Parenting in Post-conflict Burundi: A Study on the Current Parenting Support Needs of Parents with Children Accused of Witchcraft
Marilaine A. Bareño
Student number: 10416269 Faculteit der Maatschappij- en Gedragswetenschappen Pedagogische en Onderwijskundige Wetenschappen Master’s Thesis Opvoedingsondersteuning Supervisor UvA: Dr. A. van den Akker Second assessor: Dr. H.M.W. Bos Manager EAFS-International: Eva Smallegange, MSc Date: 7 August 2014 Words: 12,869
Abstract
Many children in Burundi have been accused of witchcraft, which likely caused their parents a great deal of stress. This study examines the support needs of parents with a child accused of witchcraft. Interviews were conducted with 15 parents with a child accused of witchcraft, 15 general parents, and 18 supporters. The results show that parents with an accused child experienced parenting challenges, but there were no major differences when compared to the general parents. Supporters were also interviewed to study their perceptions of parenting needs for parents with a child accused of witchcraft. The supporters mentioned that parents of a child accused of witchcraft experienced parenting challenges. In this study, parents with a child
accused of witchcraft expressed a need for parenting support. Interestingly, supporters mentioned that this type of support is available in Burundi; however, it is possible that parents were not able to gain access to the available parenting support. There is a demand in Burundi for more
professional help in order to lessen the challenges experienced by parents.
Keywords: parenting, parenting stress, parenting support, witchcraft, witchcraft accusations, witch children
Table of Contents Page
Introduction ... 5
Witchcraft Accusations and Parenting Challenges ... 7
Witchcraft Accusations and Parenting Support ... 9
Method ... 12
Setting ... 12
Participants and Procedure ... 13
Instruments ... 16
Parenting challenges and possible sources of support ... 17
Missing variables ... 17
Coding ... 18
Parenting challenges and causes for problems...18
Received and additional support...19
Analysis... 19
Findings... 20
Witchcraft Accusations in Burundi ... 20
Parents and Parenting Challenges ...20
Primary challenges ... 21
Secondary challenges...21
Other challenges... 23
Supporters and Parenting Challenges ... 24
Primary challenges ... 24
Secondary challenges ... 24
Parents and Possible Sources of Support ... 25
Received support ... 25
Additional support ... 27
Supporters and Possible Sources of Support ... 288
Offered support ... 288
Additional support ... 28
Discussion ... 28
Parents and Parenting Challenges ... 29
Supporters and Parenting Challenges ... 322
Parents and Possible Sources of Support ... 32
Supporters and Possible Sources of Support ... 33
Limitations ... 344
Recommendations ... 37
Conclusion ... 388
References ... 39
Introduction
A relatively new phenomenon arose in Africa in the 1970s: families and community leaders began to accuse children of witchcraft (La Fontaine, 2009; Tate, 2006). There are many documented cases of children accused of witchcraft in African countries, including Burundi (Foxcroft, 2009; Schnoebelen, 2009). There has been a noticeably high rise in witchcraft accusations in post-conflict African countries, particularly against children (Foxcroft, 2009). A witch child is believed to have the ability or the power to harm others in the visible/physical world by using evil forces from the unseen spiritual world (Cimpric, 2010; Stobart, 2006). Generally, this consists of sending an illness to a family member (Cimpric, 2010) or ill fortune, such as impotence, death in the family, livestock malady, and crop failure (van Eck & van Twist, 2009). Witch children have also been blamed for marriage breakups, loss of jobs, and common misfortunes (La Fontaine, 2009).
There are several characteristics that may increase the risk of witchcraft accusations against children (Aguilar Molina, 2005; Cimpric, 2010). Researchers have examined the
common characteristics of accused children to understand what makes some children a target for witchcraft accusations while others are not. Children with physical disabilities, some types of illnesses, or certain physical characteristics are often targeted, such as children with albinism (Cimpric, 2010, Aguilar Molina, 2005) or those with swollen bellies, large heads, or red eyes (Cimpric, 2010; Malmberg Eriksson, 2012; Stobart, 2006). Children with difficult behaviour are also often targeted (Aguilar Molina, 2005). These children usually exhibit behaviours that are believed to be abnormal, such as sleep walking, bed wetting, nightmares, stubbornness, aggressiveness, laziness, and extreme curiosity, among others (Cimpric, 2010; Malmberg Eriksson, 2012; Stobart, 2006). Changes in the family structure may also spark accusations against children. For example, children who have lost both parents and are sent to live with
relatives or who have lost one parent with the other remarried may be targeted. Disagreements with the relatives or the stepparent may be cause for accusation (Cimpric, 2010). Lastly, intergenerational accusation may result in accusations of witchcraft (Aguilar Molina, 2005). Intergenerational accusation is when an accusation of witchcraft is passed from one generation to the next; it is believed to be transmitted through food or drink by an elderly person or an elder possessed with the power of witchcraft (Aguilar Molina, 2005). Accusations of witchcraft may have severe consequences. Some children accused of witchcraft have been abandoned by their families (Malmberg Eriksson, 2012); abandoned children across the sub-Saharan cities are subjected to physical and sexual abuse (Cadogan-Poole, 2013), experience unlawful violence from local authorities, and are exposed to drug and alcohol abuse (Waddington, 2006).
Accusing children of witchcraft have become an increasingly common phenomenon, primarily resulting from cultural, social, and political factors, as well as the general
impoverishment resulting from the civil war in the last decades of the twentieth century
(Cimpric, 2010). The civil war in Burundi caused by the political situation, between Hutus and Tutsis, began in 1993 and ended in 2003 (Jordans, Tol, Ndayisaba, & Komproe, 2012). Although the widespread accusations of children as witches first appeared in the 1970s, the belief in witchcraft and sorcery is very common and rooted in the African culture throughout history. Witchcraft is considered a fundamental belief system in many African countries and it is anchored in their heritage, traditions, and national identity. The civil war in Burundi led to a major loss of lives, and a large number of children lost one or both parents. Consequently, their losses put these children in a particularly vulnerable group, which may have likely led to accusation of witchcraft. The general impoverishment may have also left parents unable to provide the basic needs for their children and face financial pressure. Extreme financial pressure
may have led the community to blame children for causing this poverty by accusing them of witchcraft (Cimpric, 2010).
In addition to the civil war, the increase of Pentecostal churches may be another possible reason for witchcraft accusations against children. Numerous religious movements have grown throughout sub-Saharan Africa since the 1980s, the Pentecostal movement in particular
(Foxcroft, 2009). The large number of Pentecostal churches may have led to a massive rise of children accused of witchcraft because these churches preach the gospel of deliverance, child witchcraft, exorcism, and demonic possession (Foxcroft, 2009). The cultural factors, social factors, political factors, and poverty are all possible reasons why children of sub-Saharan Africa are regularly targeted.
