The role of productive activities in the lives of adolescents: Photovoice evidence from Malawi

Abstract

Adolescence is an important transitional period, separate from both childhood and adulthood. Critical physical and mental development occurs during adolescence, including emotional skills, physical, and mental abilities. Behaviors adopted during this lifecourse period have critical implications for adolescents’ future health and well-being. The main research question of the present study is: what is the role of productive activities in the lives and development of adolescents in rural Malawi? As part of this study, selected adolescents from poor rural households were asked to take photographs of their daily (productive) activities. These photographs served as a starting point for focus group discussions. In addition to including adolescents, we conducted qualitative interviews with caregivers and teachers to triangulate and obtain a more holistic understanding of adolescent engagement in productive activities. The main themes that emerged were that 1) the work that is conducted by adolescent boys and girls inside and outside the household is not only perceived by adolescents as a product of poverty, but as a point of pride, as well as a potential means of providing for one’s future, 2) there is a tension between the needs of the family and schooling, and 3) adolescent productive activities are associated with minor although not negligible hazards and injuries. We discuss that these qualitative findings help to better understand how social protection interventions, such as Malawi’s Social Cash Transfer Program, may affect adolescent engagement in work and adolescent wellbeing more generally.

1. Introduction

Adolescence is an important transitional period, separate from both childhood and adulthood. Critical physical and mental development occurs during adolescence, including emotional skills, physical, and mental abilities. Behaviors adopted during this lifecourse period have critical implications for adolescents’ future health and well-being (Mmari et al., 2014). Additionally, in adolescence, gender norms are “solidified, rejected, or transformed” (UNICEF, 2016). According to UNICEF, “evidence shows that when adolescent girls and boys are supported and encouraged by caring adults, along with policies and services attentive to their needs and capabilities, they have the potential to break long-standing cycles of poverty, discrimination, and violence” (UNICEF, 2016).

Work carried out by adolescents in and outside the household can have significant effects on education and health outcomes and hence on adolescents’ transition to adulthood. There is an extensive literature on the tradeoff between children’s and adolescents’ labor and schooling, including reduced school attendance, worse school performance, and increased chance of grade repetition among those who work (Assaad, Levison, & Zibani, 2010; Beegle, Dehejia, & Gatti, 2009). Additionally, global studies have found that child labor significantly increases the probability of illness (Beegle et al., 2009; O’Donnell, Rosati, & Van Doorslaer, 2005).

The present study provides a nuanced qualitative picture of the role of work in the lives of adolescents in rural Malawi. Adolescents in Malawi commonly support their households both by caring for other household members and engaging in income generating activities. According to representative 2013/2014 UNICEF MICS data, about 70% of adolescent minors engage in economic activities (National Statistical Office of Malawi, 2015). Secondary school enrollment rates and school attendance rates of adolescents are low. In 2012, the largest single reason given for dropout in primary or secondary education was “family responsibility” according to the Ministry of Education, Science, and Technology (2012), including dropping out due to a need to take care of family members and/or help provide for the family.

The study complements experimental quantitative analysis examining the effect of one of the government of Malawi’s most important social protection programs, the Social Cash Transfer Program (SCTP) locally known as Mtukula Pakhomo, on adolescent’s participation in work (De Hoop, Groppo, & Handa, 2017). The SCTP currently provides regular income transfers to over 170,000 ultra-poor, “labor-constrained” households.¹ Social cash transfer programs such as the SCTP have been shown to improve vulnerable adolescents’ transition from adolescence to adulthood. In settings of severe poverty, unconditional cash transfer programs have been shown to increase the school participation of adolescents (Baird, McIntoch, & Özler, 2011; Handa, Natali, Seidenfeld, & Tembo, 2016). These programs also improve adolescents’ mental health (Baird, de Hoop, & Özler, 2013; Kilburn, Thirumurthy, Halpern, Pettifor, & Handa, 2016), delay their sexual debut, pregnancy, and marriage (Baird et al., 2011; Handa, Halpern, Pettifor, & Thirumurthy, 2014; Handa et al., 2017, 2015), and reduce both engagement in risky sexual behavior (Cluver et al., 2013) and the probability of contracting sexually transmitted disease (Baird, Garfein, McIntosh, & Özler, 2012).

Following the suggestion of Orkin (2011), this paper provides a “holistic view” of adolescents’ experiences of work to inform policy making. We provide a broad overview of productive activities in the lives of adolescents and briefly reflect on the effect of the SCTP on these activities towards the end of the paper. Our approach answers a call in recent years, particularly from scholars in Sociology, Anthropology, and Political Science, for child and adolescents-focused economic research that recognizes “children as actors in their own rights, constrained by societies’ constructions of appropriate spaces and activities for childhood but mediating the impacts of social boundaries by their choices and behavior” (Levinson, 2000, p. 125).

