Abstract
The significant challenges associated with adapting and delivering the co-design process with Bottom of the Pyramid (BoP) citizens are considered in this article, which tests a new “Empathy Building” step within the Trischler et al. seven-step co-design process model. A sequential, three-study mixed-method research design was applied to co-design a social marketing program with BoP citizens and experts, focusing on the problem of school dropout in a community in South Africa. Study 1 undertook the empathy-building step of the co-design process, which involved a 4-month ethnography in a community in South Africa. This study informed several subsequent co-design steps, including resourcing, planning, recruiting, and sensitizing. Study 2 engaged 38 participants (schoolchildren and parents) in a co-design session, challenging them to create a social marketing program for them and people like them. Study 3 delivered an evaluation session with seven key experts. This article contributes to understanding by offering a process to demonstrate how three studies used the (enhanced) co-design framework to provide a solution for a social issue, namely, prevention of school dropout. Finally, this article outlines how co-design can overcome challenges faced to work with BoP citizens through the addition of ethnography and involvement of experts at the fuzzy back-end of the co-design process to examine program implementation feasibility.
Keywords
Introduction
Social marketing aims to develop health, environmental, or social change programs that can positively influence behavior change to benefit people and the planet (Luca & Suggs, 2013). As argued by French and Lefebvre (2012), all social marketing programs should include a co-design study to empower citizens through implementation of collaborative processes that allow people to design programs for themselves and people like them (Dietrich et al., 2017; Krippendorff, 2011; Sanders & Stappers, 2008; Steen et al., 2011). However, co-design frameworks have been developed and validated predominantly in Western contexts (Dietrich et al., 2017; Trischler et al., 2019) without due consideration of applicability in other settings (Martin & Hill, 2011). For example, low- and middle-income countries (LMICs) feature vast differences in living standards compared with Western countries and this impacts how citizens engage with and experience the natural and material world (Burgess & Steenkamp, 2006; Karnani, 2007). Furthermore, a large and important segment within LMICs, referred to as Bottom of the Pyramid (BoP; Burgess & Steenkamp, 2006; Hahn, 2009; Karnani, 2007), remains understudied (Martin & Hill, 2011; Rapp & Hill, 2015). Limited understanding of BoP citizens can result in development and delivery of social marketing programs that do not address the unique needs of BoP citizens and consequently lack their acceptance (Anderson & Billou, 2007). The significant challenges associated with adapting and delivering the co-design process to a BoP citizen segment are addressed in this research.
This research focuses on the problem of school dropout in a BoP community in South Africa where only 52% of children remain in school by Grade 12 (final year; Weybright et al., 2017). School dropout offers limited opportunities, with many children being left idle (Branson et al., 2014) and children who do not complete school are more likely to engage in dangerous behaviors such as alcohol and substance abuse (Weybright et al., 2017). This research therefore aims to test a new “Empathy Building” step within the Trischler et al. (2019) seven-step co-design process, offering a process demonstrating how three studies used the (enhanced) co-design framework to provide a solution for a social issue, namely, prevention of school dropout. The Trischler et al. (2019) co-design process was developed through insights gathered across six public service design projects in a single Western country. As such, the challenges around delivering the co-design process with BoP citizens were not considered in the Trischler et al. (2019) study. The empathy-building step also aligns to the initial step in the SPRING Accelerator’s (2019) “Human-Centered Design Process” for LMICs termed “Research” which was developed by national development aid agencies and other contributors from around the world. The revised co-design process followed in this research and the Trischler et al. (2019) original framework are presented in Figure 1.
Extended Trischler et al. (2019) co-design process.
A sequential, three-study mixed-method research design was followed in this research to co-design a social marketing program that would aim to prevent school dropout. First, the empathy-building step was completed by the lead researcher and research assistant through a 4-month qualitative ethnographic study from August to December 2018 in Khayelitsha, a large township community in Cape Town, South Africa (Study 1). This empathy-building step also served to inform the next four steps of the co-design process (resourcing, planning, recruiting, and sensitizing). Second, 38 participants (schoolchildren and parents) engaged in a co-design session which relates to Step 6, facilitation (Study 2). Finally, the back-end steps (reflecting and building for change) of the co-design process were completed, involving seven experts to consider design feasibility (Study 3).