Witchcraft Accusations and Parenting Challenges
Parenting is the process of raising and caring for a child from infancy to a time where the child is considered an adult (Rindfuss & Brewster, 1996). The parenting process is conducted via socialization, which refers to the process of teaching values, skills, motivations, and behaviour patterns to children in order for them to function in the culture or community in which they grow up (Grusec & Hastings, 2008). Parents with a child accused of witchcraft are expected to
experience challenges in the parenting process, such as parent-child interaction, parental control, and stimulating child development. These parenting challenges can be caused by the child’s characteristics, beliefs, and self-expectations of parents, as well as external threats that these parents face. Consequently, these parenting challenges can result in parenting stress.
The model of parenting difficulties has focused on the different ways that parents can experience stress due to negative child behaviour, parents’ own characteristics, and financial resources. Abidin’s (1993) parenting model of stress hypothesizes that a child’s characteristics are important factors of predicting parenting stress. Parenting stress can be defined as a
mismatch in the perceived demand of the parenting role and the perception of having access to the available resources to meet the demands (Deater-Deckard, 1998); however, a simple linear relation does not exist between a child’s characteristics and parenting stress. The path of influence between a child’s characteristics and parenting stress is through the parenting role, which is related to the parents’ set of beliefs and self-expectations (Abidin, 1992). Parents’ beliefs that their child will act under the force of witchcraft, that their child’s needs differ from those of non-accused children, or the feeling that they failed as parents because of the
accusations may cause parenting stress. Furthermore, parents may have low self-expectations of their parenting abilities to help their children, which may raise many doubts for them as parents on a regular basis. Parents likely fear for their own well-being and are severely distressed by the belief that their child operates under the force of witchcraft (Aguilar Molina, 2005). The external threats that families of a child accused of witchcraft may face, such as financial pressures, change in the family composition, job satisfaction, and emotional support from other family members and/or others in the community, have caused parents or guardians to magnify a child’s individual characteristics in a negative way and view these characteristics as signs of witchcraft (Aguilar Molina, 2005).
It is important to emphasize that this study distinguishes two main categories of
challenges that parents may experience in their daily lives. Challenges related to parenting were the main point of interest; however, there was also attention to challenges related to basic human needs. Burundi is an extremely poor country (Ngaruko & Nkurunziza, 2000); its population is struggling to meet basic human needs, so it was important to be sensitive about the basic needs of parents. Parents were asked about two types of challenges: the primary challenges, which are related to basic human needs, and the secondary challenges, which are more focused on
challenges that parents face: nutrition, shelter, security, safety, health, and financial resources (Harper, Harper, & Stills, 2003).
Parents may experience great challenges related to the primary challenges, but the secondary challenges were the most important to this study. Both parents and children suffer when a child is accused of witchcraft. Parents experience fear and frustration because they may not be supported by their community and must solve their problems on their own (La Fontaine, 2009). Children may suffer abuse and neglect by their parents as a result of parenting stress (Chan, 1994). The child may be humiliated, neglected, and abused, which may lead to emotional damage (La Fontaine, 2009). It is also important to mention that once a child is accused of witchcraft, he or she will live with the stigmatisation forever (Cimpric, 2010). This
discrimination and stigmatisation leads to traumas, including emotional and psychological suffering, which can lead to more tension for parents. In this study, parents with a child accused of witchcraft are expected to experience parenting challenges even though there is no empirical evidence that indicates which parenting challenges they experience in their daily lives.
In short, the secondary challenges are difficulties experienced by parents in the process where the children are taught the values, behaviour patterns, skills, and motivation needed for adequate functioning in the community in which the child grows up (Grusec & Hastings, 2008). Parents may experience secondary challenges in their parenting behaviours, parent-child
interactions, parenting competencies, stimulating child development, and their child’s characteristics.
Witchcraft Accusations and Parenting Support
According to Abidin’s (1992) parenting model of stress, parenting stress is a motivational variable which encourages and stimulates parents to make use of available resources to support their parenting. Belsky (1984) suggested that the parenting stress can possibly be buffered by
support for the parents. More specifically, parents could have less stress when they view the support as beneficial (Belsky, 1984). The main sources of support in Burundi are pastors or prophets, traditional healers, non-governmental organisations (NGO), community leaders, family members, and neighbours. In the case of children accused of witchcraft, parents can opt to seek support through one of these support systems. Parenting support is best described as all actions, activities, structures, and facilities that focus to support, optimise and/or enrich the parenting process, with the intention to provide the best chances for optimal development of children (Vandemeulebroecke, Van Crombrugge, Janssens, & Colpin, 2002).
In Burundi, pastors and prophets are believed to be able to heal children accused of witchcraft in the name of God. Community members believe these pastors and prophets are able to identify witches through their dreams and visions, after which a treatment is offered (Cimpric, 2010). Pentecostal pastors and prophets often confirm the accusation of witchcraft (Cadogan-Poole, 2013). After the affirmation, the pastors and prophets try to heal the children through exorcism or deliverance, a process that can be severely harmful to the child. The process
includes beatings, burnings, and torture. Children are also denied food and water for several days (Cadogan-Poole, 2013; Schnoebelen, 2009). Non-harmful rituals are the norm in some cases and may include food offerings, music, and dancing (van Eck & van Twist, 2009).
Aside from pastors and prophets, traditional healers or doctors may be sources of support for parents of accused children. Traditional healers are believed to be clairvoyant, meaning they are able to detect witches and are believed to have the ability to heal children accused of
witchcraft (Cimpric, 2010). They also heal both natural and witchcraft-related illnesses with medicinal plants. In one of the practices, the traditional healers use potions made of poisonous plants. These are forcefully dropped in the eyes or ears, which results in the child being cleansed. In other cases, a series of incisions is performed using a razor blade (Cimpric, 2010).
NGOs can be sources of support for families of accused children. NGOs in sub-Saharan countries include Save the Children, International Catholic Child Bureau, Child Protection Committee in Zaïre Province in Angola, and Stepping Stone Nigeria (Cimpric, 2010). These NGOs will defend children against abuse due accusations of witchcraft. They are engaged in local activities to help these children, and they provide support for their families. The NGOs work with families to change their attitudes towards children accused of witchcraft; they set up programs and centres to facilitate psychological, physical, and social rehabilitation, and they provide support for the children. They also bring together other possible sources of support, such as church leaders, local authorities, and traditional healers by facilitating sessions to raise
awareness. Furthermore, NGOs can help establish operating principles to support families of accused children (Cimpric, 2010). Community leaders such as chefs de colline and chefs de quartier are also possible sources of support (Human Rights Watch [HRW], 2010). These supporters assist the judicial officials and police with investigations, work to create a functional system for coping with crimes, and work to educate the community (HRW, 2010).
