I have never been satisfied with anecdotal answers or conventional wisdom. When students make a comment like "Research shows … ," I demand that they show me the sources of the research and demonstrate the validity of the conclusions.

In nearly 20 years of research on the American church, our studies have uncovered some issues that defy the conventional wisdom of the day. Such was definitely the case in the breakout churches project. "Breakout Churches" is a book on churches, modeled on author and researcher, Jim Collins's "Good to Great" framework that we introduced you in the March 22 issue of the Church Health Today.

The research team will readily agree that this project was one of many lows and only a few highs. We faced monumental challenges as the process unfolded for finding the breakout churches in America. What was the process like?

Stage one: Define the criteria

What defines a "breakout" church? I am sure we all have different opinions on this question. We do not claim that we have discovered the perfect formulaic approach to defining great churches. We believe, however, that our criteria provide an acceptable screen for the churches in the United States.

1. The church has had at least 26 conversions annually since its breakout year. This number was the minimum we accepted in our previous research on effective evangelistic churches. Simply stated, we believe that any healthy church should be reaching at least one person with the gospel every two weeks.

2. The church has averaged a conversion ratio no higher than 20: 1 at least one year since its breakout year. This ratio answers the question, how many members does it take to reach one person for Christ in a year? A ratio of 20: 1 suggests that it takes 20 members one year to reach one person.

3. The church had been declining or had plateaued for several years prior to its breakout year, or the church was experiencing some type of stagnation not readily apparent in the statistics.

4. The church broke out of this "slump" and has sustained new growth for several years.

5. The slump, reversal, and breakout all took place under the same pastor. We believe that this criterion, though limiting, was absolutely necessary. We felt compelled to find breakout success stories that took place without a change in leadership. The typical solution to stagnated churches is to replace the pastor. Unfortunately, there are not enough "breakout pastors" to lead even 5 percent of the churches in America. We sought stories of changed leadership values rather than stories of changing leaders.

6. Since the breakout point, the church has made a clear and positive impact on the community, and there are numerous stories that lives have been changed as a direct result of this. While this final criterion is subjective, we wanted more than statistical measures for our breakout churches. You will see that the churches that survived our screens are clear examples of life-changing and community-changing churches.

Stage two: Find the churches

We started the project with a great deal of optimism. We received data on some 50,000 churches. We expected to find a number of breakout churches in this mix, but the yield was surprisingly much less than we thought it would be.

Many people warned us that the selection process with churches may prove more difficult than with businesses. Large companies are easy sources of information. Not only are their financial records available for public scrutiny; but many companies are written about in outside sources like magazines and newspapers. The skeptics were right. Rarely did we easily find information on these churches.

We then began the arduous process of contacting key persons around the nation who could assist us in the search. We wrote to well-known pastors, denominational leaders, church leaders selected according to their geographical area, civic leaders and others. Early on we faced frustration when a few denominational leaders chose not to help us and became protective of their churches' information.

The good news is that we found 13 churches that clearly met our strict criteria.

Stage three: Look for comparison churches

Two sets of "great" churches are omitted from this study because of our defined criteria. The first set includes churches that qualify according to our statistical criteria but do not show a period of struggle. Great churches would include Saddleback Valley Community Church in California, under the pastoral leadership of Rick Warren, and Southeast Christian Church in Louisville, Ky., led by senior minister Bob Russell. But these are not breakout churches. We were interested in churches that had broken out of a "slump." Some great churches have never experienced a real downturn.

Another group of churches were excluded as "great" because their breakout period took place under a new pastor. St. Stephen Baptist Church in Louisville, Ky., met all of our statistical criteria, but its breakout took place when Kevin Cosby became pastor. We were determined to find churches that made the change under the same leadership that had experienced struggles in earlier years.

The largest group of churches left out of the study - about 96 percent of the churches in America - were excluded because they did not meet the criteria based on our previous studies of effective evangelistic churches. These are the "good" churches. Most of them have good pastors and good people, but they have not broken out of their mediocrity.

With most of these churches, somewhere in their histories they have become satisfied with the status quo. They resist change and often seek to minister only to those inside the church. They have some or many programs. They may even have large budgets. But they are not making a significant impact on their communities, nor do they see significant numbers of changed lives in their congregations.

These "good" churches became our source for direct comparison with the breakout churches. We carefully selected three good churches to compare with each breakout church. The comparison churches were similar in size at the breakout point of the breakout churches. Their doctrinal positions and demographic compositions were similar. The differences we found in each set of churches became our focal point in better understanding the factors that move churches to greatness. We called these issues the "chrysalis factor."

There are still two more stages in our process to find and research these breakout churches. Join us next time as we continue to look at this journey that was as laborious as it was fruitful.

Thom S. Rainer, Ph.D., is president of Church Central Associates LLC and founding dean of the Billy Graham School of Missions, Evangelism and Church Growth at Southern Baptist Theological Seminary. Considered one of the leading experts on the church today, he is the author or co-author of 16 books on the church and has written articles or reviews for more than 30 publications. Dr. Rainer has also served as a pastor and interim pastor in 10 churches and has consulted with more than 300 churches, denominational entities and religious organizations.

Dr. Rainer's book "Breakout Churches" (Zondervan) is available on ChurchCentral.com and at most Christian retailers.