CHAPTER 4: METHODS
Aim
The purpose of this study is to better understand the teachings and impacts of Evangelical purity culture on the millennial generation, who were coming of age during the height of the 1990s purity movement. While we understand in part some of the harms that this has done to LGBTQ individuals, there is not much documented research on how purity culture has impacted a broader swath of former Evangelical Christians who have left their faith in the years since the purity culture movement peaked. Some of these people, like me, were expelled and forced to leave their faith group. Others could not withstand the strict rules and their repressed sexuality and left on their own. It is my intention that the results of this study shed light on the impacts and teachings of purity culture on those who have left their faith and faith communities after growing up in this culture.
Design
Qualitative studies are appropriate when the research’s primary purpose is to explain experiences and phenomena by relying on participant experiences as the source. Instead of collecting numerical data points to compare treatments and information, qualitative research helps generate the basis of information necessary for further research exploration (Tenny et al., 2018). There is little academic information compiled on the how the rigid evangelical teachings known as the purity movement of the 1990s impacted individuals then and later on in life. Seeing the harm that has been done to them with similar teachings makes it necessary to explore the impacts of this culture on other populations.
Grounded theory is the discovery of theory from data; it is a flexible methodology that works well when little is known or documented about the phenomenon in question (Chun Tie et al., 2019). This research’s goal was to construct the phenomenon of each participant’s experience and to make the abstract thoughts, feelings, emotions, and experiences concrete through building a theory or a conceptual framework based on the interpretations of the collective shared experiences of all or many participants. The researcher used a constructivist grounded theory approach, using the participants’ definitions and impressions to co-construct meaning in the study’s results.
A keystone in grounded theory is concurrent data generation and analysis. In grounded theory, researchers utilize small, memorialized documents (memos) to store the ideas generated through collecting and interacting with data. The interviewer/researcher in this study kept continual notes while transcribing and rereading the transcriptions for accuracy, writing small memos that made notes that eventually became the foundation for some of the themes that came out in the coding process (Chun Tie et al., 2019).
In grounded theory, researchers begin with an open mind. They engage with the process of initial coding, in which they started the process of reading and rereading the transcripts/data in order to compare each account and look for similarities and differences in beginning patterns of data. After engaging in initial coding and pulling out trends, similarities, and differences, the researcher engaged in intermediate coding, and then advanced coding. Grounded theory is a method that is best used when there is no hypothesis to test; it is an exploratory way to approach research (Grossoehme, 2014). Furthermore, rereading the responses upbrings new nuances and enrich the findings. By engaging in several rounds of coding and grouping constructs, researchers who work using grounded theory methodology are able to meticulously pull together broader themes while also seeing the nuance in each theme (Chun Tie et al., 2019).
Sample and Recruitment Procedures
Using purposive snowball sampling, I recruited a pool of 178 potential participants from x-Evangelical Twitter accounts, ex-Evangelical Instagram pages, and ex-Evangelical discord server. I did this by sending the administrator for each account a Google forms link with my recruitment script (see Appendix B), which outlined the study’s purpose and the inclusion criteria of the participants I was seeking. The twitter accounts and ex-Evangelical Instagram pages both shared my Google form and recruitment script, snowballing my potential sample pool and recruiting 178 potential participants. 178 potential participants submitted their email addresses via an online Google form. Gradually, over the course of several days, I sent outreach emails with the recruitment script and consent form (Appendix C) in clusters of 10 and waited for responses. As I waited for responses from each cluster of emails I sent, I eventually came to reach out to 72 out of 178 potential respondents. Of the 72 I reached out to, 12 participants responded and scheduled an interview. Participants were picked by the order they responded to the request, with the first respondents being recruited first and so on. All interviews were scheduled via email. Two participants ended up cancelling their interview after they were scheduled, changing the final sample of confirmed interviews to 10.
The inclusion criteria for this study were that participants be adults (18 years or older), of the millennial generation (between the age of 28-42), formerly Evangelical, were willing to talk about their experience in purity culture, and have left (or forced out of) their faith community. In this study, I interviewed 10 such participants who self-identified as ex-Evangelical Christians.
Recruited from ex-Evangelical Discord servers and social media groups, all participants are former members of predominately white Evangelical churches. The National Association of Evangelicals (2023) defines evangelism as:
Evangelicals take the Bible seriously and believe in Jesus Christ as Savior and Lord. The term “evangelical” comes from the Greek word euangelion, meaning “the good news” or the “gospel.” Thus, the evangelical faith focuses on the “good news” of salvation brought to sinners by Jesus Christ.