A key feature of this qualitative study is the use of a photo-elicitation approach informed by the photovoice methodology to facilitate the elicitation and inclusion of adolescents’ perspectives. Photovoice (Wang & Burris, 1997) is a participatory action research method based on health promotion principles and the theoretical literature on education for critical consciousness, feminist theory, and a community-based approach to documentary photography (Wang & Burris, 1997; Wang, Burris, & Ping, 1996). As part of the photo-elicitation approach, selected adolescents were asked to take photographs of their daily activities, especially productive activities, and to reflect on these activities based on the photographs in a focus group discussion. Our approach was informed by the photovoice methodology because it “positions [adolescents] at the centre around which key research questions, descriptions, interpretations and analyses are made” (Crivello, Camfield, & Woodhead, 2009, p. 52) and actively engages adolescents in the process of developing and implementing the research.

In addition to including adolescents, we also conducted qualitative interviews with caregivers and teachers to triangulate and obtain a more holistic understanding of the phenomenon of adolescents’ labor and how it was perceived to be impacted by the SCTP. The qualitative data helps to address the recommendations of a recent study by Krauss (2017), which emphasized the importance of combining quantitative with qualitative methods to identify a broader range of potential factors for why children work (p. 1).

The remainder of this paper proceeds as follows. Section 2 gives a general background on adolescent work globally and in Malawi, Malawi’s SCTP program, the quantitative impact evaluation examining the effects of this program on adolescent labor, and Photovoice. Section 3 describes the methodology of the study. Section 4 describes the main qualitative findings complemented with information from the caregiver and teacher interviews. Section 5 reflects on the role of the SCTP and concludes.

2. Background

2.1. Adolescent work globally and in Malawi

In 2016, approximately 14% of all children aged 5 to 17 globally were estimated to be engaged in economic activities (ILO, 2017). Child engagement in economic activities is markedly more common in Africa (20%) than in other regions. And the lions’ share of children engaged in economic activities (over 70%) works in agriculture, especially on household owned farms. Boys are more likely to engage in economic activities than girls, while girls engage more intensively in household chores. Adolescents (aged 15 to 17) are particularly likely to engage in economic activities (25%).

The National Statistical Office of Malawi (2015) provides nationally representative information on adolescents’ (aged 15 to 17) engagement in economic activities and household chores. Over 70% of adolescents engage in economic activities. Boys are slightly more likely to engage in economic activities than girls (73 v. 68%) and adolescents in rural areas are markedly more likely to engage in economic activities than their counterparts in urban areas (77 v. 37%). Most girls and boys (over 90%) engage in household chores. Following standard procedures, National Statistical Office of Malawi (2015) classifies about 59% of adolescents in Malawi as engaged in potentially harmful child labor (including engagement in long hours of work and exposure to hazards).

According to the 2017 Global Education Monitoring Report, 50% of adolescents in Malawi who should have been in Lower secondary school were out-of-school, and the upper secondary adjusted net enrollment rate was only 41% (44% for males, and 38% for females). The secondary school net attendance rate is much lower in rural areas than in urban areas (Malawi 2015–16 DHS, 2017). As mentioned in the introduction, reliance on adolescents to provide for their household and care for household members is presumed to play a role in these schooling outcomes (Ministry of Education, Science, and Technology, 2012).

2.2. Malawi’s Social Cash Transfer Program

Malawi’s SCTP is a large-scale social protection program that aims to “reduce poverty and hunger and increase school enrollment rates” in “ultra-poor, labor-constrained households” (Malawi SCTP Evaluation Team, 2016).² Beneficiary households are identified in two steps, the first consisting of a community-based shortlisting process and the second of a proxy-means test to confirm that shortlisted households are indeed “ultra-poor”. Transfer amounts increase with the total number of household members and the number of school-aged household members and are equivalent to about 20% of average beneficiary household income. The program currently reaches over 170,000 households in 18 of Malawi’s 28 districts.

2.3. Impact evaluation

Several quantitative studies document the effects of unconditional cash transfer programs, such as Malawi’s SCTP, on child and adolescent work (see also De Hoop & Rosati, 2014).³ Most of these studies find that such programs lower participation in economic activities and household chores (Covarrubias, Davis, & Winters, 2012; Edmonds, 2006; Edmonds & Schady, 2012; Handa et al., 2016; Sebastian et al., 2016). The quantitative evaluation of Malawi’s SCTP program referred to in the introduction (De Hoop et al., 2017) finds the opposite pattern. Relying on a cluster-randomized trial implemented in two districts of the country (Mangochi and Salima) and data collected from over 3200 households, the study shows that beneficiary households invest the transfers in their agricultural enterprise. All household members, including adolescents, accordingly increase their engagement in economic activities on the household farm and household chores, while lowering their participation in economic activities for pay outside the household. Interestingly, adolescents’ engagement in “hazardous activities” as measured using a standard UNICEF questionnaire module, increases. Nonetheless, broader impacts of the cash transfer program on child and adolescent wellbeing are positive: the program improves school attendance, health, food consumption and material wellbeing.

The present study was designed to generate independent findings, but also to help interpret and contextualize these quantitative results. The present study also complements qualitative interviews with 16 households embedded in the treatment arm of the quantitative evaluation (separate from the qualitative data analysed in this paper), which confirm that the SCTP reduced the need for household members to carry out work for pay outside the own household and enabled households to increase their investment in education (Malawi SCTP Evaluation Team, 2016).