To summarize, parents can opt to heal witchcraft after a child is accused through several practices provided by pastors or prophets, traditional healers, NGOs, and community leaders. It is not clear which practices parents opt to utilize most in Burundi and what additional support these parents believe they need. The main purpose of this study is to investigate the support needs of parents with a child accused of witchcraft in Burundi through interviews with parents and supporters. These interviews will reveal their perceptions of current parenting challenges faced by parents with a child accused of witchcraft and their possible sources of support. Firstly, the parents’ perceptions of their current parenting challenges were examined. Parents were also asked to provide their perceptions of possible causes of these challenges. Secondly, the
witchcraft were examined, with the intent to investigate whether supporters’ viewpoints connect with those of the parents in order to provide the appropriate support. Thirdly, the possible sources of support that these parents received and the additional support that they would like to receive were also studied. Finally, the supporters were asked about their perceptions of possible sources of support that families with a child accused of witchcraft received and the possible additional support they believe these parents need. The collected responses answered the general question of this study: “What are parenting support needs of parents with a child accused of witchcraft in Burundi?”
Method Setting
This study was conducted in the Republic of Burundi, a country located in central Africa. Burundi has an estimated population of 6 million citizens across its 27,834 km² (Ngaruko & Nkurunziza, 2000), which makes Burundi one of the most crowded countries in Africa (Ngaruko & Nkurunziza, 2000). In 1993, the country’s first democratically elected president was
assassinated. This assassination caused violence among Burundi’s different ethnic groups and resulted in political instability in the country, ultimately leading to a civil war that lasted from 1993 to 2003. This violence negatively impacted the country’s economy and disrupted social order (Jordans et al., 2012). In 1993, 40% of Burundians lived below the poverty line. By 1998, poverty in Burundi soared to 60% (Ngaruko & Nkurunziza, 2000). The same year, the gross domestic product (GDP) per capita was approximately $123 USD, making Burundi one of the poorest countries in the world (Ngaruko & Nkurunziza, 2000).
Four provinces were included in this study: Bujumbura Mairie, Bujumbura Rural, Bubanza, and Cibitoke. The latter three provinces were selected due to their proximities to the capital Bujumbura Mairie, where this study was conducted. Furthermore, these provinces receive
support services from the Association pour la Promotion et la Protection des Droits des Enfants Marginalisés (APRODEM) Girizina association. This study was part of a larger ongoing study for developing parenting support in Burundi, which was conducted in collaboration with
APRODEM-Girizina and Education and Family Support (EAFS) International. APRODEM is a nonprofit organisation in Burundi and has expertise and experience in the field of child
protection (http://www.aprodem.bi). EAFS is a Dutch organisation which primarily conducts research and co-development of educational and family support modules (http://www.eafs-international.org; see Appendix for more details).
Participants and Procedure
The first respondent group consisted of parents with a child accused of witchcraft (eight women, seven men, Mage = 51.9 years [SD = 12.1], between 35 and 80 years of age). These parents came from four different provinces in Burundi: five from Bujumbura Rural (33.3%), five from Cibitoke (33.3%), four from Bujumbura Mairie (26.7%), and one from Bubanza (6.7%). Among these parents, 10 completed primary school (66.7%) and five had no education (33.3%). The participants had an income average of 28,500.00 Burundian franc (BIF) per month (SD = 40,653.41), which is equivalent to €13.67, and an average range of 2,000 to 100,000.00 BIF per month, which is equivalent to €0.96 to €47.96. The government-mandated minimum wage in Burundi is 4,800 BIF per month, which is equivalent to €2.30 (Minimum-Wage & Marathon Studios Enterprises, 2014). In this study, three parents did not earn any income and six had unknown income values. Parents mentioned different ways of generating income: 10 cultivators (66.7%), one businessman (6.7%), three had no job (20.1%), and one was not specified (6.7%). To be included in this study, parents must have had at least one child, boy or girl under the age of 14, who was accused of witchcraft. Only one parent, mother or father, of the same family was
interviewed. The reason for this was to include as many different families in this study as possible.
A group of general parents was also recruited in order to investigate whether parents with a child accused of witchcraft experienced more challenges than other parents in Burundi. A general parent is one whose child has not been accused of witchcraft. This respondent group consisted of eight women and seven men (Mage = 35.2 years [SD = 12.0], between 16 and 56 years of age). They came from three different provinces in Burundi: six from Cibitoke (40%), five from Bubanza (33.3%), and four from Bujumbura Mairie (26.7%). Eleven parents (73.3%) in this group completed primary school, three completed high school (20%), and one had no education (6.7%). The general parents had an average income of 68,153.85 BIF per month (SD = 52,468.19), which is equivalent to €32.69, with an income range of 4,000.00 to 210,000.00 BIF per month, equivalent to €1.92 to €100.72. One parent did earn any income and one earned an unknown income value; six parents (40.1%) were cultivators, two were businessmen (13.3%), two were transporters (13.3%), one was a baker (6.7%), one worked at the hospital (6.7%), one had no job (6.7%), and two had unknown employment (13.3%).
The third group of respondents was comprised of the supporters (three women, 15 men, Mage = 47.6 years [SD = 10.11], between 28 and 70 years of age). They came from four different provinces in Burundi: six from Cibitoke (33.3%), five from Bujumbura Mairie (27.8%), five from Bujumbura Rural (27.8%), and two from Bubanza (11.1%). The supporters’ average years of experience was 17.4 years (SD = 16.3), with a range of 2 to 49 years of experience. Nine supporters (50.0%) had worked with parents with a child accused of witchcraft and nine had no experience with this marginalized group (50.0%). The supporters consisted of eight traditional healers (44.4%), eight members of NGOs (44.4%), one had no job (5.67%), and one did not specify the job description (5.6%). Among these supporters, five completed primary school
(27.8%), two completed high school (11.1%), eight completed university (44.4%), two completed church school (11.1%), and one had no education (5.6%). To be included as a traditional healer in this study, he or she had to be recognized by the community and his or her knowledge could broadly be categorized as herbalist, diviner, spiritual or faith leaders, and/or traditional birth attendant. Secondly, members from the NGOs were from Healthnet TPO and APRODEM-Girizina.
Each research assistant was interviewed by a member of EAFS international. These assistants all studied at the university level, and had worked with both general parents and parents of marginalized groups. The assistants participated in a training course of 1 week led by an experienced researcher.