Bebbington (2019) provided four identifying distinctives to help separate evangelicals from other Christian denominations. These are (1) Conversionism or the belief that lives need to be transformed through a “born-again” experience and a lifelong process of following Jesus; (2) Biblicism or a full respect and obedience to the Bible as the literal word of God; (3). Activism or the expressing the gospel through missionary and social reform efforts; and (4) Crucicentrism or an emphasis the sacrifice of Jesus Christ on the cross as making possible the redemption of humanity.
Evangelicals are a vibrant and diverse group, including believers found in many churches, denominations and nations. The Pew Research Center (2014) has narrowed down Evangelical Protestant denominations to this cluster of Christian faiths to include denominations such as the Baptist Evangelical Tryad, Southern Baptist, Assemblies of God, nondenominational churches, the Holiness Family, the Adventist family, and many more.
All participants are part of the millennial generation, between the age of 28-42. Out of the 178 participants who responded to the recruitment materials, 10 participants were interviewed; the primary investigator reached out to potential interviewees on a first-come, first-serve basis. All 10 of the participants cited in this study were raised in white conservative Evangelicalism; many of the participants’ families have been steeped in this culture for generations. Three of these participants use he/him pronouns and are cis men, one uses he/him and they/them pronouns interchangeably, and the remaining six are all cis women who identify themselves using she/her pronouns. Four participants are a part of the LGBTQIA+ community.
All of the participants are white and were raised in culturally white American families. This is by chance as the recruitment was open to all people and was advertised in numerous different social media groups and pages. This study focuses on the dynamics and potential harm done by the white Evangelical church and how it impacts its congregants.
Data Collection
The data collection period lasted two months. Each interview lasted between 1.5 and 2 hours. Participants were interviewed via the Zoom platform as they were recruited online from across the United States. Participants were from several regions and states, including Missouri, Oklahoma, South Carolina, Texas, California, Tennessee, and Michigan.The interview format was semi-formal and conversational, structured around the questions presented in Appendix A and participants were urged to ask questions as the interview progressed. The qualitative interviews were meant to reflect on participants’ past experiences as they had already left the faith communities that enforced and taught purity culture behind; therefore, there was no need for additional follow up interviews or outreach to participants after the interview was concluded.
I asked 10 base questions, divided by the main themes that I believed purity culture touched on from the literature reviewed above: relationships, sexuality, and gender. I attempted to focus on gathering information about what messaging the fundamentalist Evangelical church taught, how that messaging impacted the participants at the time, and how participants see it impacting them today. I chose to shape the sub-questions in this way in order to see what participants remembered was taught, how they internalized the teachings, and if the teachings created a lasting impact that participants could identify. I began each interview by asking basic demographic and descriptive information, such as age, race, and what gender pronouns the participants use to identify themselves. Of the participants interviewed, six identified as women, one person identified as nonbinary, and three identified as men.
The interviews began with open-ended questions about the participants’ initial engagement with the Evangelical church. As the interview progressed, more exact questions followed on the topics of purity culture, gender, sexuality, and relationships in the purity movement. The interviews were completed in single sessions of 90 to 120 minutes. One of the interviews were disconnected due to technical difficulties, likely due to a weak interconnection on either the participants’ side or my side. In this case, the video call was reconnected within minutes and data collection resumed from the point of disruption.
Administrative Arrangements
Because this study was conducted online and marketed through a recruitment form that was shared via social media, administrative arrangements were minimal. The recruitment script included on the Google form, which only asked for a participant email address, is in Appendix B.
Protection of Human Subjects
Before initiating a human subjects review process, I completed the CITI program course on Social and Behavioral Responsible Conduct of Research. I applied for and was granted IRB review exemption on February 3, 2023, IRB Protocol #852564. Before conducting each interview, I sent the consent form (Appendix C) by email to each participant. The form was used to confirm that each participant was 18 years of age (though the inclusion criteria of my study ensured this).
Immediately before conducting each interview I confirmed that the participant received and understood the consent form and that they consented to the interview and that the interview be recorded for coding and analysis. All ten participants agreed to the interview and consented to these terms.
This study, by design, allowed for sufficient protection of participants. Participants self-selected and opted-in to this study. Before each interview, I affirmed participants’ right to leave during the interview should they become uncomfortable and in such a case I would delete their data and not include their information in the study. Physical, psychological, social, legal, and other risks were minimal due to this study being entirely voluntary and online. Participants agreed to being interviewed in one 90-120 minute session, preventing repeated disclosure of any feelings or memories that feel challenging, mitigating psychological impact that could incur. Risks to confidentiality were minimal and all data was protected and unsured. All names, proper nouns, references to places, and other identifying information were replaced with pseudonyms throughout the transcription process.
