persons, it was decided to allow 50 families to be considered equal to 200 individual members. 
  26. Data was supplied by Dr. Englehard, General Secretary of the Christian Reformed Church, 2850 Kalamazoo Ave. SE, Grand Rapids, MI. Some congregations were eliminated from the database for obviously defective data. One congregation, Sunshine CRC of Grand Rapids, was eliminated because its extraordinary size skewed graphs and tables. Congregations of Canada were eliminated because a key computer program, Map Linx, could not site Canada's postal codes.
  27. Over one-half of CRCUSA is included in the 200-mile radius circle around Grand Rapids, Michigan.
  28. Sunshine CRC, of Grand Rapids, Michigan, with 843 families, was excluded from this table because its presence would skew the data very significantly. If the Sunshine CRC membership figures were returned to the table, the second row would report the "zero to 30 miles" average member families as 179 families per congregations, rather than 135 families per congregation, average. 
  29. Subject congregations located between 200 and 500 miles distance from Grand Rapids, Michigan performed with less efficiency than all other radii. Of the 8 congregations at that distance, 6 were sponsored as Type C or H congregations. 
  30. Bozeman, Montana's Gallatin Gateway had the honor of being the largest congregation of the small congregations organized in 1970. The informed reader might protest that the Gallatin Gateway CRC is, in fact, part of the Manhattan, Montana, ethnic cluster including First CRC and Bethel CRC of Churchill, as well as the Bozeman CRC. If the point is yielded, then it would be possible to say that one congregation in an area where the total CRC twenty-mile population approached 2,000, reached 46 families. But the point is not yielded, for two reasons. First, Gallatin Gateway properly fell outside the twenty-mile measuring convention. Second, Gallatin Gateway was established as a community-outreach type congregation and has been deliberately separated from the ethnic cluster. 
  31. It may also signal that CRHM has moved away from planting congregations which are geographically isolated from a CRC population cluster. At the time of writing the answer was not known.
  32. The term "Type N environment" is used loosely, here, to refer to the local sponsorship of a congregation and the many potential relationships that apparently accrue to the sponsored congregation. Sponsorship type, per se, is not thought to be a factor influencing church growth. 
  33. Even though CRHM-sponsored congregations underperformed the Type N congregations very significantly, it is not believed that the influence of CRHM can be solely responsible for the ill-fortunes of Types A, B, C, and H congregations.

This article ©1996 by David Snapper. Reprinted from Calvin Theological Journal (November 1996) by permission.