The respondents were identified through purposive sampling, where supporters of the APRODEM association or from HealthNet TPO approached the participant regarding
willingness to participate. These associations have had contact with the respondents on a regular basis in order to assist them with their needs. Prior to starting the interview, the respondents received a verbal explanation regarding the purpose of this study as well as the information sheet. Afterwards, the respondents were asked to provide verbal informed consent, which was recorded with the respondent code and agreement of participation. Due to illiteracy and fear of signing a document in the Burundian culture, a preference was made for verbal consent rather than written consent. The research assistants translated the informed consent and the information sheet into Kirundi using “The Translation Monitoring Form” of van Ommeren et al. (1999). The translation monitoring form is used to help prepare the instrument for transcultural use via a five-step translation process (van Ommeren et al., 1999). These five five-steps are (a) translation, (b) translation reviewed by a professional, (c) evaluation by local people, (d) blind back-translation, and (e) pilot testing (van Ommeren et al., 1999).
After giving informed consent, the parents and the supporters were engaged in a face-to-face interview, which lasted from 40 to 50 minutes. The interviews were conducted at the
participants’ homes, in the community office, or at the APRODEM office. One research assistant verbally conducted the interview in the Kirundi language in order to facilitate the conversations, and the research assistants recorded and later transcribed the responses. After each interview the participants received a participation fee of 5,000 BIF, which is equivalent to €2.40. The
remuneration was appropriate with local standards and organisational practices. Instruments
One questionnaire was developed for this study. There were two versions of the questionnaire: one for the parents and one for the supporters. Both were titled “Questionnaire Parenting Support in Burundi.” The interview questions were modified from Jordans, Tol, and Komproe’s (2011) “Feasibility Study for Intervention.” The interview part of “Feasibility Study for Intervention” was designed to access parents’ perceptions of their children’s psychosocial problems, the support received, and the need for support for parents in low- and middle-income countries. Therefore, this instrument was not directly applicable for data collection in this study due to the different respondent groups, so the questions that accessed parents’ perceptions of their children’s psychosocial problems were adapted to access parents’ perceptions of their parenting challenges. Questions regarding the support received and the support needed to handle the child’s problems were also adapted to address the support received and the support needed to handle parenting challenges. The adaptations of these questions were conducted after a literature study.
The questionnaires for this study collected demographic and background information. The demographic information section consisted of 18 items and collected data about the household; location of interview; respondents’ employment, age, education level; and family
composition. In addition, the section that collected the background information was composed of seven items: reasons for accusations, who made the accusations, whether the parents felt as if they were part of the community, and in general, whether or not they felt safe.
Parenting challenges and possible sources of support. Another part of these
questionnaires was designed to collect data about parenting challenges and the possible sources of support for these parents. This part consisted of five qualitative interview questions, such as “Which parenting difficulties do you face in daily life?”; “What did you do to try to solve [your] challenges when you experienced [them]?”; and “Which kind of supporter would be able to provide [you with parenting help]?” In addition, the section that accessed data about the possible sources of support was composed of three items, which collected data about the support provided to parents when experienced parenting difficulties. The questions that accessed information about the possible sources of support were “[Do you] currently [feel] supported by family?”; “[Do you] currently [feel] supported by the community?”; and “When experiencing parenting difficulties, who would you ask for help?”
The Kessler 10 (Kessler et al., 2002) was also included in this instrument. The Kessler 10 is a screening tool designed to measure populations’ psychological distress, specifically anxiety and depressive disorders (Song, de Jong, O’Hara, & Koopman, 2013). The original instrument is a five point Likert scale; however, this study used a four point scale with scores ranging between 10 and 40, with the responses Most of the time, Some of the time, A little of the time, and Never. The Cronbach’s alpha for the parents with an accused child sample in this study was .786 and for the general parents was .957.
Missing variables. Not all respondents provided data on all the variables. Of the 30 parents, 10 parents did not provide answers to two questions from the demographic section, and eight parents did not provide answers to three questions regarding their backgrounds. Of the 18
supporters, three did not provide answers to two questions from the demographic section. Regarding the Kessler 10, there were two missing questions from two of the 30 parents. The reliability tests were each conducted with one missing variable. All respondents provided data on all other variables.
Coding
Data collection during this study consisted of one questionnaire and one interview
transcript for each participant. The interviews were analysed following thematic content analysis using a code system. Categorization of responses was set in a code system, which was based on previous finding of Jordans et al. (2011). The latter assessed parental burdens and needs in Burundi (Jordans et al., 2011). The code system consisted of four main categories: parenting challenges, causes for problems, support received to improve the situation, and additional support to improve the situation.
Parenting challenges and causes for problems. The parenting challenges had three subcategories in this code system; primary challenges, secondary challenges, and other
challenges present in the family. The primary challenges are the basic human needs derived from the theory of Maslow: nutrition, shelter, security, and safety (Harper et al., 2003). The secondary challenges are those experienced by parents in the process in which the children are taught the values, behaviour patterns, skills, and motivation needed for adequate functioning in the
community in which the child grows up (Grusec & Hastings, 2008). These secondary challenges were divided into three categories: the child’s characteristics, the parent’s characteristics, and parenting. In addition to the primary and secondary challenges, a third category was developed which consisted of challenges that could be categorized neither as primary nor secondary challenges. These challenges included other possible challenges present in the family:
and discrimination, among others. Finally, the code system also consisted of codes through which to examine the causes of problems. The cause of problems category was a duplication of the parenting challenges codes.
Received and additional support. The code system also consisted of codes through which to examine the possible sources of support for parents’ daily parenting challenges. In this section, a distinction was made between the support that parents received and the additional support that parents would like to receive. Both sets of codes were based on the five capitals of resilience: financial capital, physical capital, natural capital, social capital, and human capital (Obrist, Pfeiffer, & Henley, 2010). Financial capital is the financial resources used or needed to reach life objectives (Department for International Development [DFID], 1999). Physical capital refers to the basic infrastructure needed to support a human being, such as secure shelter and buildings, sanitation, and adequate water supply. Natural capital is the natural resources from which useful help are derived, such as land, trees, water, and wild resources. Social capital refers to the social resources used to achieve life’s objectives, such as group meetings and counselling. Finally, human capital refers to knowledge, skills, and health needed to pursue and achieve life’s objectives, such as health, education, and knowledge (DFID, 1999).
Analysis
A different approach was used for both data sets to examine the collected quantitative data and the qualitative data. First, the descriptive statistics and frequencies in SPSS were used for respondents’ demographic and background descriptions. Second, the parenting challenges and the possible sources of support for parents were examined in MAXQDA in two steps: the previous mentioned code system was created and used for coding, and new variables were added to the code system as they were discovered in the transcription. Two coders conducted nearly one third of the coding and one coder conducted the remaining coding. The average intercoder
agreement for the group of general parents was 92% for code existence in the documents and 91% for code frequencies in the documents. The minimum acceptable level of agreement is 90% (Miles & Huberman, 1994).