2.4. General background on the photovoice methodology

Photovoice uses photographs and focus groups as methods for data collection. Participants document the reality of their daily lives through photos they take in their homes and communities, guided by themes identified by the researchers and/or the community. The images are used to encourage reflection on life experiences and how they relate to the topic of interest. This methodology has been adapted to the Malawi context and piloted in 2015 with SCTP youth beneficiaries in Balaka District on the topic of experiences of resiliency and vulnerability (Barrington, Villa-Torres, Abdoulayi, Tsoka, & Mvula, 2017). Particularly, Barrington et al. (2017) noticed that the youth were noticeably more engaged in the Photovoice sessions than similar youth in their teams’ prior in-depth interviews.

According to Wang (1999), the three main goals of Photovoice are (1) to enable people to record and reflect their personal and community strengths and concerns, (2) to promote critical dialogue and knowledge about personal and community issues through group discussions of photographs, and (3) to reach policymakers. According to a review by Jacquez, Vaughn, and Wagner (2013), “by partnering with youth to identify content area, research questions, data collection methods, and appropriate dissemination efforts, researchers significantly increase the chances that research findings will be applicable to children and adolescents living in the real world” (p. 50). The authors found a total of 56 projects that partnered with youth in the research process. The review outlined several research studies as exemplary portrayals of successful engagement of youth in research. Wang (2006) also argues that photovoice is an ideal way to engage youth to enhance their community’s wellbeing. For instance, Photovoice has been used with adolescents to explore themes of: condition that contribute to or prevent violence in Flint Michigan, the AIDS epidemic among African Americans in Oakland, California, and the influence of tobacco within young people’s lives and communities in North Carolina (Stevenson, 2002; Strack, Davis, Lovelace, & Holmes, 2004; Wang, Morrel-Samuels, Hutchison, Bell, & Pestronk, 2004).

Youth involvement in photovoice harnesses the developmentally-informed desire of young people to exercise autonomy, hone their critical thinking skills, and express creativity while documenting their everyday lives (Allen, Moore, & Kuperminc, 1997; Benson, Scales, Hamilton, & Sesma, 2006; Lerner, Almerigi, Theokas, & Lerner, 2005). This process is especially significant for young people who may be marginalized within their families and communities due to factors such as orphan status and poverty. Therefore, it is an ideal strategy to engage adolescents in analysing their everyday economic and household work experiences within the context of issues in their community.

2.5. Adaptation of photovoice methodology in the context of this study

In this study, we drew on the principles and practice of Photovoice. However, because of constraints on time and funding, we did not implement the Photovoice components of collaboratively assigning topics with participants, involving participants in the analysis and/or write-up of the results, nor did we facilitate an action component like some (Carter-Edwards et al., 2015) in which study participants work to address jointly identified issues. Instead, based on previous experiences using in-depth interviews with adolescents in Malawi (Rock et al., 2016), we chose a photo-elicitation approach to involve adolescents in this context in a more engaging way as we describe below.

The photovoice methodology of SHOWeD (Wang, Yi, Tao, & Carovano, 1998) was used in the study to facilitate group discussion among adolescents. In this methodology, participants answer each of the questions that form the mnemonic SHOWeD. These questions include 1) what do you See here? 2) What is really Happening here? 3) How does this relate to Our lives 4) Why does this problem or situation exist 4) What can we Do about it? Through these questions, participants move from the literal discussion of what they see in the pictures to the identification of why this issue exists and what their community can do about it.

3. Methods

The quantitative experimental evaluation of the SCTP was carried out in rural areas in the districts of Mangochi and Salima – the districts where the quantitative trial was carried out. Mangochi and Salima are both located at the southern end of Lake Malawi (roughly in the middle of the country). The districts are mostly rural and have high poverty rates similar to the rest of the country. A total of eight villages were selected from all villages in the treatment arm of the quantitative evaluation of the cash transfer program, four in each of the two districts. These villages were selected to represent the general geographical characteristics of all villages incorporated in the quantitative study. In particular, four village clusters were selected (two villages per cluster) based on 1) rural location 2) proximity to the lake 3) proximity to a trading center and 4) presence of tobacco farming in the area. These four factors were considered relevant because they can be related to the type of work in which adolescents engage in different locales such as (1) farm labor (2) fishing (3) selling goods and (4) working on tobacco estates. The sampled villages are rural and it is common for households in these villages rely at least partly on maize grown on the household farm for subsistence.

In each village, six adolescents (three boys and three girls) were selected from all adolescents aged 14–18 whose household participated in the quantitative endline survey. Adolescents were prioritized based on their exposure to work-related hazards according to the quantitative survey data (i.e. the adolescents experiencing the largest number of hazards according to the quantitative data were invited to participate). In some cases, there were not enough ranked adolescents in the village to reach a total of three boys and three girls. In this case, adolescents from nearby “backup villages” were also ranked so that a total of six participants could be reached. The caregivers of the top ranked girls and boys were asked to consent in the adolescents’ participation in the photovoice exercise. The adolescents were then approached to participate and their assent was obtained.

