Secondary Logo

Journal Logo


Exploring the Cinderella myth: intrahousehold differences in child wellbeing between orphans and non-orphans in Amajuba District, South Africa

Parikh, Anokhia; DeSilva, Mary Bachmanb; Cakwe, Mandisaa; Quinlan, Tima; Simon, Jonathon Lb; Skalicky, Anneb; Zhuwau, Toma

Author Information
doi: 10.1097/01.aids.0000300540.12849.86
  • Free



With an antenatal seroprevalence of 40.7% in 2005, KwaZulu-Natal has the highest HIV prevalence of any province in South Africa, a country with 5.1 million individuals infected [1,2]. AIDS-related mortality is high, and the consequent impact on orphaning is likely to be dramatic in the years to come, irrespective of the expansion of the antiretroviral programme. There are currently 1.2 million AIDS orphans in South Africa [2], and this number is expected to peak at a staggering 2.3 million in 2015 [3]. As such, it is no surprise that the issue of orphaning has attracted significant attention.

Much of this attention has characterized orphans as children who are growing up without the care and support of their families, who have poorer learning and knowledge levels, and who are suffering from the ‘absence of adults in their socialization’ [4]. The data reveal that most orphans, defined in the AIDS literature as having one or more parents dead, in sub-Saharan Africa have at least one parent living and either live with their surviving parent or are absorbed into other families where they have some adult supervision.

Nevertheless, the death of a parent (and the potentially long illness preceding it if the parent has died of AIDS) may have various impacts on a child's wellbeing. Numerous studies have accordingly examined the impact of orphaning on children, but have focused primarily on educational outcomes (commonly defined by school enrolment), with few studies looking at any other aspect of wellbeing such as health or labour outcomes.

Although the empirical evidence is mixed and depends on the location, data sources, and methods used, two broad themes emerge in the literature: that the impact of orphaning depends often on which parent dies and that income is often a greater predictor of outcomes than orphan status. Case et al. [5] used the Demographic and Health Survey data from 19 countries in Africa and found systematic differences in school attendance between orphans and non-orphans. The finding of difference in attendance and enrolment between orphans and non-orphans has been supported by other studies in rural Kenya through longitudinal studies [6,7], and in cross-sectional studies elsewhere in sub-Saharan Africa [8]. Beegle et al. [9], using a panel that follows children for 13 years, showed that maternal orphanhood was associated with lower educational attainment and health (as measured by height) in the long term, and paternal orphanhood was associated with lower educational attainment for certain groups only. Using Demographic and Health Survey data and other nationally representative household surveys from 51 countries in Africa, Ainsworth and Filmer [10] highlighted the variation in orphan/non-orphan differentials across countries and found that income plays a greater role in determining school enrolment than orphaning.

A lack of statistically significant differences in enrolment have been found by other researchers conducting longitudinal studies in East Africa [11–13]. Chatterji et al. [12], using longitudinal data from Rwanda, showed no differences in school enrolment and food intake between orphans and non-orphans. Adato et al. [14] found no statistically significant differences in schooling indicators but qualitatively did find some cases of discrimination towards orphans within the household.

Many of these studies consider orphans and non-orphans in general, but do not distinguish between orphans who live with other orphans and orphans who live with non-orphans in mixed households (with some exceptions) [6,7,15]. Moreover, they are an important type of household, as according to the Statistics South Africa's national 2003 General Household Survey, approximately two-thirds of all orphans in KwaZulu-Natal live in such households. These households are of particular interest also because the co-resident non-orphans make a natural comparison group for orphans as they both share household characteristics (e.g. household income, literacy of caregiver etc.).

In one of the few studies to make intrahousehold comparisons, Case and Ardington [15] showed that in Hlabisa district in KwaZulu-Natal, maternal orphans were more likely to be behind in school and had less spent on their schooling, but were equally likely to be enrolled as the non-orphans with whom they lived. Implicit in the finding that orphans are worse off when compared with other children is the notion that caregivers may prioritize their own children over the fostered child, a Cinderella approach, so to speak. For example, when facing income constraints, a caregiver might spend less on the fostered child than on his/her own child (although perhaps with less malice than Cinderella's stepmother). Gender bias in educational expenditures has been well documented in Asia; a similar bias could also apply to orphans, but very little research has examined the intrahousehold aspect of this question to date.