Literature review
BoP citizens
“Bottom of the pyramid,” also described as “subsistence marketplaces” and “base of the pyramid,” is an economic term which refers to global citizens who earn less than US$2.50 per day (Chen & Ravallion, 2012; Lappeman et al., 2019; Prahalad, 2009). With over four billion people around the world classified as BoP, this is the largest population group in the world (Lappeman et al., 2019). Most BoP citizens live in Africa, South Asia, Eastern Europe, Latin America, and the Caribbean (Pitta et al., 2008), in regions classified by the World Bank as LMICs.
BoP citizens face severe hardships, including poor nourishment, higher mortality rates, and lower educational standard attainment (Sustainable Development Goals Report, 2020). While some research progress is being made, there is considerable opportunity to extend understanding. For example, Lappeman et al. (2019) observed that BoP citizens are motivated to fulfill higher-order needs according to Maslow’s hierarchy of needs, especially as they relate to education and job opportunities. Interestingly, achieving higher-order needs has little impact on well-being unless consumption adequacy is also established (Martin & Hill, 2011). The top five expenditure categories for BoP citizens include food, energy, housing, transportation, and health (Pitta et al., 2008). Research on BoP citizen behavior remains disproportionate to Western research (Chikweche et al., 2012) and most BoP research focuses on Asia and Latin America, effectively overlooking 800 million BoP African citizens (Lappeman et al., 2019). As 35.8% of the African population are classified as BoP (Chipp et al., 2013), this understudied group provides an opportunity for research (Kolk et al., 2014).
Three key considerations require attention in a BoP context. First, BoP citizens rely on each other for daily living. Strong social ties between community members are formed, and interdependence makes group influences particularly strong (Viswanathan, 2007). Second, substantive heterogeneity exists between BoP citizen groups. For example, 70% of BoP citizens in India live in rural areas, whereas in Brazil, the majority of the poor live in urban slums (Prahalad, 2012). As such, an undifferentiated approach to different BoP markets will not be effective (Prahalad, 2012). Third, due to limited educational opportunities and lack of experience engaging in a variety of consumption activities, such as assessing marketing claims (Davidson, 2009), BoP citizens are particularly vulnerable. Therefore, it is important to extend efforts to close the gap experienced by these citizens.
Social marketing
Social marketing has a proven track record of success undertaking research that investigates group influences on decision making (Collins et al., 2010; Mitchell et al., 2017), segmenting and understanding differences among varying citizen groups (Dietrich et al., 2015; Newton et al., 2013) and working with vulnerable citizen groups (Badejo et al., 2019; Dietrich et al., 2017). As such, social marketing is primed to answer the call from the United Nations’ (UN) Sustainable Development Goals Report (2020) to create programs that attend to the needs of disadvantaged and marginalized populations.
Recent reviews have assessed the efficacy of social marketing (Stead et al., 2007), degree of stakeholder involvement (Buyucek et al., 2016), and strategies employed in social marketing programs (Luca & Suggs, 2013), as well as assessing the suitability of social marketing to address specific health and social challenges such as minimizing alcohol harm (Kubacki et al., 2015), healthy eating (Carins & Rundle-Thiele, 2014), and physical activity (Fujihira et al., 2015). However, a paucity of social marketing research undertaken in LMICs (Gordon et al., 2016) and among BoP citizens remains. Furthermore, social marketing programs implemented within LMICs have faced criticism for overreliance on principles only validated in high-income countries (HICs; Alday et al., 2010). For example, Gordon et al. (2016) argued that “ideas that emerge from voices from the south (LMICs) need to be heard, listened to, acknowledged and integrated into social marketing theory and practice” (p. 1068).
