This study was commissioned by the Family Ministries Department of the Seventh‐day Adventist Church in North America. The director is Dr. Willie Oliver. The research was implemented by the Center for Creative Ministry, one of the resource centers affiliated with the North American Division of the Adventist Church. The team for this project included Monte Sahlin, primary investigator and analyst; Paul Richardson, executive director of the center; Petr Cincala, data processing and statistics; and research assistants Norma Sahlin, Elizabeth Salisbury, Jeannie Hartwell, Carol Spence, Robert Seal and Melissa Sahlin.
This report is based on a total of 1,397 questionnaires returned from a random, double‐ blind sample of families in the Adventist Church in the United States, Canada and Bermuda. This sample was obtained with the collaboration of 267 pastors, who were selected through a stratified, rolling random sample of telephone interviews and agreed to distribute questionnaire packets to heads of households in their congregations. Email contact was made to get a count of the number of packets not distributed. The response rate was 39 percent from the distributed packets, well within accepted standards for a probability survey sample. The stratified sampling method assured that congregations from every one of the 58 local conferences in the North American Division of the Seventh‐day Adventist Church were included in this study.
The double‐blind method means that the research team had no way of knowing which individuals or families were handed a survey packet, and the pastors had no way of knowing which families actually responded nor any opportunity to review the completed questionnaires. This method was used to assure confidentiality in a study that touches on many sensitive topics. The pastors were instructed to use the Nth name select method with a list of their members to randomly distribute the questionnaires within their congregations. They did so almost entirely among active members, so this study generally excludes those members who have stopped attending church.
A standard allowance for sampling error must be made in all probability surveys. At the 95th percentile of reliability the allowance for sampling error for this particular sample size is three percentage points, plus or minus. See the technical appendix for a more detailed discussion of this topic.
Comparisons are made throughout this report from a similar survey conducted in 1993‐ 94 and a similar study conducted by Charles Crider and Robert Kistler of Andrews University in 1974‐75. Published results of these studies are included in the bibliography.