In summary, the research to date is limited in two ways. First, few studies explicitly assess intrahousehold differences between orphans and non-orphans. Second, both intra and interhousehold comparisons have used limited indicators such as school enrolment, grade progression, and height, and thus do not provide a comprehensive picture of other important aspects of child wellbeing.

To respond to these shortcomings, this paper compares the differences in wellbeing between orphans and non-orphans who live with each other using longitudinal data from 2004–2006 from Amajuba District in KwaZulu-Natal. This paper addresses the question ‘do orphans and non-orphans living within the same household fare differently in terms of wellbeing?’ We define ‘wellbeing’ to include education, health, and labour outcomes. We conclude with potential explanations for the observed results and trends.


Study area

The Amajuba district was chosen because it included a broad cross-section of urban, peri-urban and rural areas. The district has approximately 470 000 inhabitants and is poor. The economy used to be driven by the coal mining industry, but the closure of coal mines has led to high unemployment in the region and consequently high rates of migration. Additional details are provided elsewhere (Bachman DeSilva et al., manuscript in preparation).

Data and sample selection

The data come from the first 2 years of an ongoing 3-year cohort study that commenced in 2004. The study was designed, and the data collected by the Amajuba Child Health and Well-being Research Project, a joint initiative between the Health Economics and HIV/AIDS Research Division (HEARD) at the University of KwaZulu-Natal and the Center for International Health and Development at the Boston University School of Public Health. The Institutional Review Board of Boston University Medical Center and the Ethics Committee of the University of KwaZulu-Natal provided ethical approval for the study.

The annual survey has four components: a household and demographic information questionnaire administered to the household; a questionnaire for the primary caregiver of the study child; and two questionnaires administered to the study child, one on general wellbeing including self-reported health, educational attainment, and the use of time, and a second that assesses the self-reported psychosocial wellbeing of children.

Sample selection took place using randomized stratified cluster sampling from 60 of 252 schools in the district. The study population were predominantly Zulu-speaking children aged 9–16 years, resident in the district and attending school at the time of sampling. Only ‘recent orphans’, defined as children who had lost one or both parents to any cause during a 6-month period between March and August 2004 were included. This was done in order to capture the incidence of orphaning and measure the exposure period of parent death. Three comparison non-orphan children were randomly selected from the same school, grade, and age as index-orphan children. In households in which there were both orphan and non-orphan children, a secondary comparison child was selected in the same age range as that household's primary study child. A cohort of secondary comparison children was thus also constructed in order to investigate intrahousehold differences in children's wellbeing.

Study households were classified into three groups: orphan-only, non-orphan-only, and mixed households. The overall baseline sample includes 50 orphans from the 50 orphan-only households, 377 non-orphans from 377 non-orphan-only households, and 298 children from the 210 mixed households (a total of 725 children). Of the 210 mixed households, 87 had non-orphans of comparable age. The analytical sample used for the current analysis thus consists of 174 children at baseline, 87 orphans and 87 non-orphans who live together in mixed households; and 124 children, 62 orphans and 62 non-orphans, in the second round. Of the 25 pairs that dropped out in round 2, six dropped out as the comparison non-orphans were orphaned (therefore the household no longer remained mixed); the remaining 19 pairs had at least one child moved to another household as a result of a change of caregiver, return to living with parents, or move for family reasons. The movement of one child led to the exclusion of the comparison child from the analysis.