Brenkert (2002) argued that due to the concept of social good being contentious, value-laden, and perceived differently by different people, social marketing may present ethical concerns when naïve assumptions are made by funding agencies. Hence, involving target group members in decision-making is one way to avoid community reactance through the adoption of approaches that empower citizens. Furthermore, behavior change is more effective when driven by what the target group wants as opposed to what researchers develop (Lefebvre, 2013). Researchers can elicit the views of the target group within program design in a variety of ways (Van Velsen et al., 2013; Yen & Bakken, 2012), including, for example, co-design (Dietrich et al., 2016; Trischler et al., 2019).
Co-designing social marketing programs for BoP citizens
Co-design as an approach involves the collaboration of members of the target group and researchers to design a program (Dietrich et al., 2017; Krippendorff, 2011; Sanders & Stappers, 2008) that results in an offering that meets the needs, wants, and preferences of the target group (Ind & Coates, 2013). Co-design gathers information that cannot be collected from traditional market research through empowering ideation and gaining firsthand citizen behavior insights (Dietrich et al., 2017). Social marketing as a behavior change discipline distinguishes itself from other behavioral change approaches on the principle of exchange. Exchange thinking ensures that all design, implementation, and evaluation efforts seek to gain both a program (internal) and a citizen (external) viewpoint. Given that two parties need to voluntarily come together for exchange to occur and that multiple exchanges are required for outcomes to be observed, social marketing programs need to be valued by the target group. Co-design is therefore a tool that can be applied to help improve the likelihood of outcomes being achieved.
Anderson and Billou (2007) advocated the necessity of adapting programs to the unique requirements of BoP citizens as there is substantive diversity in relation to their cultures, ethnicities, abilities, and market needs, and a standardized approach to serving these citizens is unlikely to be effective (Prahalad, 2012). One way to enhance the relevance of a program is through involving the target group in the design process (Dietrich et al., 2017; Trischler et al., 2018). Dietrich et al. (2017) outlined a six-step process for co-designing with vulnerable citizens, which consists of resourcing, planning, recruiting, sensitizing, facilitation, and evaluation. More recently, Trischler et al. (2019) extended this co-design process, replacing Step 6 “evaluation” with “reflection” and adding a seventh step “building for change” to allow for both the front- and back-end steps to be iterative in their approach. The front-end of the co-design process is usually described in the literature as a series of steps (Donetto et al., 2015; Steen et al., 2011). Given that challenges in the field can require multiple iterations (Trischler et al., 2019), the final two steps outlined in the Trischler et al. (2019) co-design process (i.e., reflecting and building for change) encourage a collaborative and iterative approach with experts to determine a viable solution for implementation that will last over time.
An opposing approach to co-design, which places experts at the center of the design process, is expert-driven design. While experts’ core competencies are derived from knowing what is technically feasible, ideas generated by citizens in consultation with experts are more implementable (Magnusson et al., 2003) and may challenge existing expert biases. Although generating ideas that are more actionable is positive, there are drawbacks to citizens consulting with experts during the design process, given the process can negatively influence originality of ideas expressed by citizens. Involving experts during the co-design process may run the risk of citizens losing their unique perspective and becoming “copycat professionals” (Magnusson et al., 2003, p. 119). However, the involvement of key experts at the concept-testing stage is crucial for program success (Tate et al., 2018; Trischler et al., 2019). Following the extended seven-step co-design process outlined by Trischler et al. (2019), experts are involved in program design after citizens have engaged in the design process, ensuring that original ideas can be expressed, copycat ideas can be reduced, and biases challenged through generation of unique concepts by potential program users. Through a sequential involvement process, more implementable solutions that still meet the needs of citizens may be uncovered.
Research design
A sequential, three-study mixed-method research design aligned to the extended Trischler et al. (2019) co-design process was applied. See Table 1 for an overview of the research design. This research was approved by the Griffith University Ethics Committee (REF: 2018/621).
Research design overview.
Study 1: ethnography
The objective of Study 1 was to examine the usefulness of ethnographic fieldwork as a form of “empathy building” employed to overcome challenges associated with the front-end stages of the co-design process.