The photo-elicitation process consisted of two sessions, one on the first day of data collection in the village and the other on the last day of data collection in the village with a period of one day in between for the participants to take pictures for the photo assignment. Each session had one interviewer as the discussion leader and one as a note taker. The first session included a general discussion on the daily activities in which the adolescents participate (including economic activities and household chores). The first session also included a thorough discussion on the photo assignment, which was to take ten photographs of the work and the chores carried out by children and adolescents in the community. The first session also included a discussion of the different ways to take photos and the upmost importance of asking for permission to take photos of people and getting explicit assent for photos that include people’s faces. The participants were provided with an assent form, a pen, and an ink pad for obtaining assent for photographs. Additionally, if the photo was of a child (under the age of 18), the participant also sought assent from the guardian of that child. The participants then learned how to use the cameras and practiced taking photos.

During the second session, the participants uploaded their photos to a computer and each participant then selected the photo that they most identified with. This photo was printed on a battery powered printer. At the beginning of the session, participants went through each of the 6 photos and briefly explained where the photo was taken and why they took the photo, they also linked each signature on the assent sheet to a specific photo, which was noted by the interviewers. After each participant shared their photos, the adolescent participants collectively selected two of the six printed photos to discuss. The interviewer then facilitated the focus group discussion following the SHOWeD methodology. Additional questions were included in the guide for interviewers to ask in the case the discussion on the photos did not include information on hazards to their satisfaction.

We triangulated the photo-elicitation data with in-depth interviews (IDIs) with the caregivers of young people (age 14–17 at endline) and key informant interviews with teachers. In each village, three caretakers and one teacher of the main primary school was interviewed. The qualitative interviews with caretakers and teachers provide valuable information about the context of adolescent engagement in paid and unpaid work in the community and at home. The caretakers were selected from a sub-sample of the households from the quantitative impact evaluation that (i) had participated in the quantitative impact evaluation survey waves, (ii) were able to participate in an interview in Chichewa (the most widely spoken local language), and (iii) had at least one child aged 5 to 18. We aimed to have one caregiver interviewer from each of the following categories, mapping onto the trends in child labor observed from the quantitative SCTP impact evaluation data: (1) The household exhibiting the strongest increase in agricultural work children and adolescents engaged in over the three quantitative surveys, (2) the household exhibiting the strongest decrease in informal wage labor that children and adolescents engage in over the three quantitative surveys, and (3) the household in which children were most heavily engaged in hazardous economic activities in the final wave of the quantitative study. The teachers interviewed were selected through an initial conversation with the head teacher and/or other key staff from the school. The teacher selected was intended to be a senior teacher who lived in the community and had worked at the school for at least a few years (Table 1).

Table 1.

Overview of study design.

Method	Study population	Sample size
In-depth interviews	Caregivers	24
Key informant interviews	Teachers	8
Photovoice focus-group discussions	Adolescents (14–17)	8 FGDs (45 participants)
	Total participants	77

All interview and focus group discussion recordings were simultaneously translated and transcribed from Chichewa into English. We engaged in an in-depth and inductive approach to the analysis process, following a sequence of interrelated steps including: reading, coding, displaying, reducing, and interpreting (Tolley, Ulin, Mack, Succop, & Robinson, 2016). The first step of this process involved immersion-reading and rereading transcripts and summaries. We identified emerging themes from the transcripts and compared them to a priori themes developed while creating the interview guide. We subsequently compiled these themes into a codebook with both deductive and inductive codes.

After the codebook was initially finalized, we added it to the Atlas.ti software (Berlin Scientific Software, 2013). We then read through the interviews again, attaching the codes to text (ranging from phrases to whole paragraphs of text) that fits the thematic definition of the code from the codebook using Atlas.ti. At the beginning of the coding process, two researchers coded a sample of transcripts and examined for discrepancies in coding to establish a standardized method of applying codes. After this, one researcher coded the remainder of the transcripts, conferring with the other researchers and/or translators if any questions arose in this process.

Once the transcripts had been coded, we explored each thematic area, first presenting in detail the information relevant to each category through generating reports on each code that include all the text that the code has been applied to. We then reduced this information to its essential points. During this process, we created memos capturing the various general themes and concepts, an important iterative step in the research process.

4. Results

As planned, around half of the 45 participants were girls and half were boys. The average number of participants per focus group discussion was 5.6, with a maximum of 6 and a minimum of 4. The average age of participants was 15, with a minimum of 14 and a maximum of 18. Eight participants had dropped out of school, either having dropped out during primary school or not having pursued secondary school after completing their primary education. This is a comparatively low number of dropouts in our sample, compared to average school dropout rates, but our sample is small and purposefully selected.