Analytical methods

This paper examines the education, health and labour outcomes of orphans compared with the non-orphans with whom they live. For education, we assessed two indicators: grade normalized for age, and annual expenditures on schooling for the child. For physical health, we examined body mass index (BMI), which was calculated then translated into z-scores and percentiles using the United States Centers for Disease Control and Prevention age and sex-specific reference curves. As malnutrition often manifests itself as obesity, analysing BMI as a continuous variable can be misleading. We therefore looked at malnourishment as a categorical variable in which malnourishment was defined, in accordance with Centers for Disease Control and Prevention definitions, as being in the bottom 5 percentile or the top 5 percentile of the age and sex-specific BMI distribution. For labour outcomes we examined the categorical variables: whether the child had worked outside the house in the last week and whether the child had done chores within the house in the last week.

Bivariate relationships between child type and demographic variables were assessed using Mantel–Haenszel chi-square tests for categorical variables and t-tests for continuous outcomes. For the multivariate analysis, we estimated a linear regression model using household fixed effects. The household fixed effect allows for the comparison of children (with different characteristics) within households by controlling for all observed and unobserved child invariant household characteristics such as income, assets, household size, distance to school, etc. As co-resident orphans and non-orphans will have the same household characteristics, this method allows us to identify the within-household differences between orphans and non-orphans.

For continuous variables, we estimated the following linear model:

where Yijt represents the educational attainment/BMI for child i from household j at time t; male is an indicator variable for whether child is male or not; age is the age of the child; mobility is an array of categorical variables reflecting when the child moved into the house; orphan type is the type of orphan (maternal orphan, paternal orphan, double orphan); resident parent is whether the surviving parent of the orphan or the parent of the non-orphan is living at home (father lives at home, mother lives at home). Hj is the child invariant household fixed effect; and eijt is the error term. Several models were estimated using the different independent variables in different combinations, but results from only one such model are presented in this paper.

For categorical variables, we estimated a logit model:

where Yij represents whether child i from household j at time t is malnourished or not/works at home or not/does chores at home or not; and the remaining variables are as defined above.

The decision to include the variables on mobility and resident parent was informed by the literature. This literature shows that fostering and mobility are high even for all kinds of children in South Africa because of high levels of adult migration and children born out of wedlock [14,16,17]. We thus feel it is important not only to introduce these variables as controls.

Examining children who are co-resident using household fixed effects allows us to control for common household characteristics. It is impossible to know, however, with these data, whether the children being compared were indeed comparable before the death of the parent because we do not have information on the orphan's characteristics before being orphaned (which may have themselves been affected by orphaning). We are also unable to consider the fixed and unobserved characteristics of the child him/herself, even though this is longitudinal data, as most of the variables of interest have remained constant over time. As a result of this limitation of the data, this paper does not try and isolate the impact of parental death on children. Rather, it compares orphans and non-orphans on a range of indicators and tries to identify some causal pathways for the results.


At baseline, bivariate analysis of sociodemographic characteristics shows few differences between the 87 orphans and 87 non-orphans in the analytical sample (Table 1). There are no statistically significant differences between the demographic, educational, health, or labour outcomes between orphans and co-resident non-orphans. Whereas attendance was near 100%, approximately half the sample had repeated a grade once. Average body mass was within normal range and was comparable for orphans and non-orphans. Approximately 10% of the children work outside the house and 91% report assisting with chores in the household (with no differences between orphans and non-orphans). Although mobility of the sample was high, it was equally high for orphans and non-orphans, with approximately 30% of the sample having moved at least once and approximately 12% in the past 2 years. Living arrangements were different, however, with a parent being the primary caregiver for only 39% of the non-orphans and 12% of the orphans. Grandparents were primary caregivers for 56% of the orphans and 43% of non-orphans, although this difference was not statistically significant. These findings of non-difference remained unchanged in round 2 (results not shown). Table 2 shows the demographic characteristics of the sample that was lost in round 2. None of the characteristics listed were significant predictors of attrition (results not shown).

Table 1:
Demographic characteristics of orphans and non-orphans at baseline (2004–2005).
Table 2:
Characteristics of the 25 children lost to attrition in round 2.