Method
One method to help researchers gather culturally relevant insights (Brennan et al., 2015; Weidner et al., 2010) and provide a deep understanding of the lives of citizens (Nailer et al., 2015) is ethnography, which remains largely underutilized in social marketing (Brennan et al. (2015). Nailer et al. (2015) and Belk (2014) identified the value of qualitative and ethnographic research in LMICs, noting the limitations of quantitative research for gaining deep understanding of context-sensitive issues. Weidner et al. (2010) argued that research approaches such as surveys and experiments developed for Western citizens may not translate effectively to a BoP citizen group due to communal interdependencies, low levels of literacy, and difficulty responding to traditional attitudinal measures. Given its strengths, ethnography may be an effective approach, allowing social marketers to overcome challenges associated with completing the front-end stages of the co-design process with BoP citizens.
The ethnographic study utilized maximum variance purposive sampling to select research participants (Etikan et al., 2016). This sampling strategy involves selecting a broad range of participants and allowed understanding of the problem of school dropout from many different angles and perspectives. Data were collected through engaging, observing, and participating in day-to-day activities within the community and with community members. Some of these key activities related to living and interacting with a host family, working as an assistant teacher at both a primary school and high school, and visiting community organizations. Following procedures outlined in Belk and Kozinets (2005), Knoblauch (2005), and Pink and Morgan (2013), data collection was intensive: using a variety of methods including semi-structured in-depth interviews, focus groups, and informal conversations (n > 100) that were documented in field notes and videos.
Findings
The ethnographic study established a useful platform assisting to build empathy and informing the resourcing, planning, recruiting, and sensitizing stages to overcome challenges associated with the front-end stages of the co-design process. For example, the ethnographic study facilitated empathy-building through embedding the lead researcher with knowledge on what types of school programs would appeal to participants for inclusion in co-design sessions (e.g., peer mentoring program, personal development program, and professional development program) and ensuring the words (e.g., life skills) used in interactions with research participants were culturally appropriate. Furthermore, the time spent in the community during the ethnographic study allowed the lead researcher to plan the co-design session with community partners and the research assistant. Participants for the co-design study were recruited through networks and relationships established during the ethnography. Students participated in a focus group interview prior to undertaking the co-design session, giving them familiarity (sensitized) with research processes and confidence that their voice matters (Dietrich et al., 2017; Durl et al., 2017). Participants for the parent and expert evaluation sessions were also recruited from networks established during the ethnography and, consequently, were already familiar with (sensitized to) the research problem.
The ethnographic study also provided empirical support for the afterschool programs to be the focal area for co-design. Many participants noted that children are left idle for many hours of the day, especially during the afternoons, due to parents working long hours in the city of Cape Town (often in low-paying roles such as household staff). During this time, children look for ways they can fill their time: looking for “instant gratification.” Furthermore, parental support for education was not always evident: Your parents wake up and leave early in the morning and come home late. So it kinda makes no difference if you went to school or not. (High school student, 17, female). When I refer to instant gratification, it’s plain out sex, drugs, alcohol, nightlife, and going back to the home of course, which I probably didn’t mention, is how absent parents are. So if I’m like a working timeframe of parent getting up as early as four in the morning, five in the morning, going to work, coming back way in the evening. So that absence, that cap, that frame of time, it’s kind of destructive mode for some. (Community program manager, 30s, male) Parents, they easily give up on their children. That’s the other thing that I have learnt when I’m working with the kids. Like for instance the kids that are doing drugs, they are violent, they are rude, they are rude when speaking with their parents. Then the parents, they just give up and say, “I cannot stay with this kind of a person anymore.” (Community program manager, 40s, female).