Findings Witchcraft Accusations in Burundi
In the process of locating children accused of witchcraft, the researchers discovered that the parents have been accused of witchcraft and it is believed that the children have inherited this curse from their parents. In this study most of the recruited parents with an accused child had also been accused of witchcraft (n = 14). Most parents (n = 11) were accused by their
neighbours, several parents were accused by a family member (n = 2), and the other parents had been accused by friends or extended family. The accusations had an average duration of 5.7 years (SD = 3.9), with durations ranging from 3 months to 13 years and 3 months. Initial
accusations were made at an average of 8.7 years (SD = 4.9) prior the date of the interview, with a range from 6 months to 20 years and 3 months.
Some questions were created to collect respondents’ background information regarding the consequences of the accusations. Most parents (n = 11) with a child accused of witchcraft mentioned that they felt they were part of the community; however, the majority of parents (12 out of 15) felt stigmatised. Half of the parents with a child accused of witchcraft (n = 7)
mentioned that they experienced negative consequences because of the accusations, and more than half of the parents (n = 10) expressed that they felt less capable of taking care of all their children due to the accusations. Even though parents experienced negative consequences, more than half of those with a child accused of witchcraft (n = 9) mentioned that they felt safe. Parents and Parenting Challenges
Primary challenges. In this study, parents with a child accused of witchcraft were compared to a group of general parents in order to investigate whether parents with a child accused of witchcraft experienced more challenges than other parents in Burundi. All parents (n = 15) with a child accused of witchcraft and the general parents (n = 15) expressed that they experienced primary challenges in their daily lives. The primary challenges most experienced by the parents with an accused child were nutrition (n = 11) and financial resources (n = 8); the general parents also experienced challenges of nutrition (n = 7) and financial resources (n = 11). One parent from the marginalized group said, “The first challenge is having little food for my children.” Another respondent said, “It is not easy to raise a child while we do not have enough food.” Concerning financial resources, one parent said, “The main problem that I have is
thinking how my husband will find a job.” Another respondent said, “Sometimes my children get sick and I do not have the resources to pay for the treatment.”
Secondary challenges. The majority of the parents with a child accused of witchcraft (14 out of 15 parents) faced secondary challenges, and all parents from the general group also
experienced secondary challenges. Most parents from the marginalized group experienced challenges regarding the characteristics of their children (14 out of 15 parents), which was similar compared to the general parents. The most commonly mentioned challenges regarding the child’s characteristics for the parents with a child accused of witchcraft were disobedient behaviour (n = 12), impoliteness (n = 10), aggressive behaviour (n = 9), stealing (n = 8), and emotional problems (n = 8). Half of the parents (n = 7) mentioned the difficult behaviour of the child, laziness (n = 7), and six parents (out of 15) mentioned vagabondage as a challenge. The general parents experienced fewer challenges than the parents with a child accused of witchcraft regarding their children’s disobedient behaviour (n = 9), impoliteness (n = 2), aggressive
behaviour (n = 4), stealing (n = 3), emotional problems (n = 3), difficult behaviour (n = 9), laziness (n = 1), and vagabondage (n = 5).
It is notable that a few parents with a child accused of witchcraft mentioned lying (n = 2), prostitution (n = 3), illness (n = 2), alcoholism (n = 3), and psychosocial well-being (n = 3) as challenges. Likewise, the general parents experienced fewer challenges regarding prostitution (n = 1), lying (n = 2), illness (n = 3), alcoholism (n = 1), and psychosocial well-being (n = 4). Furthermore, parents with a child accused of witchcraft did not mention drug abuse, physical disabilities, or child’s temperament as challenges; the general parents only mentioned drug abuse (n = 3) as a challenge.
More than half of the parents with a child accused of witchcraft (nine out of 15) faced difficulties regarding aspects of their own characters, whereas fewer general parents (five out of 15) experienced challenges with aspects of their own characters. Furthermore, parents with a child accused of witchcraft mentioned parental well-being (n = 9) and alcoholism of the caregiver (n = 5) as challenges regarding parents’ characteristics. The general parents only mentioned parental well-being (n = 3).
Some aspects related to parenting were mentioned as challenges face by parents with a child accused of witchcraft. However, there was barely a difference between the parents with an accused child (n = 7) and the general parents (n = 8). Furthermore, less than half of the parents with a child accused of witchcraft (four out of 15) mentioned having difficulties in their parent– child interaction; this was similar for the general parents (n = 4). Only one parent with a child accused of witchcraft mentioned having difficulties stimulating child development, while four general parents mentioned this challenge. Also, two (out of 15) parents with an accused child mentioned having difficulties in their parent–child relationships, compared to only one (out of
15) general parent. In addition, several parents with an accused child mentioned they
experienced problems regarding their parenting competence (n = 1) and parental control (n = 2). Other challenges. Although the focus in this study was on secondary challenges, parents mentioned other difficulties that they experienced in their daily lives that could not be
categorized either as primary challenges or secondary challenges. More than half of the parents with an accused child (nine out of 15) experienced difficulties with their child’s education, compared to six general parents (out of 15). One respondent said, “I have a child who does not want to go to school.” Several parents with a child accused of witchcraft (five out of 15) mentioned that their children dropped out of school. As for the general parents, this was mentioned only once. One respondent said, “What hurts me the most is that my child left
school.” Another respondent said, “My son abandoned school when he was in fifth grade.” Less than half of the parents with a child accused of witchcraft (n = 5) mentioned stigmatisation and discrimination as challenges, which none of the general parents indicated as a challenge. Several parents with a child accused of witchcraft mentioned teen pregnancy (n = 4), parental disputes (n = 2), and absence of caregivers (n = 2) as a challenge. Teen pregnancy (n = 1), parental disputes (n = 3), and absence of caregivers (n = 2) were mentioned several times by general parents as well. One parent with a child accused of witchcraft mentioned lack of social support. None of the general parents mentioned lack of social support as a challenge.
The average scores of the Kessler 10 for the two different respondent groups were compared. Parents with a child accused of witchcraft had an average of 25.64 (SD = 6.98) with scores ranging from 14.00 to 37.00; the general parents had an average of 28.21 (SD = 9.93) with scores ranging from 15.00 to 40.00.
The majority of the parents with accused children (nine out of 15) stressed that most of their problems were related to the accusations against them. Others mentioned lack of nutrition
(six out of 15) and lack of financial resources (six out of 15). The general parents (four out of 15) mentioned financial resources as the cause for their problems; lack of nutrition was not
mentioned as a cause.
Supporters and Parenting Challenges
Primary challenges. The supporters (13 out of 18) stated that parents in Burundi faced primary challenges to a great extent in their daily lives. Supporters mentioned nutrition (n = 8) and financial resources (n = 10) as the two main primary difficulties that parents encountered. The supporters (n = 3) also mentioned that parents with a child accused of witchcraft faced difficulties with security and safety.