Of the orphans at baseline, 13 were maternal-only orphans, 30 were paternal-only orphans, 26 were double orphans, and 19 had missing information on which parent had died. This changed to nine maternal-only orphans, 21 paternal-only orphans, 21 double orphans, and 12 with missing data in the following year. Table 3 summarizes the demographic and socioeconomic characteristics of children by orphan type.

Table 3:
Demographic characteristics of maternal, paternal and double orphans at baseline (2004–2005).

As mentioned in the methodology section, several models, each controlling for a different combination of independent variables, were estimated to examine the orphan/non-orphan differentials across different outcomes. Table 4 shows results from two such models. The results were consistent across specifications: paternal orphans are more likely to be behind in school than non-orphans with whom they live, they are on average a third of a year behind in their grade. Maternal orphans are on average half a year behind in schooling, but this effect is statistically insignificant. Mobility within the past 2 years is seen to have a positive effect on grade progression. Orphanhood does not seem to have any effect on expenditures on schooling. Recent mobility is associated with a substantial negative effect on schooling expenditure. The impact of parents being present at home is insignificant (results not shown).

Table 4:
Educational and health outcomes for orphans and co-resident non-orphans (household fixed effects).

Bivariate analysis demonstrates no significant differences in nutrition, health proxy, and labour outcome indicators such as going to bed hungry the previous night, being sick in the past 6 months, and working both within and outside the house (see Table 1). Analysis of BMI, presented in Table 4, shows that BMI is lower for orphans. This is also robust when controlling for mobility. As malnutrition also manifests itself as obesity, however, lower BMI is not necessarily informative, especially in adolescents. Logistic analysis (Table 5) shows that maternal and double orphans are at greater odds of being malnourished than non-orphans but this is not statistically significant. Maternal and paternal orphans are at greater odds of doing chores within the house and at lower odds of working outside the house compared with co-resident non-orphans, but again these differences are not statistically significant.

Table 5:
Health and labour outcomes for orphans and co-resident non-orphans (logistic model with household fixed effects).


The results show some statistical differences in educational outcomes and no differences for health and labour outcomes between orphans and non-orphans who live in the same households.

Case and Ardington [15], in their intrahousehold analysis also set in KwaZulu Natal (albeit in a poorer district), found that maternal orphans are ‘on average, 0.12 of a year behind in their schooling and have 7% less spent on their education’ compared with the non-orphans with whom they live. Although the differences in expenditure are moderate, the magnitude of the difference between orphans and non-orphans in terms of schooling is small: 0.12 of a year behind equates to orphans being behind by a little over a month. They found no difference for paternal orphans. Our results, on the other hand, show that paternal orphans are behind in school.

Given the fact that a large proportion of South African fathers are absent or not linked to the household [16], the significant effect of paternal orphanhood is curious. The coefficient on the father being a resident within the household is insignificant (results not shown), suggesting that the impact of paternal orphanhood may be caused by the fact that the death of a father could be an economic shock that may have, at some point, resulted in children dropping out of school. We, unfortunately, have no way to test this hypothesis and can only offer it as a potential explanation.

What could explain the lack of overall differences between orphans and non-orphans? First, temporal issues may be driving the result. It is important to remember that we are merely looking at incident (recent) orphans, i.e. children that have lost at least one of their parents in the 6 months before the survey and in the same year the survey was conducted. This may not be sufficient time to see a large effect on children, or the households may have effective short-term coping mechanisms to mitigate the effect on children [9]. On the other hand, one can argue that the critical period for an orphan child is the terminal illness period as a result of the trauma of seeing a parent wasting away and sometimes having to miss school in order to attend to sick parent(s), and that the impacts may diminish over time.

Second, Case and Ardington [15] have argued that differences in outcomes between orphans and non-orphans are driven by the tendency to live with distant or unrelated caregivers. All the orphans in the sample are either living with close relatives (aunt or grandmother) or their surviving parent, as in a study from the same province by Adato et al. [14]. Therefore, they are likely to receive the same support as the non-orphans with whom they live. With time, destination households may become oversaturated and could struggle to absorb more children and this may change. South Africa's extensive social grants system potentially mitigates against this phenomenon and assist families in coping.

