We began in the fall of 1996 to solicit nominations of those church youth groups and para-church youth ministries who may have evidenced teen-to-teen evangelism as a normal occurrence for at least three previous years. Our inquiries led us to denominational youth directors, youth ministry educators, writers, and para-church youth ministry leaders. The requests for nominations came via e-mail and phone conversations, seminars and conferences, in group settings as well as individual conversations. After four months of such efforts a total of 109 different youth ministries across the United States had been suggested for our research. This relatively small population represents a limitation of this study which was recognized from these earliest stages.

In the meantime, a plan was agreed upon which would allow us to use the research task as an excuse to become meaningfully involved with a team of undergraduate ministry majors/minors in a shared growth experience. And so, parallel to this activity of seeking youth ministry exemplars, a team of ten students was selected for participation in the project. Grants for undergraduate travel expenses related to this research were supplied by the MacLeod Foundation.

In order to find out whether those youth ministries nominated for further investigation met our criteria for effective student leadership (a stable and immediate history of 3 or more years where peer-to-peer evangelism was a normal occurrence) we distributed these 109 groups among the team for the purpose of conducting a phone interview. Our process was relatively simple after we designed a master interview template. Initial contact was made with youth pastors seeking a phone appointment of no more than 20 minutes where we could ask them questions about their use of student leaders in youth ministry. Nearly one out of every six calls found the youth sponsors to be unresponsive. We reviewed each case separately and, in most cases, dismissed these locations from consideration for our project. In a few instances the contact was passed to another team member for another attempt at contact before such a decision was made.

Upon completing the interview and filling out the three page questionnaire, the team member conducting the interview assigned a "grade" to the ministry. Sites receiving "A's" deserved further serious consideration while those receiving "B's" might be worthy of a closer look. Those ministries which didn't receive either an "A" or "B", had experienced a recent youth minister change, or were unwilling to help were removed from the dwindling list of locations to be visited for our project. One of the important practical considerations of this project involved the limitation of time and money. After careful planning we decided that various team members could visit as many as 20 sites between March and May in 1997. With major consideration balanced between the strength of the location and the effective use of funds, we deliberated which of the 20 sites offered us the highest probability of effective return and attempted to set up team visits with each. Logistical limitations reduced our site visits to a total of 17 in this first year.

When visiting a location, nearly always in teams of two or more, team members employed a combination of three data-gathering devices. It was agreed that these distinct approaches would supply us with data for both quantitative and qualitative analyses, as well as affording the youth ministries multiple avenues to communicate to us in a relatively short period of time.

Forced-choice surveys were designed and administered on location. These instruments asked students to answer eight questions about themselves before asking them to indicate what the relative frequency of their practice was for 24 different evangelism-related behaviors. For example, in response to the stem "I pray for one friend to accept Christ" students could choose responses from one of seven frequency options, ranging from "never happens" to "a few times daily."

We also designed and implemented a small group data gathering tool that we called "timed team responses." The larger group of student leaders gathered for our visit was divided into small groups of equal size and composition with regard to gender and age. Adult volunteers were invited to form their own group and professional youth ministers were excluded from the process. Each team was given the same question and ask to come to consensus and record their top three answers within a strict five minute time period. Our intent was to surface commonly agreed upon important elements in the life of each particular youth ministry in a way which would allow us some control with regard to peer influence during discussion.

Finally, we also employed large group discussions that helped us to discover the expectations and spirit of the groups. Extensive notes were take at each location. Adults were allowed to participate in this discussion but asked to defer to students with regard to how often they contributed comments. Youth ministers were, again, asked to listen only. Before leaving each site we also collected some descriptive data about each youth ministry from the youth minister.

Data from the forced-choice surveys were entered into a computer and analyzed through separate one-way ANOVAs and subsequent Scheffé post-hoc tests. Data from the timed team responses were immediately analyzed for internal (within group) consistency. They were also used with large group discussion data as comparative analyses were made between groups. Frequent and extensive team interaction helped us to verify perceptions and judgments made throughout the research procedure.

Because it became apparent that there were some worthy sites which, for various reasons, could not be personally visited, we developed self-administered packets to pass along to those who might be willing to help us. This was necessarily more limited than our on-site data collection, but it was felt that descriptive data and forced-choice survey responses would, at least, contribute to the quantitative analysis done in this project. Unfortunately, while a number of youth ministers agreed to help us in this regard, they did not follow up and we received no data through this means.

After our visits were concluded, we engaged in analytical work and hypotheses development during the summer of 1997. Eight preliminary conclusions were developed and discussed among the research team for their validity. After concluding that further study was warranted, including some revisits to test our hypotheses and new visits to diversify our samples, a second grant was solicited and approved for another year of research.

In this second year of study our undergraduate research team was pared from ten to four. This decision reflected more limited funds while rewarding those students who had distinguished themselves by "over and above" dedication to the project. This smaller team worked together to develop on-site strategies that would help us to check our hypotheses. In addition to employing the earlier three data gathering strategies, an individual interview template was developed and employed with student leaders at each location. Large group discussion questions were also modified a bit in this second year to assist us in uncovering the truth with regard to our hypotheses. A total of seven site visits were done from October to March in this second year, with two of them being revisits of locations from our first year. In total, 22 different youth ministries participated in our study.