Engaging with community members through ethnography enabled exploration of the effect of afterschool and alternative programs (i.e., a parental support program or professional development program for teachers) on reducing school dropout. Participants shared how afterschool programs support children in the community: So, including in our art, man, it’s a safe space to express and show you scars and don’t have to be perfect and don’t have to be insecure or over protective. It’s okay to say, “Guys, I’m feeling vulnerable now because of this and this and that reason.” (Arts afterschool program manager, 20s, female)
Although the role that afterschool programs could play in delivering the social support needed for children seeking to advance themselves through education was clearly apparent, the number of these programs to support children in the community was insufficient. Afterschool programs often run with a 50-1 ratio between program staff and children attendees and many children have to walk long distances, in the dark, to access these programs. One community program manager commented, A lot of kids are missing out because the population is very high. And when it comes to that . . . when it comes to facilities, there’s so many less to accommodate. You will find that one team is struggling, transporting their kids, because we don’t have facilities . . . we don’t have transport . . . and it’s not safe for these kids to actually travel quite far. We are struggling when it comes to facilities. If ever we can have more programs that would be great. (30s, female)
Discussion
Study 1 sought to examine the usefulness of ethnographic fieldwork as a form of “empathy building” employed to overcome challenges associated with the front-end stages of the co-design process. Previous studies (Dietrich et al., 2017; Hurley et al., 2018) reported conducting literature reviews to complete the resourcing for their co-design studies. However, due to underrepresentation of BoP citizens in the literature (Martin & Hill, 2011), and BoP citizen heterogeneity (Prahalad, 2012), adopting this approach was unlikely to be effective. Instead, this research employed insights gathered through ethnographic field work to understand the root cause of the focal problem (school dropout), inform the identification of the most relevant program to be co-created (afterschool program), and select an array of activities for consideration within the sensitization phase of the co-design process (e.g., word association and afterschool programs available within the BoP community). Furthermore, the ethnographic study provided substantive time to plan the co-design sessions with team members and local partners. The ethnographic fieldwork was also essential in building networks with experts in the community, which supported subsequent stages of the co-design process, including the fuzzy back-end. This study has therefore provided insights into how ethnographic fieldwork was applied in a new stage of the co-design process, namely, the empathy-building stage.
Study 2: co-design sessions
The objective of Study 2 was to demonstrate application of the (enhanced) co-design framework demonstrating how solutions for a social issue, namely, prevention of school dropout, could emerge.
Method
The co-design sessions attempted to develop an implementable solution addressing the problem of school dropout identified in Study 1 by designing a user-centered and stakeholder-supported afterschool program that could be delivered in the community aiming to reduce school dropout. Four co-design sessions were carried out involving 31 participants aged between 11 and 16 (see Table 2 for list of participants), with two classes at both a primary and high school. In addition, one co-design session that ran with seven parents was completed (see Table 2). Each co-design session ran for approximately 60 min. This duration is consistent with other co-design studies (e.g., Dietrich et al. 2017), ensuring that participant fatigue is avoided. All participants were chosen using purposive sampling made possible through networks with schools, teachers, and community members established through the initial ethnographic fieldwork completed in Study 1. Co-design sessions with primary and high school students were undertaken inside a school meeting room or classroom during school time and the co-design session with parents was completed at a community child care center during the evening.
Overview of participants.
Following procedures outlined in Sanders (2006), each co-design session commenced with an activity designed to sensitize participants to the topic of the study. Participants completed a word association activity, during which they were asked to write down the first thing that came to mind in response to each word. Words were related to afterschool programs and included (but were not limited to) “homework,” “maths,” and “afterschool.” Students and parents were then given an opportunity to review existing afterschool programs. These programs were presented on laminated cards, which provided participants with imagery and a short description of the program (see Figure 2 for examples). In groups of three to four, participants were given the opportunity to (re)design their own program on paper (see Figure 3 for group design examples). To support participants in creating new ideas, tools such as large sheets of paper, pens, sticky notes, and markers were provided. Each co-design session concluded by each group presenting their afterschool program design to the entire audience.
Program card examples.
Participant afterschool program designs: (a) Primary School Participant Afterschool Program Idea and (b) High School Participant Afterschool Program Idea.
In-group discussions and afterschool program ideas were audio and video recorded in all co-design sessions. Program ideas created in the sessions (a total of 10) were analyzed by assessing the key components of each idea (see Table 3 for summary of design ideas generated through this study) following the approach outlined in Hurley et al. (2018). Finally, fieldnotes were made both during the co-design session and at the end of the session to document group dynamics, challenges groups faced, and any important information shared during ideation in small groups that was not shared during design presentations (Dietrich et al., 2017). Thematic analysis to identify key themes in the collected data was undertaken using NVivo Version 10 (Guba & Lincoln, 1994).
Afterschool program designs.