Third, in Amajuba District's context of high adult migration, having a parent alive does not equate to the presence of a parent at home, thus orphanhood itself may not be associated with lower educational or health outcomes. The majority of both orphans and non-orphans live without parents present at home. Table 1 indicates that only 38.64% of non-orphans have parents as primary caregivers, and a large proportion of both orphans and non-orphans live with their grandparents. Even single parent orphans tend not to live with their surviving parent. Migration for employment was the most frequently cited reason for parents' not living at home. Furthermore, approximately a third of the fathers were not living at home because they were not married to the mother. This figure also calls into question the role of biological parents (especially fathers) in caregiving, and supports other literature that shows that the absence of fathers is high in South Africa, with 55% of fathers being absent in rural South Africa in 2002 [16]. Current definitions of orphan inaccurately privilege the biological parent in a context in which even non-orphans do not live with their parents. This calls into question our thinking on the category of orphan in South Africa.

Fourth, when orphans have moved from their original households (i.e. they were fostered into the survey household, which has non-orphans), there may be endogeneity in placement decisions, in that orphans are strategically moved to better-off households and this may bias the results towards the null. The positive and significant coefficient on recent mobility (within the past 2 years) supports the idea that children are often moved for schooling. It is important to note, however, that mobility is equally high for both orphans and non-orphans. Table 1 shows how there is no statistically significant difference between orphans' and non-orphans' mobility. Of the 174 children at baseline, 23 children moved houses in the preceding year, 13 orphans and 10 non-orphans. Moreover, in round 2, sample attrition was equally high for orphans and non-orphans. Therefore, although there may be endogeneity in placement decisions, we do not believe it is disproportionately so for orphans when compared with non-orphans. Child migration is a historical/cultural phenomenon, and fostering literature shows how children have lived away from their ‘nuclear’ families (although this may be exacerbated by AIDS mortality) [17].

Fifth, it is possible that the indicators used may not be sensitive to differences, particularly because the orphans were so recently orphaned. There may be some limitations of BMI, but in general it is difficult to identify good measures of health of older/school-aged children because this age group is generally very healthy (self-reported or otherwise). In terms of schooling, there maybe differences in performance within a grade that are not captured by these instruments. Whether an orphan child is truly learning, as opposed to progressing, like non-orphaned children, may not be fully captured by the data.

The study has some limitations worth mentioning that may bias the results towards the null. First, the tests have low power as a result of the relatively small sample size, and this may contribute to not finding statistically significant effects. The attrition in round 2 only further reduced the sample. Comparisons with much larger studies should be made with this in mind.

Second, the study sample was drawn from a random sample of schools in the district. Using schools as our sole recruitment source for study participants was both methodological and practical. Drawing a sample of school-going ‘recent’ orphans and non-orphans introduces a sampling bias that potentially biases the intrahousehold results towards the null as worse-off orphans may have been excluded from the sample. What is important to note is that school enrolment rates in KwaZulu-Natal are extremely high. The national 2003 General Household Survey conducted by Statistics South Africa showed school enrolment rates to be 97.5% for non-orphans and 95% for orphans. Upon further analysis of this survey, we found orphans who are not enrolled neither worse off than the non-orphans with whom they live nor are they are worse off compared with enrolled orphans in terms of health and labour outcomes. Their schooling outcomes do differ, however, and this may bias the results towards the null.

Finally, Evans and Miguel [6] argued that studies that do not take into consideration child fixed effects are likely to be biased to the null because they are unable to account for omitted variable bias as well as endogeneity because we do not have information on the child's characteristics before orphaning. The results therefore serve as a lower bound on the differences between orphans and non-orphans.