4.1. Common activities

There was a clear gender division in the activities carried out by adolescents for their own household without pay. The responsibility for most of the household chores fell on the girls within the family. Chores commonly performed by girls included fetching water, fetching firewood, fetching grass/reeds for the house, sweeping, cooking, washing dishes, washing clothes, and caring for younger children in the household as well as elderly household members. At time, boys would help girls in the house with their chores, such as collecting water. However, boys were mostly responsible for supporting the household agricultural enterprise including sharing the responsibility for farming the household plot, tending animals, and procuring materials such as molding bricks. Non-livestock agricultural activities included clearing land, making ridges for planting, weeding, planting, and harvesting. Boys also collected grass/reeds to use in building and maintaining the household structure, fence, and latrine and were sent on household errands such as going to the maize mill or picking up vegetables at the market.

Both boys and girls engaged in activities for pay and income generating activities for the household. Common activities included selling surplus crops (such as maize, tomatoes, pigeon peas, sweet potatoes, cassava, and groundnuts) at local markets or in the markets of villages that are further away from the household. Adolescents also cut grass/reeds in the wetlands or and/or firewood in the mountains and often sold the surplus after accounting for household needs for fuel and roofing. Male participants also earned money for the household through construction work on roofing, smearing dung and mud on the exterior of houses and in digging pit latrines for others in the village. Additionally, boys were involved in molding bricks and girls would collect water for brick making. Adolescents also worked in cooking and/or selling scones (mandasi or zitumbwa) or selling juice for their parents.

Both girls and boys also engaged in what was referred to as Ganyu labor, which is mostly informal, piecemeal wage labor outside of the home. Ganyu activities included planting and harvesting crops and tending livestock for others in the village. Participants also worked on the larger farm estates adjacent to their villages that mostly grow cash crops such as tobacco. Adolescent participants mentioned that some are employed daily at the estates, while some work monthly. Adolescents get the most Ganyu opportunities at the estates harvesting tobacco when it is mature. Participants also travelled to work in fishing or factories in Mozambique or South Africa. Other participants were involved in working in shops in trading centers, such as helping to charge phones.

According to the adolescents, most of the proceeds from Ganyu labor are used to buy clothes, groceries, and provide for other household needs such as soap. However, there is some difference between responses of adults and adolescents regarding how adolescents spend their money. While some caregivers mentioned the economic contribution of adolescents to household expenses, some of the caregivers referenced giving adolescents some of the money from their labor, or letting them keep the entire amount. Caregivers even reported giving some of the cash transfer funds to the adolescents for pocket money to spend on clothes, soap, or spending time at the local tea room. This discrepancy between the adolescents and caregiver’s responses may have been related to social desirability bias on both sides, some adolescents wanting their peers and the facilitators to know that they help support the household, while some parents wanted the interviewers to know that they allowed their children to keep their earnings.

4.2. Perceptions

Throughout the photo-elicitation focus group discussions, adolescents expressed a pride in fulfilling their designated role in the household through work inside and outside of the household. According to a 16-year-old male participant who is a paternal orphan who is heavily engaged in household chores and is not enrolled in school, “when parents decided to have us, they said I should have a child to help me at home.” [Salima V1 P3] In explaining why adolescents do chores to support their parents, the participant perceived that the primary reason they were brought into the household was to help their parents in this way. The adolescents in the photo-elicitation session saw that the purpose of their household responsibility was “to relieve them (parents/guardians)” [Salima V1 P1, 14-year-old maternal orphan living with extended family]. Though the act of doing chores was not referred to as being enjoyable, chores were mentioned in a positive manner by girls, in household work, this was expressed in helping providing food for the household through keeping the house clean and hygienic through activities such as washing dishes. Adolescent participants saw performing chores as an important positive role in the household, especially for girls (Photo 1).

Photo 1.

Adolescents sweeping household compound.

Photo credit: Salima Village 1 Participant 3.

Chores for boys were mentioned as relating to maintaining the structure of the home, such as digging pit latrines and repairing roofs. Roles for girls and boys were clearly delineated. For instance, according to a 15-year-old female participant, who spends many hours on household chores, “it is not good for boys to wash plates, it is not their role” [Salima V1 P4].

Adolescent participants also had a positive perception of economic activities for pay and for the household. They felt proud to help provide food for the household through farming and to help their parents in buying food, goods for the household, and in paying for school expenses such as school fees, uniforms, notebooks, and pencils. According to a 15-year-old female double orphan living with extended family [referring to Photo 2], “I feel good because I started this work some time back and that’s how we manage to have food in our households.” [Mangochi V4 P4]. Economic activities are seen as especially important during the period between November and February, which is the most labor-intensive time of the year on farms, but also coincides with the lean season, where there is low availability of staple foods, such as maize. According to another participant, a 16-year-old female participant, “Ganyu is not only for the money, we do the Ganyu at times, especially in the months of December and January, to get food instead of money.” [Mangochi V4 P5].

Photo 2.

Adolescents in a maize field.

Photo credit: Mangochi Village 31 Participant 5.