As referenced in Appendix C, there was no benefit to the participant aside from feeling pleased that they can contribute to an academic understanding of purity culture and fundamentalism. Participation in this study helped me better understand how purity culture impacted others, which may benefit participants indirectly as this research will hopefully serve in helping clinicians understand this type of religious trauma. Though I assumed that participants would experience no negative psychological symptoms from participating in this study, participants were encouraged to reach out to a clinician to address negative symptoms.
Process for Capturing and Storing Data
I conducted video interviews via Zoom to facilitate a nationwide reach of participants. After finalizing my instruments and consent form, I reached out to a colleague whom I had a relationship with prior to this research who also fit the criterion for this study. I conducted a pilot interview via Zoom with this colleague to test my process using the Zoom format. This first interview took longer as the participant agreed to provide input to make small changes to the wording of my instruments in order to improve clarity. After the pilot interview I evaluated the feedback and made some changes to my wording before moving on to interview study participants.
To set up each interview, I emailed each participant directly with the recruitment script reiterating the purpose of the study and attached the consent form. Before engaging in the interview, I confirmed with the participant that they had read the consent form and confirmed their eligibility according to the inclusion criteria, their knowledge of their rights during and after the interview, and their consent to record the interview for transcription and analysis. I then interviewed each participant via Zoom while using the platform’s record function. At the conclusion of the interview, I thanked each participant profusely, asked if they had any questions about the process, and encouraged the participant to engage in some sort of self-care activity of their choice to celebrate their participation in this study.
The interviews were scheduled over the course of a two week period and were conducted over the course of a seven week period, averaging approximately 2-3 interviews per week. As stated above, 12 initial participants agreed to interviews, but two of those initial 12 cancelled prior to their interview dates. I conducted the 10 remaining interviews with the participant pool that volunteered.
Data Analysis
I transcribed the Zoom recordings as soon as possible after the conclusion of each interview. I used the AI-driven digital transcription software included in the Zoom recording in order to get an initial draft of the words spoken in the interview. I copied that digital transcription off of the Zoom platform into Microsoft Word and read through them while listening to the audio of the Zoom recordings to ensure accuracy. This served as a first read-through of the interviews, giving me an opportunity to take some initial notes on my impressions of the data. As I transcribed, I assigned pseydonyms to each participant.
After transcribing the interviews and changing identifying information, I began to harvest the answers to the questions I asked in my instruments. I did this by creating a Microsoft Excel document with a question (and sub question) on a different sheet of the document. I copied participants’ demographic information and the answers they gave to the interview questions in each sheet, differentiating answers between participants by color coding them according to each different pseudonym. Coding of transcripts was completed in the order of the interviews. I listened to each interview multiple times, allowing me time to reflect on and pick out emerging themes that began to show up in the data.
My instruments provided a starting point to begin grouping and generating initial codes. This process helped me frame the data within the aims of this study. I then read the interviews several times, pulling out additional themes, which became codes. I began with the process of initial coding, in which they started the process of reading and rereading the transcripts/data in order to compare each account and look for similarities and differences in beginning patterns of data. After engaging in initial coding and pulling out trends, similarities, and differences, I engaged in intermediate coding, in which I sorted initial codes into relevant categories, refining the analysis into broader themes and constructs. Finally, I entered the advanced coding phase of grounded theory, in which they organized the constructs and themes into categories that represent overall conceptual terms in the data (Chun Tie et al., 2019). Doing this allowed me to be able to integrate theory and connect research findings to potential clinical recommendations. Throughout this process, I scrutinized each individual interview, combining and refining codes that felt redundant and omitting codes that seemed to fall outside the scope of this study.
Through this process, I ended up creating a document in Microsoft Excel that separated each participants’ answers by pseudonym and question. I sifted through the data in those Excel sheets and sorted them into the codes. This then led to the creation of a code book in my Microsoft excel document, color coded by pseudonym and grouped into different sheets according to the codes that emerged rather than the initial draft of the excel document, in which each sheet sorted the answers by question. During this process, I wrote notes in a bucket (text document) to record ideas and explore larger these that would eventually help create the Findings of this study.
After finishing the analysis and submitting the final draft of my dissertation to my committee, I deleted the transcripts of all of my interviews.