We have shown a lack of systematic difference in education and health outcomes between orphans and co-resident non-orphans, with a few exceptions. Policy responses (and literature) do not always distinguish between orphans living with other orphans or orphans living in mixed households, even though they may live in completely different circumstances. Comparing the results presented in this paper with the results in the literature reveals the heterogeneity of the category of orphans. The contradiction of the results of Case and Ardington [15] highlights the need to refrain from generalizing to the country based on district or even province level results. The paper thus illustrates the need for context-specific approaches that pay attention to definitions, as opposed to sweeping global responses to the crisis. The remarkable similarity of orphans' and non-orphans' living arrangement calls into question the category of orphan in South Africa. Evidence from the study thus far challenges the Cinderella assertion in the case of orphans in South Africa. Unstable households with absent fathers are common and mobility is high, as also suggested by other literature. Reframing the discussion around orphanhood to be appropriate to South Africa's social context is warranted.

Sponsorship: This project was funded by the National Institute of Child Health and Development of the United States National Institutes of Health under its African Partnerships programme (grant R29 HD43629).

Conflicts of interest: None.


1. Department of Health. National HIV and Syphilis Antenatal Sero-Prevalence Survey in South Africa 2005. Pretoria: National Department of Health, South Africa; 2006.
2. UNAIDS. 2006 Report on the global AIDS pandemic. Geneva: UNAIDS; 2006.
3. Dorrington R, Bradshaw D, Johnson L, Budlender D. The demographic impact of HIV/AIDS in South Africa: national indicators for 2004. Cape Town: Centre for Actuarial Studies, South Africa Medical Research Council, Actuarial Society of South Africa; 2004.
4. Meintjes H, Giese S. Spinning the epidemic: the making of mythologies of orphanhood in the context of AIDS. Childhood 2006; 13:407–430.
5. Case A, Paxson C, Ableidinger J. Orphans in Africa: parental death, poverty and school enrolment. Demography 2004; 41:483–508.
6. Evans D, Miguel E. Orphans and schooling in Africa: a longitudinal analysis. Demography 2007; 44:35–57.
7. Yamano T, Jayne TS. Working-age adult mortality and primary school attendance in rural Kenya. Econ Dev Cultural Change 2005; 53:619–653.
8. Bicego G, Rutstein S, Johnson K. Dimensions of the emerging orphan crisis in sub-Saharan Africa. Soc Sci Med 2003; 56:1235–1247.
9. Beegle K, Weerdt JD, Dercon S. Orphanhood and the long-run impact on children. Oxford: Centre for Studies of African Economies; 2006.
10. Ainsworth M, Filmer D. Inequalities in children's schooling: AIDS, orphanhood, poverty and gender. World Dev 2006; 34:1099–1128.
11. Lloyd CB, Blanc AK. Children's schooling in sub-Saharan Africa: the role of fathers, mothers and others. Popul Dev Rev 1996; 22:265–298.
12. Chatterji M, Dougherty L, Ventimiglia T, Mulenga Y, Jones A, Mukaneza A, et al. The well-being of children affected by HIV/AIDS in Lusaka, Zambia and Gitarama Province, Rwanda: findings from a study. Working paper no. 2. Washington, DC: Community REACH; 2005.
13. Crampin A, Floyd S, Clynn J, Madise N, Nyondo A, Khondwe M, et al. The long-term impact of HIV and orphanhood on the mortality and physical well-being of children in rural Malawi. AIDS 2003; 17:389–397.
14. Adato M, Kadiyala S, Roopnaraine T, Biermayr-Jenzano P, Norman A. Children in the shadow of AIDS: studies of vulnerable children and orphans in three provinces in South Africa. Washington, DC: IFPRI; 2005.
15. Case A, Ardington C. The impact of parental death on school outcomes: longitudinal evidence from South Africa. Demography 2006; 43:401–420.
16. Posel D, Devey R. The demographics of fathers in South Africa: an analysis of survey data, 1993–2002. In: Richter L, Morrell R, editors. Baba: men and fatherhood in South Africa. Cape Town: HSRC Press; 2005. pp. 38–52.
17. Madhavan S. Fosterage patterns in the age of AIDS: continuity and change. Soc Sci Med 2004; 58:1443–1454.

caregivers; children; HIV; intrahousehold; mortality; orphans; South Africa

© 2007 Lippincott Williams & Wilkins, Inc.