Table of Contents
This chapter details the research design and approach that I used to understand how FL is conceptualized and communicated by teachers and parents in NS school communities. As presented in Chapter Four, this qualitative study was conducted using a critical research paradigm, with tenets of Critical Ethnography. This chapter summarizes the research design employed, including the approaches used to facilitate participant recruitment and my approach to data analysis. This chapter concludes with a section on researcher reflexivity, which is essential for research that is conducted using a critical ethnographic approach.
Research approach
This study was situated within a broader multi-component and phased study; titled Building on successes and learning from challenges: A comprehensive evaluation of the school food and nutrition policy (SFNP) in Nova Scotia with the following research objectives:
- Research Objective 1: Describe the system level dissemination of the SFNP; and
- Research Objective 2: Assess factors influencing the adoption of the SFNP.
This study contributes to Research Objective 2 of the larger study through the following research question:
How is food literacy conceptualized and communicated?
To achieve this, I employed a qualitative inquiry using a case study approach while drawing on tenets of critical ethnography. In the proceeding sections of this chapter, I will summarize the study setting, study participants (recruitment and sample size), and the sources of data.
127 our experts are availableright now
Setting
The setting for this study was in the province of NS, Canada. The province of NS provides a unique case to explore how FL is conceptualized and communicated for two reasons: 1) its rich (but challenging) history of economic growth and development derived from its agriculture, mining, and fishing sectors; and 2) its history of political commitment towards improving children’s health through the provision of substantial and ongoing financial support for school food policies and programs since 2005 (McIsaac et al., 2019). Furthermore, NS was one of the first provinces to position nutritionists in communities and public health units across the province; therefore, there is a long history of addressing nutrition issues through education efforts directed to food and health.
NS is one of the four provinces that forms Atlantic Canada; it is the second smallest of Canada’s ten provinces and the most densely populated province in Atlantic Canada with a population of nearly one million residents. NS is surrounded by the ocean and is comprised of freshwater lakes which makes it an excellent habitat for fishers. Fishing in NS dates back many centuries and has been the livelihood for many families in their communities. Lobster is the most valuable seafood export for NS as more than half of the lobsters fished are exported (Province of NS, n.d.).
Agriculture is another important sector and began in NS over 400 years ago. Since that time, “farmers have been improving agricultural practices by testing crops and livestock to find the best suited to the climate, soils, and management practices of the province. Agricultural models that work well in western Canada or other places may not work well” (Province of NS, 2012, p. 3) for NS. |The largest agricultural sector in NS “in terms of number of farms is fruit farming, consisting mostly of blueberry and apple operations. Nova Scotia fruit farms represent a healthy 12% of the total number of fruit farms in Canada” (Province of NS, 2012, p. 9).
Despite the strong economic growth related to agriculture and fisheries, NS has some of the highest disease burdens across the country and is further afflicted with one of the highest rates of food insecurity in Canada (as previously described in Chapter Two). Given the complexity of the above stated factors, the province of NS is an excellent reference for a case study to explore how FL is conceptualized and communicated.
Participants
Participants who are familiar with the area of inquiry and who are willing to impart information are preferred in ethnographical studies (Hammersley & Atkinson, 1983). For the purpose of this research, I employed criterion sampling, purposive and/or opportunistic sampling. Criterion sampling involves establishing criteria for studying select individuals (Creswell, 2013). This strategy is typically applied when considering quality assurance issues (Creswell, 2013). Purposive sampling aims to select participants based on their particular relationship to the area of inquiry (Creswell, 2013).
This strategy allowed for the exploration of FL in the daily experiences of school communities and how their experiences are connected to and shaped by the broader environment. Opportunistic sampling is flexible and takes advantage of the unexpected, such as new leads during fieldwork (Creswell, 2013). This strategy allowed the sample to evolve on its own, which is a characteristic of carrying out an ethnographical study (Teddlie & Yu, 2007). Therefore, this approach allowed me to remain open to inviting a group or individual I hadn’t considered to participate if they met the inclusion criteria.
Inclusion Criteria
Teachers and parents were invited to participate in this study in order to gather different perspectives on how FL programs, practices and initiatives are integrated into the school community based on the following criteria:
- someone who has undergone, or who is undergoing, the experience of school food and nutrition policy, programs, and initiatives inside and outside schools;
- someone who is able to reflect and provide detailed experiential information about the phenomenon in relation to the school food environment;
- someone who is willing to critically examine and self-reflect on the experience (in relation to SFNP, programs and initiatives) and his/her response to the situation (school food environment); and
- someone who is able to participate in a lengthy (30-60 minute) interview process that will take considerable uninterrupted time (Morse, 1991).