Secondary challenges. All supporters (n = 18) mentioned that parents faced secondary challenges in their daily lives. The majority of the supporters (16 out of 18) suggested that parents had difficulties regarding their children’s characteristics. The children’s characteristics which clearly stood out to supporters were impoliteness (n = 10), vagabondage (n = 10), difficult behaviour (n = 15), stealing (n = 9), disobedient behaviour (n = 9), and psychosocial well-being (n = 8). Several supporters mentioned laziness (n = 5), aggressive behaviour (n = 4), alcoholism (n = 4), drug abuse (n = 3), and prostitution (n = 3). A few supporters also mentioned that parents with a child accused of witchcraft specifically experienced problems with the impoliteness (n = 1) and psychosocial well-being (n = 5).
From the parents’ characteristics, several supporters mentioned that parents experienced challenges with alcoholism (n = 2), difficult behaviour (n = 2), and parental well-being (n = 2). In addition, one supporter (out of 18) mentioned that parents with a child accused of witchcraft experienced challenges regarding their own well-being.
The majority of the supporters (16 out of 18) mentioned that parents face challenges in their parenting. The conspicuous parenting challenges were stimulating child development
(n = 13), parent–child interaction (n = 10), parental control (n = 5), and parenting competence (n = 5). Supporters also mentioned that the group of parents with a child accused of witchcraft faced difficulties stimulating child development (n = 1), parenting competence (n = 1), and parental control (n = 1).
Other challenges. Although the focus in this study was on secondary challenges, most supporters (n = 17) also mentioned other difficulties that parents experienced in their daily lives that could neither be categorized as primary challenges nor secondary challenges. Most
supporters (n = 11) mentioned that parents had difficulties with children dropping out of school. Eight supporters (out of 18) stated that parents experienced some problems related to the
education of their children, such as parents unable to pay school fees or some children skipping classes. Supporters also mentioned teen pregnancy (n = 5), children who were financially responsible for themselves (n = 4), lack of money to attend school (n = 4), polygamy (n = 3), absence of caretakers (n = 3), and lack of social support (n = 3) as challenges. Supporters mentioned that parents with a child accused of witchcraft were more likely to experience a greater significance of stigmatisation and discrimination (n = 9) within the community, and that their children dropped out of school (n = 4) more frequently than parents without accused children. Other challenges mentioned included conflicts concerning land (n = 1), educational problems (n = 3), and lack of social support (n = 1).
Parents and Possible Sources of Support
Received support. Most parents with a child accused of witchcraft and the general parents appeared committed to overcoming the challenges that they experienced in their daily lives. Only five parents (out of 15) with a child accused of witchcraft mentioned that they did not seek support when they experienced parenting difficulties, compared to the general parents, of whom only one parent (out of 15) did not seek support. Almost half of the parents (eight out of
15) with a child accused of witchcraft mentioned that they sought support from their families, while six general parents (out of 15) mentioned that they sought support from their families. In the background section of these interviews, parents were asked specifically if they sought
support from a family member. Different answers were reported. Three parents (out of 15) with a child accused of witchcraft mentioned that they sought support from family, compared to nine of the general parents (out of 15). Parents were also asked if they felt supported by their families. Half of the parents with an accused child (n = 8) mentioned that they did not feel supported by their families, compared to five (out of 15) general parents. Furthermore, five parents with an accused child (out of 15) mentioned that they felt supported by their families, whereas two general parents mentioned that they felt supported by their families.
Participants were asked if they sought support within the community. More than half of the parents (n = 9) with a child accused of witchcraft mentioned that they sought support within the community. Parents of accused children expressed that they sought support from their neighbours (n = 2), friends (n = 1), community leaders (n = 3), and church leaders (n =1); none of the parents sought support from the NGOs or from a traditional healer. The general parents (n = 7) mostly used the same sources of support within the community as the parents with a child accused of witchcraft; these were neighbours (n = 2), friends (n = 2), community leaders (n = 1), and church leaders (n = 1). None of the general parents sought support from a traditional healer. In the background section of this interview, parents from both respondent groups were asked specifically if they sought support from a community member. Different answers were reported. Parents with a child accused of witchcraft mentioned that they sought support from neighbours (n = 3), friends (n = 2), community leaders (n = 2), NGOs (n = 2), and church leaders (n =1). It is notable that neither respondent group mentioned seeking support from a traditional healer.
Parents were also asked if they felt supported by the community. Several parents with an accused child (n = 5) mentioned that they did not feel supported by the community; of the general parents, four did not feel supported. Less than half of the parents with a child accused of witchcraft (n = 5) mentioned that they felt supported by the community compared to a few general parents (n = 2). Most parents defined community as neighbours: 11 parents with a child accused of witchcraft and 13 general parents.
Remarkably, the majority of parents with a child accused of witchcraft (n = 13) were not supported by a professional organisation. Parents with a child accused of witchcraft reported that their two main supports were human capital support (three out of 15 parents) and social capital support (four out of 15 parents). The general parents received the same type of support as the parents with a child accused of witchcraft: human capital support (three out of 14) and social capital support (six out of 15).
Additional support. All parents with a child accused of witchcraft requested support to address their secondary needs; the general parents mentioned similar support needs. Parents with an accused child asked for social capital support (n = 10). One respondent said, “Help me to advise my children so that they can give up their bad behaviour.” Another respondent said, “The support that I need is counselling.” Parents with a child accused of witchcraft also asked for human capital support (n = 9). One respondent said, “The support that we need is to show us how to raise our children.” Another respondent said, “We need teaching on how to educate our children.” Similarly, the general parents requested social capital (n = 9) support and human capital (n = 12) support. Parents with a child accused of witchcraft (n = 7) also requested financial support, as did the general parents (n = 6).
Supporters and Possible Sources of Support
Offered support. Almost every supporter (17 out of 18) stated that social capital support was offered to parents. One supporter said, “We advise the father to support his family at home.” Another supporter said, “We conduct group sessions and home visits to show parents the
[negative] impact of polygamy,” More than half of the supporters (12 out of 18) mentioned that human capital support was offered to the parents. One supporter said, “Parents are taught about children’s rights.” Another supporter said, “Sometimes I give children traditional medicine and this helps them to regain their confidence.”
Additional support. According to the majority of the supporter (15 out of 18), there is a need for additional secondary support to parents. Furthermore, most supporters suggested social capital support (13 out of 18) as an additional support. One supporter said, “The first step to support parents is to advise them.” Another supporter said, “There is a need to reinforce parents’ ability.” Also, more than half of the supporters suggested human capital support (11 out of 15) as an additional support to parents, respondents said: “Parents need knowledge to resolve the problems with their children”; “Parents need training on parenting.”