Data from Study 3, Expert Evaluation Study.
Findings
The key findings emerging from Study 2 indicated that the ethnographic fieldwork conducted in Study 1 provided a rigorous platform for the lead researcher to successfully facilitate co-design sessions with BoP citizens. The sessions provided important insights around three main themes: program content, program facilitators, and parental involvement.
Program content
Students’ and parents’ descriptions focused on the importance of having a variety of activities to appeal to a diverse group of students. One participant commented, “it is important because some of the learners are not as good in their academics, they are good at art and communicating with others” (primary school student, 15, female, Co-Design 3). Significantly, 90% (9/10) of design ideas created by citizens had more than one activity included. For example, Sometimes not everyone is good with the books . . . the academics so if there is a chance for them on the sports field . . . where they can demonstrate . . . where they can do things on their own, show their talent. I think they can be a better person tomorrow, they can do things in order [for] them to be like Ronaldo, so their role model. (Father, 50s, male, Co-Design 5)
In addition, the importance of students developing life skills was reflected in 80% (8/10) of the afterschool program designs from students and parents. One parent commented, It gives them life skills . . . so, they can get their lives on track and know how to tackle problems in our area. If you have a clear mind and life is going good, you will focus more on school. So, I like that part. (High school student, 16, male, Co-Design 5)
Program facilitators
Students and parents stated preference for a peer to facilitate the afterschool program with 70% of design ideas including peer facilitator. One student commented, I like the fact that it is people close to our age group. That it is not teachers and or parents that are involved. It is people our own age. It is much easier to talk to somebody like that. Because they relate to you, because they have just been there, they have just been your age. So, they can give you advice ’cause they have been there. (High school student, 17, female, Co-Design 2)
Another student similarly commented that “Because all of us are peers we can understand each other without shouting or raising our voices to each other” (primary school student, 15, female, Co-Design 3).
Parental involvement
Insights from the co-design sessions indicated that program outcomes could be enhanced if parents were involved. One student commented, “Seeing my parent involved, kinda motivates me to do better, because . . . like they show interest” (high school student, 17, female, Co-Design 2). Participants discussed the disconnection between parents and children in the community and the potential for the program to act as a medium for greater communication. One student commented, I like the fact that they include parents, because this will create communication between you and your parents. Because some kids are not really close to our parents, but, maybe if the parents are aware of the activities you are doing . . . when you get home there is something to talk about with your parents. (High school student, 17, female, Co-Design 2)
Discussion
Dietrich et al. (2017) argued that co-design facilitators should have certain skills, including the ability to engage and empower individuals to share their experiences and perspectives. A potential barrier to a researcher exhibiting these skills in the context of this study could be due to them being perceived as an outsider lacking commonality with research participants (Dwyer & Buckle, 2009) for many reasons, including economic disparity between more advantaged researchers and participants who may earn little more than US$2.50 per day (Lappeman et al., 2019). As a result, any researcher from outside the BoP will likely lack comprehension of BoP lived experiences, cultural practices, social norms, and styles of communication, to name just a few important considerations. Consequently, researchers from outside the BoP can only apply an etic (outsider) approach, which entails taking a culture-free standardized approach (Kubacki & Skinner, 2006), which may prevent the resultant social marketing program from capturing the unique needs of the target group (Anderson & Billou, 2007). This etic approach may act as a barrier to establishing rapport with participants during the co-design process. By undertaking ethnographic fieldwork, the lead researcher acquired substantial social capital and some cultural understanding while living with a host family in the community. In addition, working with a local research assistant and working as an assistant teacher in the community assisted the lead researcher to build relationships that in turn facilitated delivery of co-design. Earlier interactions between the lead researcher and participants during the ethnographic study allowed both the researcher and participants to relate to each other during the co-design design sessions assisting facilitation.
Study 3: expert evaluation session
The objective of Study 3 was to examine expert views and consider whether experts felt ideas expressed by potential program users were feasible as a means to reduce school dropout.