For the purpose of the school communities, the following criteria were applied: - Diversity of the nature of the school community (based on level of school (elementary, junior and high school), and the circumstances surrounding school community such as divergent political, social and economic context), and
- Geographical dispersion of school communities (catchment was determined by teachers and parents who expressed interest in participating in the study).
Participant Recruitment
Ethnographic studies typically start with key informants who may act as gatekeepers, thus enabling the researcher to initiate contact with potential participants (Creswell, 2013; Higginbottom, Pillay, & Boadu, 2013).
Prior to recruiting participants, I worked as the Coordinator of School Food and Nutrition with the NS Department of Health and Wellness and had established networks with the school boards. I had intended to purposefully select the gatekeepers from my established networks to get into the school communities; however, due to a labour dispute, I was not able to attend to this approach. Instead, participant recruitment occurred through an iterative process by way of social media recruitment; this was adjusted based on feedback from my supervisor and committee members, personal reflection as well as information I received from participants themselves.
Once I received ethics approval for this study in April 2017, participant recruitment commenced. My primary recruitment tool for this study was a recruitment poster (Appendix B), which was shared via social media (Facebook and Twitter). This approach to recruitment was critical to the success of participant enrolment since I gained most of the interest, and subsequent enrolment, in my research study through this method. The secondary approach to participant recruitment was word of mouth and direct email to colleagues asking them to share the poster with their networks.
Sample Size
Qualitative sample sizes should acquire sufficient data to address the research question and describe the phenomenon of interest. Sample size often depends on the research questions, data collected, data analysis, and the availability of resources; human and financial (Merriam, 2009). For the purpose of an ethnographic education study, sample size ranges from 6-33 participants and for the purpose of health science case study research, a sample size ranges from 2-420 participants (Guetterman, 2015). Another approach to determining sample size is to investigate until no additional data or further themes are found; known as data saturation (Glaser & Strauss, 1967). Some studies have reported data saturation after as few as 6 interviews (Fugard & Potts, 2015).
Furthermore, sample size for thematic analysis (discussed later) ranges between 2-400 participants (Guetterman, 2015); determined by the type of data collection and the size of the project. For small projects, 6–10 participants are recommended for interviews, 2–4 for focus groups, 10–50 for participant-generated text, and 10–100 for secondary sources while 400+ participants are recommended for large projects (Braun & Clarke, 2013).
Furthermore, a recent methodological study found that code saturation (the stage at which no additional thematic issues are identified and codebook stabilizes) was realized at 9 interviews; but rather 16 to 24 interviews were required to reach meaning saturation (the point in time where one has a comprehensive understanding of the issues raised in the data) (Hennink, Kaiser, & Marconi, 2017). Taking this into consideration, it seems there are no agreed upon guidelines to determine sample size for qualitative inquiry. For that reason, I originally proposed to recruit between 20-30 participants in this study in order to obtain an information rich case study aligning with my research aim to explore the meaning of FL and attain depth in my findings related to socio-cultural context.
More specifically, a variety of reasons were explored for proposing this range. First, I understood that an ethnographic education study sample size ranged from 6-33 participants; second, I recognized that recruiting potential participants to observe may be challenging; third, the interview required an investment of uninterrupted time from busy teacher and parent participants; and finally, I anticipated generating a large amount of data for my analyses due to my two-pronged research approach (testing my conceptual model and exploring socio-cultural context of FL in schools). Given these factors, my rationale for the participant sample size was to ensure data quality, while achieving breadth and depth of the data, in order to provide a rich understanding of FL issues.
I encountered some challenges in recruiting and enrolling participants throughout my data collection process. The recruitment and enrolment of participants during the labour dispute and over the summer months posed scheduling challenges; many interviews had to be rescheduled several times.
Once the recruitment poster was shared with potential participants and they identified their interest to participate by leaving a voice message or emailing me directly about the study, I followed-up to inquire further about their interest in participating and provided them with additional details about the study. Figure 3 provides a timeline that represents the recruitment process. Over the course of four months (April to August), I was contacted by a total of 21 individuals throughout my data collection process; three individuals did not respond after I provided them with additional information and followed up one additional time while one individual declined due to time constraints.
A total of 17 participants (9 teachers and 8 parents) met the inclusion criteria and were eligible to participate. These participants were considered enrolled in the study once they provided written or verbal informed consent to participate in the interview (Appendix C). If written consent was not obtained prior to the interview, informed consent was completed at the beginning of our interview. Due to challenges with securing any additional interviews, it was decided by my supervisor and members of my committee that data recruitment could cease mid-July 2017.
