Stakeholders in Adventist education―teachers, administrators, church leaders, pastors, parents, alumni, and parishioners―have noted with growing concern the gradual decline of enrollment in K-12 Adventist schools in the North American Division (NAD) over the past few decades (see Figure 1). In 2014, the highest level of administration within the North American Division (NAD) charged a select group―the NAD Education Taskforce (NADET)--with assessing the state of the division’s K-12 educational system and making recommendations that could strengthen and improve the schools. The members of the NADET spent hundreds of hours in videoconferences and focus groups examining issues and seeking out solutions that could possibly plug the holes in the proverbial “sinking ship.” There must be a cause, a reason for low enrollment, they reasoned, and there seemed to be no better place to point that finger than at things the schools supposedly lacked—quality, or innovation, or . . . something.

These conversations have been mirrored at the local level as well. As a former teacher and principal in the Adventist school system, I’ve sat through many board meetings in which my school, my staff, and/or my curriculum were all under intense scrutiny, and our efficacy was called into question. Concerned parents and church members were always quick to reference the other private schools in the area, encouraging us to mimic their style or type of education in order to draw more students to our campus. I recall one particularly enthusiastic parent sitting down in my office to detail his plan of starting a Christian drama program through our school and the throngs of new families who would allegedly flock to our campus as a result.

These well-meaning suggestions were not necessarily wrong. As with any system or institution, there will always be faults or areas for improvement. However, I began to wonder if blame was being cast in the wrong direction. What if the quality of our schools has not changed? What if it is the characteristics of the churchgoing member that have changed? What if Adventist education is simply not a priority within our denominational culture anymore?

In that same aforementioned principal’s office in which I used to work, I also would often be regaled with tales from constituents about their parents or grandparents who had worked three jobs to ensure that their children could receive an Adventist education. I heard story after story about the heroic, herculean efforts by these staunch church members to keep their child attending an Adventist school, including moving their family to be physically closer to a church school or, in one case, literally building a new school, brick by brick, on the local church property with their personal funds.

Based on numerous anecdotes similar to these, it can be surmised that a generation ago, being a solid, committed member of the Seventh-day Adventist Church in North America meant that you always sent your child to an Adventist school regardless of circumstances or educational needs or even desire. The church identity of parents used to, it seems, extend into their choice of school for their child. This seems to contrast starkly with the current reality in North America.

What if that gauge is different now? What if the measure of a solid, committed member of the Adventist Church no longer involves enrolling one’s child in a denominational school? What if both parishioners and church leaders today feel they can still be engaged and involved and participatory in their faith community even if their children attend a non-Adventist school? Viewed in this light and in juxtaposition of the results of this study,1 the general enrollment in NAD K-12 Adventist schools may not reflect the quality of their academic offerings, but rather the church members’ paradigm of denominational identity and commitment.

And what exactly is that paradigm of denominational identity and commitment? How does one discuss, much less quantify, that feeling or understanding that comes from membership in any given community? There is a significant divide, I believe, between doctrine and culture. A church’s doctrines serve as its backbone, its guiding light, and are generally held in profound regard. The Seventh-day Adventist Church has clearly defined 28 Fundamental Beliefs that it believes are core to the present truth that it seeks, which can be categorized into six different themes: God, humankind, salvation, the church, the Christian life, and last-day events (see These doctrines lay a foundation for the parishioner’s worldview, clarify a theology of faith, and answer essential questions such as “Who is God?” and “Where do I go when I die?”

This, though, is not what seems to be at the crux of conversations about school choice. In the NAD, at least, choosing to enroll one’s child in an Adventist school does not appear to hinge on beliefs about the existence of God or the nature of humankind. Rather, I began to wonder if school choice might instead be a reflection of culture―and, more specifically, Seventh-day Adventist culture.3

Adventist Culture

But what is this Adventist culture? Individuals who have been a part of the Seventh-day Adventist Church for a number of years, who were “born and raised in the church” and grown up immersed in this community of believers sometimes appear to be privy to an inner circle that outsiders may find daunting or overwhelming. They casually toss out Adventist jargon―terms that for the general public require definition and explanation―and embrace lifestyle choices and habits that aren’t identified in any church manual. For instance, take the following descriptive paragraph:

During my years at boarding academy, we were required to stay in our church clothes until potluck. In the afternoon, we would go colporteuring or Ingathering   and then come back to the dorm for Pathfinders. Then, after sundown vespers, the game of choice was often Rook―and always, always accompanied by haystacks for dinner.