Method
Subsequent to the co-design sessions with students and parents, a final expert evaluation session was completed with a group of experts, including teachers, an NGO (nongovernmental organization) worker, and a government official (see Table 2). Following the procedures outlined by Trischler et al. (2019), the final expert evaluation session study attempted to explore the feasibility of citizen-driven design alternatives (generated by students and parents) and to produce a final afterschool program design believed to be both feasible and impactful in addressing school dropout in the community. The session with experts was different from the co-design sessions with students and parents in two ways. First, experts evaluated the design ideas generated by students and parents, rather than considering existing afterschool programs as occurred in the student and parent sessions. Second, all seven expert participants worked as one group to form consensus for the final design of the afterschool program, in contrast to earlier sessions which split students and parents into smaller groups. Data collection and analysis procedures followed were the same as employed in Study 2. Co-design sessions with community experts were completed at a community child care center during the evening.
Findings
The key findings emerging from Study 3 indicated that experts could comment on the feasibility of citizen-generated ideas. After reviewing the afterschool program design from children and parents, the experts determined an afterschool program with two focal content areas: academics and life skills would be optimal. This differed from the afterschool programs created by students which averaged 3.2 content areas (see Table 3). The designs by experts were more focused on education alone, whereas student designs acknowledged the need for activities that deliver social support and life skills, which enhance resilience and in turn facilitate academic endeavors (e.g., sport). The purpose of involving experts was not to change citizen-driven designs, but rather provide key insights around how expert viewpoints differed from potential program user preferences given that stakeholder support is needed for program implementation to occur. The expert evaluation session contributed to the development and refinement of each theme presented in Study 2: program content, program facilitators, and parental involvement aligning program user preferences with stakeholder views.
Program content
The expert evaluation session provided concrete suggestions on the program content in the afterschool program. One suggestion was that one day each week be dedicated to engaging in a variety of extra-curricular activities (sports, arts, etc.) and life skills. Experts, many of whom were teachers, noted the importance of time for academic study, indicating beliefs this would reduce the high incidence of students failing and thereby prevent school dropout (City of Cape Town, 2011). Experts expressed that two days “need to be academic” (principal, 50s, expert evaluation session) and they identified the subjects that should be the focus: “Maths, Physics and English,” noting these have traditionally high failure rates (community program manager, 40s, expert evaluation session).
Program facilitators
The experts agreed that peers would make the best program facilitators for the afterschool program and suggested that they be recruited from a high-performing school in the community. The experts also identified a primary school that would benefit from the program and recommended Grades 6 and 7 be the focus grades “as everything starts from primary” (teacher, 30s, expert evaluation session). One expert commented, You see those kids who are at [Name of High School in Khayelitsha] . . . they are already in tune . . . they’re doing things on their own. Did you see them? Their presentation is pure . . .it is like they are not in Khayelitsha . . . because that is how they are you see. But we want to take these ordinary kids . . . the ones that are here and make them to become like them. So meaning you can use them to be the ones who will be teaching the ones that we can get here . . . My kids all want to be there . . . so when those kids are coming to this place . . . it is very interesting. (Principal, 50s, expert evaluation session)
Parental involvement in program
Participants in the expert evaluation session suggested how the afterschool program could integrate parental involvement. For example, once a month, parents could attend a meeting with a community leader where they can hear about the activities their child has participated in during the afterschool program. The experts suggested using the meeting as a development workshop for parents, with a focus on helping parents enhance their interpersonal communication with their children: Focus on educating them [parents] . . . on how do they speak with their children. For an example, I come from work and my house is not clean, how do I approach them? How do I talk to them? You give them funny names . . . You fat. You lazy. That thing is killing them. When you call them stupid . . . [they] think there is nothing good that can come from me. Then they don’t want to go to school . . . what we must concentrate on is how we speak to children and the role we need to play as parents. (Teacher, 30s, expert evaluation session)
Taken together, experts agreed that peer led programs supported by parents that included extra-curricular activities (sports, arts, etc.) and life skills be implemented in a program aiming to reduce school dropout. A key tension area identified in this study is beliefs about program content, with experts believing more time should be dedicated to academic areas experiencing lower performance rather than extra-curricular activities, which have been shown to develop core academic skills and contribute to improved academic performance.