According to Creswell (2013), data are typically gathered “in the research site, respecting the daily lives of the individuals at the site, and collecting a wide variety of materials” (p. 95). Common data collection methods used in ethnography often consist of observation, interviews, and documents. Data were collected in the context of the school community. Data collection was an iterative process and a maximum of four months was allotted for data collection. Data from multiple sources needed to be organized, sorted, and retrieved for analysis purposes; for this reason, data were transcribed into Excel to assist with this process.
One of the important tenets in ethnographic research is the ability of the researcher to become embedded in the culture that is being studied; this is often done through observation. However, it is important to note that “being embedded is not necessarily linked to the amount of time spent at the research site” (Fusch & Ness, 2017). For that reason, non-participant observations, characterized as a method where the researcher follows the flow of events (Adler and Adler, 1994; as cited in Reeves, Peller, Goldman, & Kitto, 2013) can be beneficial in acquiring knowledge of values, policies, roles, structures, processes, practices, and tools used in the field.
When I began this research study, I was embedded in the school food culture by means of my employment; I was employed by the NS Department of Health and Wellness as the Coordinator of School Food and Nutrition. My position had a focus on food and nutrition policy and programs while also supporting health promoting schools and school health curriculum. This work-related experience afforded me the opportunity to cultivate a critical and complex systems lens. With this, I am cognizant of the context and culture within the school food environment and how this environment impacts students, their families, and their communities.
I am also aware that cultural behaviours and language vary throughout the education system from school to school and between health and education which can strengthen or weaken the opportunities provided in the school community. I was no longer working in this role while I was conducting data collection and due to a labour dispute within the province, I was unable to conduct participant observations in schools. Nevertheless, my work-related experience allowed for an embedded perspective and non-participant observations.
Furthermore, I was also embedded in the culture since I am a parent of three children who were in the NS public school system at the time of my investigation. As such, I was exposed to other parent views of school food as well as teacher and principal approaches to FL. I also sat on the School Advisory Committee (SAC) at my children’s elementary school for two years which allowed me to observe how the school operates and their priorities. This is considered the ‘emic’ perspective in ethnography; which is perspective taken by a researcher who is a member of the community being studied.
Data Sources
The primary source of data collected for this study was semi-structured interviews with research participants. My secondary data sources included review of policy, guidelines, standards and supporting documents as well as my field notes. The following sections outline these sources and their applicability to the research objectives.
Primary Data Source: In-depth semi-structured interviews
Ethnographic interviews are often semi-structured (Hasselkus, 1990; Schensul, Schensul, & LeCompte, 1999; Spradley, 1979). In-depth semi-structured interviews are complex and require consideration of multiple dimensions of culture, boundaries, ethics and format (Fontana & Frey, 2005).
According to Patton (2002), “researchers interview people to find out from them those things we cannot directly observe…feelings, thoughts and intentions. The purpose of interviewing, then, is to allow us to enter into the other person’s perspective” (pp. 340–341). Participants completed an informed consent (Appendix C) to participate in the interviews with an understanding that their contribution would be shared but their personal information would be kept confidential.
Participant interviews were conducted over a period of four months between April 2017 and August 2017. In order to conduct the interviews, I held the position that the most important aspect of the interviews with participants was that they were comfortable to express their experiences and perceptions; therefore, interviews were conducted in a quiet, comfortable, non-judgemental neutral location (Creswell, 2013) either in person or by telephone to accommodate participant requests.
A semi-structured interview guide (Appendix D) with broad questions was used to facilitate the conversation. Consistent with an ethnographic approach to interviewing, Spradley (1979) identified three major types of ethnographic questions to consider when conducting interviews: descriptive, contrast, and structural. The first set of questions I asked participants were descriptive in nature; this allowed me to better understand their role in the school food environment. The second set of questions were contrast questions which afforded me the opportunity to better understand the barriers and facilitators to a healthy food environment.
The last set of questions were structure questions, which enabled me to discover information about the various actors, organizations, and factors that influence school food environments. Furthermore, an active interview approach (Holstein & Gubrium, 1995) was employed for this research, focusing on conversational dialogue and interaction between the teacher and parent participants and myself; this allowed for flexibility throughout the course of the interviews which also aligns with an ethnographic approach.
At the start of the interviews, I explained that I was studying for a Doctorate in Interdisciplinary Studies degree through Dalhousie University, and that I was carrying out an exploratory investigation into FL which was part of a larger research project investigating school food environments. I explained that I was interested in listening to their experiences and perspectives and that there were no right or wrong answers to the questions. The goal of the interviews was to help build an understanding of how FL is conceptualized and communicated in order to inform policies, programs, practices and initiatives as well as how FL is influenced by socio-cultural forces and factors. Interviews ranged from 45-71 minutes, with an average of 50 minutes. All interviews were audiotaped and transcribed verbatim using labels taken from the words of participants.