If non-Adventist individuals were to read that, how many words or phrases might cause them to raise an eyebrow or ask for clarification? Are church clothes a type of uniform? Pathfinders―clearly the speaker doesn’t mean a model of car? What on earth are haystacks? And yet most members of the Adventist Church, at least in the NAD, would not only understand all that was being said, but perhaps might also smile at their own memories of some activities that were mentioned.

Ruminating over this concept of Adventist culture, I began to talk with my friends, family, and colleagues. I asked them about their own cultural context―we compared notes and laughed at the similarities. References spanned meat substitutes, summer-camp experiences, and Sabbath afternoon activities. In these informal conversations, there appeared to be a general consensus that these cultural norms, many of which were lifestyle related and regional, were both (1) commonly shared and (2) hard to explain to someone outside the Adventist Church. There seemed to exist a shorthand, an understanding that belied these exchanges that spoke of a connection difficult to explain, but easy to identify.

But this was the very thing that I was suggesting could be a factor in parents’ choice of school for their children. In my experience as an elementary school principal at an Adventist school, there was something other than doctrine at play in these decisions. And so, I hypothesized, if there was a way to bring this idea of culture into the light and give it a formal place at the table, then perhaps a different perspective could be provided about current trends in Adventist education in North America.

Cultural Consensus Analysis

The first and most obvious problem with any attempt to employ Adventist culture as a variable in a research study is that allusions to it are purely anecdotal in nature. Shared stories, knowing winks, affirming nods―all certainly indicators, but not necessarily empirical evidence of this so-called culture.

In 1986, Romney, Weller, and Batchelder introduced cultural consensus analysis (CCA) by using the theory of culture as an aggregate construct.4 CCA restated the basic premise that individuals behave in certain and specific ways based on their understanding of shared behavioral and social norms within that certain and specific culture. For example, Americans behave differently at baseball games than they do in board meetings because they anticipate and expect different social and behavioral protocols at different venues. The culture informs their behavior, both of which are rooted in a shared understanding of that particular environment.

By assuming a fixed knowledge base or information about a proposed experience, CCA first identifies agreement within this experience from key participants. Researchers ask a sample of the population to list and then rank items that are salient to a specific culture. Agreement among the respondents serves to validate the cultural domain and then construct a cultural model. For instance, one study asked Brazilians to define a successful lifestyle.5 Responses from the sample produced a list of 25 items that were identified by at least 10 percent of the sample. The respondents were then asked to rank the items, which produced another list with the average assigned rank.

By comparing the responses of the individuals, researchers could identify which respondents had higher correlations or, to put it another way, agreed more with one another. In CCA, those respondents are considered more “culturally competent”; that is, their knowledge of the cultural domain is greater and more correct than the others. This is an important aspect of CCA, as subsequent calculations will give more weight to those respondents than to others who are not as “culturally competent.” In the example above, there was clear cultural consensus within the domain of what respondents considered to be successful Brazilian lifestyle as evidenced through a high ratio of the first-to-second eigenvalue.6 From this, researchers were able to derive a cultural “key,” or average value, for instance, of owning a DVD player or a refrigerator, as identified by the respondents. This key is crucial as it paves the way for further analysis of the cultural domain and the people who inhabit that domain.

The beauty of CCA is that it provides a valid and tangible way of connecting a collective, shared culture with individual understanding and behavior. Being able to quantify culture thus can provide researchers with newfound freedom to operationalize this construct. 7

Methodology: Research Design

In 2018, I embarked on a study to tie all these pieces together. There have been a few research studies on Christian denominational culture, but the majority of those have been solely qualitative―utilizing focus groups, individual interviews, or small case studies. This study specifically sought a way to turn the vague and elusive concept of culture into a concrete, quantifiable variable. Cultural consensus analysis emerged as the most appropriate method to use for this initial step; therefore, while the preponderance of data were collected quantitatively through the distributed survey and analyzed through various statistical analyses, the first part of the study was wholly qualitative. In order to arrive at a valid measure that could quantify this cultural component of religion, an emic (insider)8 approach was first taken to develop a cultural model based on the responses from the community itself. That domain and derived cultural key were then embedded into the survey instrument and used as a quantitative measure.