Adolescent participants expressed a sense of responsibility for providing for their household, especially in the context of the outmigration of many of the more traditional caretakers. In reference to Photo 2, in a discussion of why adolescents engage in farming Ganyu, according to an 18-year-old male participant who is not enrolled in school and only completed 4 grades of primary education, “the one I relied on moved out for marriage very far from here. The other went away for work. I am the only man left at home, so when something is needed, it is me who has to go look for Ganyu. When I find it and get paid, I go to the market to buy what is needed.” [Mangochi V3 P2].

Related to the situation of absence of traditional caretakers, participants expressed a value of independence and self-reliance, of being able to provide for themselves in case of the absence of a caretaker. According to a 15-year-old female paternal orphan, in reference to why adolescents do farming work (in reference to Photo 2) it was important to learn farming skills because “if we were to just sit and watch the parents do it, we would be in trouble in future after the parent’s passing” [Mangochi V1 P4]. In reference to Photo 3, a 16-year-old male participant mentioned: “as we can see that in this picture the boy is building a house, this in our life helps us to learn and sharpen the building skills so that when we marry one day and our in-law is asking us to construct a house for them, we should be able to do that without problems” [Salima V3 P3].

Photo 3.

Staged photo of adolescents building a house.

Photo credit: Salima Village 3 Participant 2.

4.3. Poverty and future opportunity

Frequently, the discussions on why the situations in the photos exist centered on poverty. According to a 16-year-old female participant, in reference to Photo 4, “in our country, we depend on farming. We are poor. If we dare to stop farming, we can get poorer or die due to poverty, we can’t wait. In Malawi, we depend on Nsima (maize porridge) and nothing else.” [Mangochi V4 P5]. According to a 16-year-old male participant, “the household we are staying in is failing to provide us with our daily needs, we engage in different kinds of economic activities.” [Salima V3 P1]. Participants referenced Ganyu in the context of basic needs including food, soap, and clothing. According to a 17-year-old female participant who spends over 10 h per week on Ganyu, “sometimes we may have no clothes, and Ganyu is the only way we can get money” [Salima V1 P3].

Photo 4.

Adolescents clearing maize from household plot.

Photo credit: Mangochi Village 4 Participant 6.

While adolescent participants mentioned the importance of work for providing for the immediate needs for the family, not for their own consumption, participants also mentioned the aspiration to do Ganyu to save money for a passport to work in South Africa. In reference to Photo 3, in discussing why adolescents are involved in farming Ganyu, according to a 15-year-old female participant “both boys and girls go to South Africa. As for me, my siblings did some tobacco production and got money for a passport and they are both in South Africa... they left when they were 22.” [Mangochi V1 P3]. However, moving to South Africa could also be related to the desire to provide for the family since many of the families in the study were supported primarily through remittances from family members working in South Africa or Mozambique.

Participants also mentioned the desire for resources to start their own businesses. According to one participant who is engaged in selling mandasi, “for some of us who can manage to do business, we need money to start business so we can buy what we need” [Mangochi V3 P5]. A 15-year old female participant who is a paternal orphan involved in 7 h of economic activities per week, said that they would use money to “buy things and sell them. For example, I would buy flour, make Mandasi and sell them to make money. When I have made money, if I don’t have clothes, I would buy, the rest I would use to buy household basic needs” [Mangochi V3 P4]. However, participants mentioned having neither skills nor capital to start these businesses. According to a 16-year-old female participant, “there is need for organizations to take part to assist the youth, because on our own, we can’t manage. They should bring interventions that will assist the youth to gain skills that can bring food on their table, that way people will have improved livelihood activities, hence improved households and that will help the youth even to assist other youth in the community to also get the skills since they will be employing each other. They can bring the materials for carpentry, wielding... then youth will stop doing Ganyu” [Mangochi V4 P5]. This participant also talked about the lack of productivity of adolescent farming due to lack of farming tools to improve their yield. According to this participant “once we yield a lot we will have food security and we will be able to employ others. Then our challenges will be gone in no time” [Mangochi V4 P5].

4.4. Schooling

There was a theme of the overall tension between the importance of education versus the need to provide for oneself and one’s family in the discussions with adolescents, parents, and teachers. However, the view of teachers appears to diverge from that of parents and adolescents. Teachers mentioned that parents who had not gone to school did not understand the importance of education. They argued that these parents push economic activities on their children and encourage them to discontinue their schooling. However, both the caregivers and adolescents repeatedly emphasized the importance of education. Caretakers generally indicated that they prioritize child engagement in economic activities only in situations of severe need. However, families in this context are often in precarious states. Caretakers indicated that prioritizing children’s engagement in economic activities is accompanied with feelings of guilt and regret. A 33-year-old participant taking care of five children under the age of 14 said, “I feel guilty that I’m killing the child’s future… in March I got very sick, I am the one they rely on to bring food on the table, my husband drinks a lot and doesn’t do any-thing to support the family so the child was really pressed. Him being the eldest at home, he was supposed to do everything alone and when it’s too much, he could miss classes.” [Mangochi V3 CG 2].