Discussion
As outlined previously, the purpose of this study was to examine parenting support needs of parents with a child accused of witchcraft. In short, this study showed that parents with a child accused of witchcraft experienced primary challenges and secondary challenges in their daily lives. Concerning the secondary challenges, parents with an accused child experienced more challenges regarding the characteristics of their children than challenges related to their own characteristics and to their parenting. The supporters’ perceptions of parenting challenges faced by parents with a child accused of witchcraft were similar to the parents’ perceptions. Supporters mentioned that parents experienced both primary challenges and secondary challenges. With
reference to the secondary challenges, supporters mentioned that parents’ most experienced challenges related to their children’s characteristics and challenges related to their parenting. With regard to the possible sources of support, parents with a child accused of witchcraft
mentioned that they did not receive professional secondary support and expressed a need for this type of support. Remarkably, the supporters mentioned that secondary support is available in Burundi. It is possible that parents were not able to gain access to the available secondary support.
Parents and Parenting Challenges
The first aim of this study was to examine the perceptions of parents with a child accused of witchcraft in their daily parenting challenges. The two main categories of challenges were primary challenges and secondary challenges. Although the focus of this study was on secondary challenges, parents also mentioned that they experienced primary challenges. Concerning the primary challenges, the findings show that parents experienced significant challenges related to lack of nutrition and financial resources. Parenthood requires full responsibility of an infant who is fully dependant on his or her caregiver (Deater-Deckard, 1998). In other words, children demand comfort, food, and attention, which can be stressful to parents. Parenting can present new challenges, such as psychological distress that results from the demand of parenting (Deater-Deckard, 1998). Additionally, parents’ income was also associated with parenting stress
(Baldwin, Brown, & Milan, 1995). Children rely on their parents to satisfy their basic needs and parents in this study were in positions where providing these basic needs was difficult. This difficult situation may cause heavy parental burdens and stress; consequently, parents may experience challenges in their parenting. Further research is needed in order to better understand whether primary challenges lead to challenges in parenting a child accused of witchcraft and how this phenomenon occurs.
In addition to the primary challenges, other studies show that nutritional disadvantage is more likely to exist in impoverished settings (Grantham-McGregor, Walker, & Chang, 2000). This finding was also consistent with the findings in this study, which was executed in one of the poorest countries in the world (Ngaruko & Nkurunziza, 2000) where the most commonly
mentioned primary challenges were nutrition and financial resources. Low social economic status brings with it feelings of low self-esteem, powerlessness, and helplessness (Bradley & Corwyn, 2002). Parents in this study felt powerless and helpless without food and financial resources, and they were desperate for solutions to satisfy their basic needs, which could be a reason that primary challenges were the first problems that parents in this study mentioned when asked about challenges they face in their daily lives. The difficult conditions in which parents with a child accused of witchcraft live presented barriers to providing proper care for their children.
Although parents mentioned that they experienced a significant number of primary challenges, the main concern in this study was the secondary challenges. Given that the majority of parents in this study mentioned that they experienced secondary challenges in addition to primary challenges demonstrates that these parents experienced great challenges in their daily lives, which could be an important viewpoint for future support offered to these parents. Parents who have experience difficulties may benefit more from the support they receive than parents with low initial level of problems (Hautmann et al., 2010). Furthermore, parents who have experience difficulties may be more motivated to seek support for their parenting than parents with no difficulties (Abidin, 1992). That the parents interviewed for this study experienced primary and secondary challenges also suggests that parents need support in the near future, as the greater number of challenges parents experience, the more authoritarian discipline they will
use (Deater-Deckard & Scarr, 1996). Furthermore, the authoritarian discipline that parents use is related to the misbehaviour of the child (Deater-Deckard & Scarr, 1996).
In addition to the secondary challenges, parents in this study mentioned that they
experienced more difficulties concerning the characteristics of their children than with aspects of their own characters and of their parenting. The children’s characteristics identified were
disobedient behaviour, impoliteness, aggressive and difficult behaviour, stealing, laziness, emotional problems, and vagabondage. In addition, parents also mentioned that they felt less capable due to the accusations of witchcraft. These findings are consistent with the parenting model of stress, which hypothesizes that the child’s characteristics and the parenting role predicts parenting stress that parents may experience in their daily lives (Abidin, 1992). Furthermore, parenting can be more of a challenge for parents who feel less competent and have less knowledge (Mash & Johnston, 1990). It seems that when parents experienced challenges with their children’s characteristics alongside feelings of incompetence and lack of skills, they experience parenting stress. This parenting stress may have a negative effect on their parenting, as parents who experienced challenges regarding their children’s characteristics also expressed a need for secondary support.
Although most parents with a child accused of witchcraft felt that they were a part of their communities, most parents felt stigmatised and some felt negative consequences as a result of these accusations. People who are stigmatised live in fear and face negative consequences in their environments (Corrigan & Watson, 2002). It may be possible that the fear parents
experienced due to the stigmatisation may not be overruled by feelings of belonging, as it did not help reduce the parenting challenges they experienced. In other words, when parents felt
stigmatised, while still feeling as if they were parts of the communities, they still experienced parenting challenges. Further research is needed to explain this discrepancy.
Supporters and Parenting Challenges
The second aim of this study was to examine the supporters’ perceptions of parenting challenges that parents with a child accused of witchcraft experienced. Supporters identified almost the same challenges as those that were mentioned by parents with a child accused of witchcraft. Regarding the primary challenges, supporters mentioned that parents experienced challenges concerning lack of nutrition and financial resources, and several supporters mentioned security and safety. Regarding the secondary challenges, supporters noted that parents
experienced challenges with their children’s characteristics and faced challenges in their parenting. Based on these findings, it seems that parents blamed their children as a means of defending their parenting abilities (Tennen & Affleck, 1990), as blaming their children may help parents to retain their positive self-images (Snyder & Higgins, 1988). Parents may have failed to mention their own mistakes in their parenting in order to maintain their positive self-image; conversely, supporters were more objective about parents’ challenges than the parents themselves. Further research is needed in order to learn more about this discrepancy. Parents and Possible Sources of Support
The third aim of this study was to examine the perceptions of parents with a child accused of witchcraft regarding possible sources of support for their parenting challenges. Furthermore, parents were asked about the support they received and support that they would like to receive. Parents mentioned that they did not receive professional support; however, half of the parents with an accused child had sought support from a family member, and more than half of the parents with a child accused of witchcraft sought support from the community. Even though parents sought support in their environments, some parents expressed that they did not feel supported by their families or their communities. In the parenting model of stress, parenting stress is a motivational variable which encourages parents to use available resources (Abidin,
1992); parents can experience less stress if the support is seen as beneficial (Belsky, 1984). Although parents in this study sought the support of a family member or the community, the support received did not help lessen the challenges they experienced, as parents in this study still mentioned that they did not feel supported by their environments. In other words, parents did not view the support they received as beneficial. A study showed that parenting can be more of a challenge for parents who lack instrumental and emotional support from their communities, friends, and families (Mash & Johnston, 1990). That parents sought support but did not feel supported shows that the instrumental and/or emotional support they received from families and their communities was insufficient to help them cope with the parenting challenges they faced; these parents required more professional support.