Discussion
In this research, the final afterschool program design generated by experts overemphasized core academic areas, which was not fully aligned to citizen preferences regarding “program content.” This research indicates that expert-built programs would have limited program performance, given that expert designed programs would not align to program user preferences. This finding is consistent with Rundle-Thiele et al. (2019) and highlights the need for diversity in participation during program design to ensure efforts remain focussed on enhancing program uptake and retention rates, given these are needed to achieve intended program outcomes. In the present research, students asked for activity variation, seeking a focus that was beyond spending time purely on the pursuit of academic excellence. Initiatives preferred by students are crucial in building the social support needed to attain academic excellence. For example, Durlak et al. (2010), in a meta-analysis of afterschool programs implemented in the United States, argued the importance of programs including activities that promote social and personal skills. Given that parental support for academic pursuit is missing for many students, a cross section of activities known to enhance educational outcomes delivers support needed for individuals to attain program outcomes (Coatsworth & Conroy, 2007; Durlak et al., 2010). Experts want their key outcomes, that is, educational attainment, and show a reluctance to challenge their inherent biases by considering the ideas generated by students and parents. Creativity remains the key and partnerships with social marketing and other human-centered design disciplines can yield improved outcomes through finding creative delivery methods to increase program appeal—for example, using sports to teach youth life skills (Hemphill et al., 2019) or using virtual reality to improve attitudes and intentions relating to adolescent alcohol drinking (Dietrich et al., 2019).
Study 3 provides support for a cost-effective strategy to improve the feasibility of citizen-driven ideas through expert involvement prior to afterschool program build. However, the process of involving a homogeneous expert group and giving them “the final say” may limit incorporation of citizen-driven insights into the final afterschool program design. Therefore, citizens should be involved in the reflecting and building for change stages of the co-design process (Trischler et al., 2019) and/or more diverse experts ensuring different professional backgrounds should be included to ensure consensus and acceptance of citizen ideas can be attained.
Conclusion
Addressing known challenges in co-design that could reduce the applicability of the co-design process and inhibit co-creation of engaging programs in BoP communities, this study aimed to test a new “Empathy Building” step within the Trischler et al. (2019) seven-step co-design process model. The study was conducted within the context of developing an afterschool social marketing program in a South African community where school dropout rates exceed 60% (City of Cape Town, 2011). Study 1 determined that prior ethnographic fieldwork built empathy and it helped to overcome challenges associated with the resourcing, planning, recruiting, and sensitizing stages of the co-design process. Therefore, this article advanced co-design application with inclusion of an additional, initial step, termed empathy-building. Study 2 determined co-design sessions could be facilitated by a Western researcher with BoP citizens through application of an extended co-design framework. Therefore, this study provides a process Western researchers can follow when working with BoP citizens to co-design social marketing programs. Finally, Study 3 identified which program characteristics generated by BoP citizens were supported by key experts. This article contributes to understanding offering a process to demonstrate how three studies used the (enhanced) co-design framework to provide a user-generated solution for a social issue, namely, prevention of school dropout, demonstrating which elements were supported by stakeholders holding vested interests.
Limitations and future research
This study is limited to a specific setting and a narrowly defined sample. The co-design study involved 45 participants from one low-income community in South Africa and hence its generalizability to other BoP contexts will require investigation in future research. Investigation of the appropriateness of this co-design process is especially warranted, given the substantive diversity which exists in LMICs (Prahalad, 2012). This study is also limited to specific design tasks, expert types, and the facilitators involved. It utilized word association and an activity requiring participants to evaluate the positives and negatives of existing program designs and was conducted by a nonlocal researcher with the support of a local research assistant from the community. Future research should test the efficacy of different design tasks and a local lead facilitator. In addition, future research should consider involving the composition of experts and program end-users involved in the fuzzy back-end stages of the co-design process to determine which approach can yield optimal outcomes. We do note this agenda would be ambitious, given alternate program designs emerging would require field-testing to examine program performance. Finally, all participants within the co-design study were previously involved in the ethnographic study. Hence, future research should recruit participants who had not previously participated in the ethnographic study to determine whether co-design ideas differ.