This tension between work and school was also expressed by adolescents. According to a 14-year-old paternal orphan, “there are some children that might be given clothes for siblings to wash and be told to do dishes, fetch water and cook and one may fail to go to school” [Mangochi V1 P6]. When asked what they think should happen in this situation, a 16-year old female participant said they “should refuse to do the work” [Mangochi V1 P5], while a 14-year old female participant said that they should “help each other” [Mangochi V1 P6]. Adolescents also referenced that they should work hard in school. According to an 18-year-old girl, “I think we need to work hard in school so that we should do better in the future and help our parents” [Salima V1 P4]. Though most of the adolescent participants were in school, some of them said that it is a violation of their rights to engage in economic activities instead of going to school, and that this frequently happens in their community due to poverty. According to a 16-year-old female participant, “…during the rainy season we go for Ganyu in the estates that are far away. Some adolescents even go to Mozambique to search for farm Ganyu and that results in school drop-out” [Mangochi V4 P5]. Adolescents also mentioned that they should be in school and not working. According to a 16-year-old male participant, “we should work hard in school and not engage in selling reed. School is the future.” [Salima V3 P6]. Another adolescent, a 15-year-old male participant, mentioned that they should not miss school to do Ganyu “so that we get educated and get good jobs in the future just like you guys (referring to focus group discussion facilitators)” [Salima V2 P2].

One particular challenge that came up in regards to education was paying for school uniforms and exam fees. Some students were sent back from school until they could pay the fees to write their exams to pass to the next year, which ranged from 350 to 450 kwacha (~0.5 USD), depending on the grade attended by the child. According to a 17-year-old female participant, adolescents engage in Ganyu because “during exam period, they need money and we go do Ganyu knowing that if you don’t pay you will not write exams” [Salima V1 P3]. According to a 15-year-old female participant who has only completed standard 3, “this relate to our poverty, our parents do not manage to provide us with all our needs so we engage ourselves in this type of Ganyu to source money and buy books, pencils and sometimes the school uniform” [Salima V3 P4].

4.5. Hazards and health

The most common hazards that were mentioned by caregivers, adolescents, and teachers were carrying heavy objects, using dangerous tools, and being exposed to dust. Adolescents carried heavy loads during chores such as carrying heavy buckets of water and bundles of firewood. Adolescents also carried heavy loads of maize and sugarcane while farming and heavy loads of sand and water while brick making.

Injuries were commonly reported by caregivers and adolescents. Injuries tended to be minor in nature, such as cuts from sharp grass, washing broken or warn metal plates, or tools used when farming (such as a panga knife), although an exception was when a child of a caregiver participant had soil and rock collapse in on him when he was digging a latrine. Most of the injuries were framed by caregivers as happening when an adolescent was not careful in their work.

Cases were also commonly mentioned of adolescents facing bullying, violence, and/or intimidation during the course of their chores and/or economic activities. Intimidation and bullying was commonly mentioned at the borehole, where older women and children would berate younger girls and cut them in line. It was mentioned in one village that a committee was formed to protect children from this at the borehole. One caregiver had a son whose bicycle was stolen on the road when he was transporting maize back from the farm. Earnings and harvested produce were sometimes also stolen from adolescents who could not defend themselves. Other adolescents mentioned that some-times they are not paid for a long period of time after working for others doing Ganyu.

Adolescent participants mentioned that children and adolescents in the community face violence from their family when they do not carry out chores. For instance, they may be yelled at or deprived of food. They mentioned that orphans in the community are more vulnerable to this treatment. According to an adolescent participant

Participant 2: “most of the children that are being violated without apparent reason are those that do not have parents, the orphans” Facilitator: “The orphans, what type of abuse do they experience?”

Participant 3: “even if they work and do the chores in the household they are always being shouted at and denied food.”

Facilitator: What else?

Participant 1: “the children that are orphans are denied good clothes. When the guardians want to buy clothes they will only do that for their biological children and leave out the orphan.”

[Mangochi V2]

Caregivers and adolescents did not mention any exposure to chemicals. While some households used pesticides and the adolescents could have been exposed to them on the farms, they were not the ones directly using the chemicals. Participants mentioned wearing gloves while working in the farm.

Most of the dust, fumes and gas exposures that were mentioned were exposure to dust while farming due to the dry land or exposure to dust while sweeping the house or exposure to dust kicked up from cars on the road. According to male caregiver, in reference to his 14-year-old son “Harvesting the sweet potatoes involves the digging of the ridges that to get to the tuber crop. This make the dust and cause him to get sick with a cough” [Mangochi V2 CG3]. Adolescents were also exposed to smoke while cooking.

While caregivers and adolescent participants mentioned adolescents working on the farms or doing construction in the mid-day heat, or collecting water early in the morning in the cold weather, there were no references to any activity that would fit a definition of being conducted in extreme heat or cold. The only exposure that was consistently mentioned to extreme noise was working near the maize mill. However, unless the family owned the maize mill, the adolescents were typically exposed to the noise for a short period of time.