Parents with a child accused of witchcraft expressed the need for additional secondary support, especially human capital and social capital support. The fact that parents asked for additional support illustrates that they are motivated to make changes in their parenting. Parents who are motivated to change will be able to access support faster than those who are not
motivated to change. Parental motivation and willingness to receive support may result in positive outcomes from the support they receive (Baydar, Reid, & Webster-Stratton, 2003). In other words, when parents receive the support that they asked for, they will probably experience fewer secondary challenges in their daily lives.
Supporters and Possible Sources of Support
The fourth aim of this study was to examine the perceptions of supporters regarding the support offered to parents for their daily parenting challenges and the additional support that the supporters believe these parents need. The supporters mentioned that secondary support is available in Burundi, but parents mentioned that they would like to receive additional secondary support. In other words, parents do not seem to be able to find the support that they asked for. A
study by Deater-Deckard and Scarr (1996) showed that parents experience parenting stress when there is a discrepancy between the perceived demand of the parenting role and the perceived availability of resources to meet these demands. Furthermore, there is a discrepancy between the demands of parents with a child accused of witchcraft and access to the available resources to meet those demands. A possible explanation for the lack of access to secondary support is that the main concern of parents is to provide the basic needs for their families and not to address their parenting challenges. The basic human needs may have caused the parents to place all their energy and time to locate primary support rather than secondary support. Another possible explanation is that parents are not aware of how to approach or where to find secondary support. Further research is needed on the lack of accessibility to the secondary support in order to better understand the reasons that parents are not yet accessing the support that they asked for, while this support is being offered in Burundi.
Supporters mentioned that there is a need for additional secondary support for parents with a child accused of witchcraft. Furthermore, supporters mentioned that parents need
additional social capital and human capital support. Even though secondary support is available for parents in Burundi, it seems that the available support is not sufficient to satisfy the current needs of parents. In other words, the needs of parents in Burundi are greater than the available support to meet those needs. It is evident that there is a demand for more professional help in Burundi in order to lessen the challenges that parents experience.
Limitations
This study is the first to examine parenting support needs for parents with a child accused of witchcraft in Burundi, which provides knowledge of the parenting needs for parents with a child accused of witchcraft. There are several limitations that should be taken into consideration when interpreting the results. In the process of locating parents of children accused of witchcraft,
researchers found that the parents themselves had also been accused of witchcraft. Parents who have been accused of witchcraft may experience challenges different than parents whose children have been accused of witchcraft but they have not. It may be possible that the findings in this study are based on a target group other than the one desired, as parents could have mentioned challenges from the perspective that they were the ones accused of witchcraft and not from the perspective of parents who have a child accused of witchcraft. Even though there is a possibility that these findings are based on another group, researchers chose to continue with this study because the intergenerational accusation states that when parents are accused of witchcraft their children will likely be accused as well. Nevertheless, this was a substantive finding which requires more research in order to better understand the intergenerational accusations and the reasons for these accusations against these parents.
There is a high level of poverty in Burundi (Jordans et al., 2011). People in Burundi have developed a mind-set that when they see NGOs working on a project in the community, they expect to receive money or some kind of help. This study attracted people while it was being conducted. Some parents and supporters who came to take part in this research did not meet the criteria to be included. These people came without invitations because they expected to receive support in exchange for participation. To detect these uninvited people, members of the NGOs had to identify the respondents they personally invited prior to the interviews. Although the uninvited people were detected prior to the interviews, it may be possible that some parents participated in the interviews without meeting the criteria and provided socially desirable responses to meet the interviewer’s expectations. Even though face-to-face interviews decrease the tendency to give an invalid answer to questions, it creates a high risk to provide socially desirable answers to meet the research expectations (Hoyle, Harris, & Judd, 2002). In order to
prevent socially desirable answers, the research assistants were attentive to the respondents’ facial expressions and their body language during the interviews.
Another limitation in this study was the translation work from Kirundi to English.
Translations were not conducted with van Ommeren et al.’s “The Translation Monitoring Form” (1999). The translation work was needed in order to examine the collected data. It is not clear whether the translated responses were interpreted in the manner they were expressed in Kirundi. Although the interviews were not translated with “The Translation Monitoring Form” (van Ommeren et al., 1999), experienced translators translated these interviews, which validates and makes the translations reliable.
The limitation regarding the study size is also important to mention. This study was conducted on a small sample size of parents with a child accused of witchcraft in four provinces of Burundi; as a result, generalization is likely difficult in the other provinces in Burundi and other countries (Marshall, 1996). It is also difficult to generalize these results to parents with a first generation child accused of witchcraft. Future research may focus on parents with a first generation child accused of witchcraft. Furthermore, this study collected some quantitative data, but it was not possible to inspect statistical differences among groups due to lack of power to detect even small differences. Therefore, this study chose to not impute the missing values.
Another shortcoming in this study was the slightly different answers given to questions that were very similar in both the quantitative part and the qualitative part of the instrument. In the quantitative part, parents were asked if they sought support from a family member, the community leaders, neighbours, or the traditional healer when they experienced parenting difficulties. In the qualitative part, parents were asked what they did to solve their problems when they experienced parenting difficulties. A possible explanation for this discrepancy is that the questions were asked in a different format. In the quantitative part, the question was a closed
question with multiple answers. In the qualitative part the question was open (“What did you do to try to solve the challenge?”). Finally, as mentioned in the analysis part of this study, less than one third of all transcripts were double coded. Even though the average intercoder agreements for the double-coded documents were above the acceptable level of agreement, a reliability check would be more accurate if all the transcripts were coded by two coders (Miles & Huberman, 1994).
Recommendations
The majority of the parents with a child accused of witchcraft in this study mentioned that they faced primary and secondary challenges, both of which seem to be related. Other studies show that poverty may diminish parental responsiveness (Bradley & Corwyn, 2002), which may lead to disruptive behaviour in a child (Bradley, Corwyn, Burchinal, McAdoo, & Garcia-Coll, 2001). This study also shows that the majority of the parents who mentioned lack of financial resources also mentioned that they faced secondary challenges.
It would be ideal to provide support for these parents which help solve both primary and secondary challenges. Other studies have shown that parent training was less effective for
families with financial disadvantages (Lundahl, Risser, & Lovejoy, 2006). Therefore, if the basic needs of parents are covered, there is a possibility that they can focus more on their parenting (Leijten, Raaijmakers, Orobio de Castro, & Matthys, 2013). However, if the basic needs of parents cannot be satisfied when they receive ideal support, the possibility still exists that they may achieve positive outcomes from individual parenting training. Families with financial disadvantages benefit more from individual training compared to group training (Lundahl et al., 2006). Thus, future support for parents with a child accused of witchcraft should be a