Construction of the Cultural Domain

In order to measure culture, one first needs a cultural model. Following the steps outlined for cultural consensus analysis, this domain was constructed in two phases with two different samples, using an emic approach. The first phase utilized the qualitative inventory methods of free listing and rank ordering to help identify salient characteristics of Seventh-day Adventists in the United States. The resulting inventory was then used to create the survey questionnaire that was distributed to a larger population sample.

Free Listing

In qualitative research, free listing helps individuals create categories based on their emic or insider understanding of a given concept.9 A snowball sampling procedure10 was used in the first stage of the study. With snowball sampling, the researcher begins with a small number of participants, but as each respondent refers other potential participants, the number of individuals begins to increase or “snowball." In this study, individuals in the first sample (n = 61) were a network of Seventh-day Adventist friends and colleagues from whom additional names were procured of individuals who are active and involved members of the Seventh-day Adventist Church. Steps were taken to ensure that the sample was geographically representative of the North American Division (NAD) of Seventh-day Adventists by including approximately 7-8 individuals from each of the eight U.S. unions involved in this study. Of the 61 participants, 41 were female and 20 were male; 18 were over 50 and 43 were under 50. Because the data collected was used to assess shared cultural knowledge, the sample did not need to be random.11

The individuals in the first sample were contacted by phone or e-mail and the interviews were conducted either face-to-face or via Skype, Zoom, or telephone. After a brief explanation of the study, each participant was provided with the prompt: “Imagine a traditional Seventh-day Adventist who lives according to the prescribed Adventist culture. What behavior or characteristics would you expect to see in such an individual?”12 Based on that prompt, the respondents were asked to free-list (categorize) all items that came to mind. They were specifically instructed to answer on behalf of their knowledge of the Adventist community and not of themselves personally.13

Each interview was recorded, and a spreadsheet was created, itemizing the responses from each individual. At the conclusion of the interviews, this spreadsheet was examined in its entirety, and a codebook was created from the notes. Similar items were reduced to single statements. For example, one respondent remarked, “Adventists don’t intentionally seek interactions with non-Adventists.” Another stated that Adventists have “a bit of an exclusive mindset and are drawn to people we are similar to. . . .” Phrases like those were merged and coded into “socializes with other Adventists.”

From this first sweep through the respondents’ lists, the codebook consisted of 165 items. By continuing to parse and combine, the list was further reduced to 45 traits or characteristics of a traditional, upstanding American Adventist (see Table 1). New columns in the spreadsheet were created for each respondent, with his or her corresponding edited list that used the codebook terms. Twenty-seven of the most salient items were chosen for the rank-ordering task that followed.


Once this list was created, a second sample was chosen (n = 63).  Once again, a snowball sampling procedure was employed, using “referrals” from Adventist friends, colleagues, and acquaintances across the country (n = 63). All participants were self-described as active and involved church members and were also parents of K-12 school-aged children. This sample also provided a fair representation of the eight U.S. unions studied, with roughly seven to eight participants from each region. Of the 63 individuals in this sample, 44 were female and 48 were under 50 years old. For this phase of the construction of the cultural consensus domain, the participants were given the task of rank-ordering the list of items derived from the first sample. The purpose of this second step was to assess the degree of agreement―or consensus―among these items, which had been identified as being key elements in the culture of Adventism in the first phase.

As with the first sample, each participant in the second sample was first contacted either through phone or e-mail. At the appointed meeting time, a brief summary about the study was provided to the participants as well as an explanation as to how these 27 items were identified. The participants were told exactly what the prompt had been for the first sample; in other words, what those participants had been responding to and how this list had been developed. They were then instructed about the task before them―to rank-order all 27 items, beginning with what would be most important to a traditional Seventh-day Adventist in good standing.

For those with whom this task was conducted in person, Dengah’s approach to rank-ordering was used.14 Those participants were given 27 small cards―each card had one of the items written on it. Respondents were encouraged to first sort the cards into three categories―very important, somewhat important, and not at all important. Once they had three piles, the respondents were asked to order them within each of the categories. When they were finished, the result was a complete rank-ordered list of all 27 items. Some respondents chose to complete the task as described above; others simply rearranged their cards from left to right and ordered them from 1 to 27.

Like the first sample, all respondents in the second sample were instructed to rank-order the statements according to how the community perceived their importance, not how they would prioritize them personally.