5. Concluding summary and reflection on the role of the SCTP

The present study shows that adolescent perceptions of the role of work in their lives are nuanced. Adolescents broadly perceive their engagement in work as a point of pride, as well as a potential means of providing for their future, as earnings can pay for school uniforms, exam fees and passports to migrate internationally for work. However, they may perceive a need to work as a violation of their rights when it interferes with their education. Youth indicate, moreover, that work may expose them to (mostly minor but not negligible) physical hazards.

Productive activities that keep children and adolescents out of school and expose them to hazards appear to be the result of underlying challenges faced by the household, particularly chronic poverty and adult labor constraints. Caregivers, for instance, mention that they may rely on their children at the expense of their education in case of severe need. These findings suggest that, in the setting we study, work is (perceived by adolescents and caregivers as) a potential pathway through which underlying challenges limit the development of adolescents and their transition to a productive adulthood (a contested area of research, see for instance Edmonds, 2008). A detailed qualitative description of the activities in which adolescents engage, such as the one presented in this study, is key to understanding how this pathway operates. For instance, while some productive activities are conducive to school participation (e.g. activities that enable adolescents to pay for school-related expenditures) other activities may act as an impediment to school attendance (e.g. work to generate income for the households’ basic needs during school hours).

The findings of this study can be understood within the context of theories used in adolescent development research. For instance, within Social Cognitive Theory (SCT), human behavior is seen as the “product of the dynamic interplay of personal, behavioral, and environmental influences” (McAlister, Perry, & Parcel, 2008, p. 170). One of the key concepts of SCT is observational learning. In the context of this study, adolescents learn to perform productive activities through parental and peer modeling (Bandura, 1986).

Additionally, behaviors learned during adolescence are implemented throughout the lifecourse. Life Course Theory emphasizes the Principle of Timing, where the consequences of certain life transitions and events vary according to their timing in a person’s life, and also the Principle of Time and Place, that the life course of individuals is embedded and shaped by the context in which they live (Elder, Johnson, & Crosnoe, 2003). In this study, the adolescent participants are embedded within the context of families that are often in precarious states economically. However, though their context often causes pressure for them to engage in productive activities, we found that adolescents often responded to a perceived financial need from their families in accordance with the life course Principle of Agency. They often independently pursue financial opportunities for Ganyu (Elder et al., 2003) as seen also in other qualitative studies with adolescents on labor. For instance, Orkin (2011), who advocates for children’s perspectives in economic research, found that in rural Ethiopia children between the ages of 11 and 12 question cultural norms about their work or schooling by negotiating with their parents. Orkin (2011) also challenges the notion that children and parents either uniformly prefer that children do not work or uniformly prefer that children work, a finding reflected in our study.

The effects of policy interventions that attempt to address the challenges underlying the phenomena discussed above (e.g. poverty and adult labor constraints) may be complex and indirect. As described earlier, the quantitative impact evaluation finds that the SCTP cash enabled beneficiary households to expand their agricultural activities. Household members of all working ages (including adolescents) shift out of Ganyu into work on the expanded household farm. While the net labor supply of adolescents increases as a result, the qualitative findings presented in this paper show that working on the farm reduces the exposure to potential security and health hazards associated with Ganyu. Work on the household farm can also be more flexible and require less travel. If so, it is likely to be more compatible with school attendance. Indeed, the quantitative evidence shows strong positive impacts on broader adolescent wellbeing indicators including school participation, health, and material wellbeing despite the increase in overall adolescent economic activity (De Hoop et al., 2017; Kilburn, Handa, Angeles, Mvula, & Tsoka, 2017). The qualitative perspective provided in this paper thus improves our understanding of the effects of the SCTP by illustrating the reasons that adolescents work, and their continued desire to attend school despite a perceived responsibility to contribute economically to the household.

Interestingly, household investment in agricultural activities and a concomitant increase in child and adolescent participation in farm activities did not feature in a relevant way in the focus group discussions. Some caregivers, when probed, mentioned that the SCTP helped them expand their businesses or start new ones. However, caregivers did not mention cascading effects on the economic activities of adolescents in their households. We do not view the latter as contradictory with the quantitative results. There are many confounding factors that make it hard for interviewees to recognize such cascading effects in their own household, especially given the high prevalence of child work among poor rural households in this region. Instead, we interpret this finding as underlining the importance of combining quantitative and qualitative methods to understand how policy interventions affect beneficiaries.

We end by highlighting several limitations. Despite how conducive photovoice is to being used for our purposes, Photovoice is considered to be one story among many and cannot be generalized to all adolescents in the community. Moreover, though the photo-elicitation approach for this study is adapted from photovoice, it is important to highlight that it is subject to multiple forms of bias. For instance, though adolescents are engaged in the process of taking and describing the photos, they may change their approach based on what they think the researchers want to hear, a phenomenon that is not limited to Photovoice of course. Additionally, adolescent participants may be reticent to talk about certain topics such as hazardous work conditions or working interfering with school for fear of repercussions for themselves or their parent and/or embarrassment for missing school. Conversely, since this study does not implement the full photovoice methodology, including participant analysis, the researchers may add bias through directing what topics are talked about during the focus group discussions and/or through the analysis.