Establishing a Domain

Using the ordered lists from each respondent in the second sample, a correlation matrix was created of respondents and their ranking of each item. The degree to which respondents agreed with one another was quantified as a cultural competence coefficient; essentially, it determined how well each individual understood the culture. Those who ranked items similarly to most others had a high coefficient and were said to have a high degree of cultural competence.15 This is a consensus model, which means that “competence” is not defined as correct answers, but rather the level of agreement and shared knowledge among respondents.

Using those cultural competence coefficients, a factor analysis was run on the items, the respondents, and their ratings, and examined for the ratio of the first eigenvalue to that of the second. A cultural domain is established based on an examination of the ratio of the first and second eigenvalues. The first factor denotes the largest shared intersection among a set of variables (as composed by the free lists), and the second factor accounts for the residual agreement.16 Cultural consensus theory maintains that if the ratio between the first and second eigenvalues is higher than three, it can be inferred that the sampled population is referencing and utilizing the same shared knowledge and that there indeed exists a cultural domain.

The results of the factor analysis of the respondents (n = 62) produced a ratio of 3.28 between the first eigenvalue (19.357) and the second eigenvalue (5.901). While it is a modest ratio, it still indicated there existed a shared set of cultural knowledge within the population of Seventh-day Adventist Church members in the United States.

By next calculating the rankings of all the items based on the average of all the respondents’ rankings, while also giving more weight to those respondents with higher cultural competence coefficients, a “cultural key” was identified, providing a touch point from which the rest of the study could proceed.

Survey Design and Distribution

With the cultural key in hand, I was able to move forward with developing the final survey instrument. Following the factor analysis, which ordered the items starting with the most salient trait/behavior/characteristic of a traditional Seventh-day Adventist and ending with the least salient, I took the top 14 items and turned them into survey questions/statements. Respondents were given the option to answer Strongly Agree, Agree, Disagree, or Strongly Disagree. I only included 14 items in order to be sensitive about the length of the survey. This culling or selecting of items is not an unusual approach for in the cultural consensus model and has been employed in numerous other studies.17

In addition to the measurement of culture, I included two other components: general religiosity and Adventist doctrine. The Duke Religion Index (DUREL) as developed by Koenig, Meador, and Parkerson18 was used to measure the respondents’ general religiosity over three dimensions: organizational religious activity, non-organizational religious activity, and intrinsic religiosity,19 and the answers to these five questions were averaged to create the religiosity variable.

Commitment to and belief in the doctrines of the Seventh-day Adventist Church were measured using another short, five-question instrument that was previously used to study religiosity and public issues among Seventh-day Adventists by Dudley, Hernandez, and Terian.20 The answers to these five questions were averaged to create the variable for doctrine.

I distributed the final survey instrument in the summer of 2018 through a variety of channels across the United States, targeting Seventh-day Adventist Church members in the United States who had K-12 school-aged children. The North American Division, which includes the United States, Guam, Bermuda, and Canada, is subdivided into nine unions and one mission, which are further divided into 59 conferences. Because of the significant cultural differences found in Guam, Bermuda, and Canada, these areas were omitted from this study. When the survey closed, more than 1,000 responses had been submitted and of this number, 991 entries were deemed viable and of this number, 830 were used in the subsequent analyses.

Figure 2 provides a representation of school choice among the respondents from eight U.S. unions represented in this study. In general, it appeared that most of the survey respondents sent their children to a K-12 Adventist school. Both the North Pacific and Pacific unions had fairly high percentages of respondents who chose Adventist schools for their children. At 17.2 percent, though, North Pacific Union had a higher percentage of children who were homeschooled than the Pacific Union, with 10.3 percent. The Pacific Union (26.2 percent) also had one of the higher rates of children enrolled in non-Adventist schools, along with Lake Union (32.4 percent).


Part 1 of this article attempted to articulate the purpose for this study and provided an explanation of its methodology. The intersection of school choice and Adventist culture has indeed provoked a fascinating thought process in regard to our church, our heritage, and our community within the United States, and demands closer examination for any significant ramifications on the decisions we make for ourselves and for our families.

Part 2 of this article will appear in the October-December 2020 issue of The Journal of Adventist Education® and will explore in more detail the key findings of the study as well as implications for the future.

This article has been peer reviewed.

Aimee Leukert

Aimee Leukert, PhD, is Associate Professor of Curriculum and Instruction at La Sierra University, Riverside, California, U.S.A. An educator in the Adventist school system for more than 15 years, Dr. Leukert has taught at the elementary, academy, and university levels, and has also served as a principal in the Southern California Conference. Her work at the Center for Research on Adventist Education has allowed her to share her passion for teaching through different channels, including assisting the NAD Education Taskforce and launching Ambassadors for Adventist Christian Education (AACE), an organization designed to recruit, organize, and support volunteers for K-12 schools across the NAD.

Recommended citation:

Aimee Leukert, “Adventist Culture, Adventist Choices: The Relationship Between Swimming on Sabbath and Adventist Education, Part 1,” The Journal of Adventist Education ® 82:2 (July-September 2020).


  1. A detailed description of the study begins on page 30 of this article.
  2. General Conference of Seventh-day Adventists, “28 Fundamental Beliefs,” (n.d.):
  3. Although there may be shared characteristics and beliefs held by Seventh-day Adventists, there is no one, single definition of Adventist culture, since social and behavioral norms can vary depending on the country and, more specifically, the ethnic and cultural backgrounds of church members within a given country or region. This study specifically sought a way to turn the vague and elusive concept of culture into a concrete, quantifiable variable through the method of Cultural Consensus Analysis (CCA).
  4. William W. Dressler, Culture and the Individual: Theory and Method of Cultural Consonance (New York: Routledge, 2018).
  5. Ibid.
  6. Ibid.
  7. Ibid.
  8. In anthropological studies, emic is defined as an internal perspective. For example, members of a specific group being studied will serve as the primary sources of information about the group. It is often referred to as the “insider” approach. For more information see “Two Views of Culture: Etic & Emic” (n.d.):
  9. For more on free listing, see Marsha B. Quinlan, “The Freelisting Method” (August 2017):; and Beverly Peters, “Qualitative Methods in Monitoring and Evaluation: Free Lists” (n.d.):
  10. Statistics How To (2020):
  11. W. Penn W. Handwerker, Quick Ethnography: A Guide to Rapid Multi-Method Research (Lanham, Md.: AltaMira Press, 2001).
  12. For more in-depth discussion about how the author collected data for this study, see Aimee Leukert, Choosing God, Choosing Schools: A Study of the Relationship Between Parental Religiosity and School Choice. PhD diss., Claremont Graduate University, 2018, 68-82. Available at
  13. Dressler, Culture and the Individual: Theory and Method of Cultural Consonance.
  14. H. J. François Dengah II, “The Contract With God: Patterns of Cultural Consensus Across Two Brazilian Religious Communities,” Journal of Anthropological Research 69:3 (Fall 2013): 347-372.
  15. Dressler, Culture and the Individual: Theory and Method of Cultural Consonance.
  16. W. Penn Handwerker, “The Construct Validity of Cultures: Cultural Diversity, Culture Theory, and a Method for Ethnography,” American Anthropologist 104:1 (March 2002): 106-122.
  17. William W. Dressler, personal communication, 2018.
  18. Harold G. Koenig, G. R. Parkerson Jr., and K. G. Meador, Duke Religion Index (DUREL, DRI) (1997): APA PsycTests:
  19. Permission was granted for the use of the Duke Religion Index scale. The Duke Religion Index (DRI or DUREL) defines three key dimensions of religiosity: organizational religious activity (ORA), non-organizational religious activity (NORA), and intrinsic religiosity (IR). ORA includes religious activities that are public, such as attending a religious service, prayer group, Bible-study group, or small-group meeting. NORAs are private religious activities, such as such as listening to/watching religious programs on a radio or television, private prayer, study of Scripture, and other religions activities that do not take place in groups. Intrinsic religiosity (IR) is an individual’s personal religious commitment and motivation not influenced by external factors such as social status or being seen. For more, see Harold G. Koenig and Arndt Büssing, “The Duke University Religion Index (DUREL): A Five-Item Measure for Use in Epidemological Studies,” Religions 1:1 (December 2010): 78-85:
  20. Permission was granted for the use of the scale by Roger L. Dudley, Edwin I. Hernandez, and Sara M. K. Terian, “Religiosity and Public Issues Among Seventh-day Adventists,” Review of Religious Research 33:4 (June 1992): 330-348: