Printer Friendly

Understanding fatherhood and father involvement in South Africa: insights from surveys and population cohorts.

Calls for research on men's roles in families in sub-Saharan Africa, especially as fathers, and for greater efforts to engage men in family-based interventions and policy requires detailed data on family functioning, parenting and parenthood. However, in resource poor countries within the region few family or birth cohort studies have been conducted and household and population-based studies continue to be the main sources of empirical data about families. In this brief report, we review the data on fathers and father involvement collected in South Africa (e.g., national household surveys, household panel surveys, demographic surveillance systems, birth cohort) and suggest ways to improve ongoing empirical data collection efforts.

Keywords: fathers, families, sub-Saharan Africa, measurement, study design


There have been calls for increased research on men's roles in families in sub-Saharan Africa, especially as fathers, and for greater efforts to engage men in a wide variety of family-based HIV prevention and reproductive, maternal and child health interventions and policy (Engle, 1997; Mbivzo & Bassett, 1996; Richter et al., 2009). Similar advances in scholarship, policy and intervention research related to men, particularly fathers, in the United States and Europe were accompanied by improvements in the data available about parenting and families in large, nationally representative family surveys and birth cohorts (Batty, 2009; Cabrera et al., 2002; Eggebeen, 2002). In contrast, very few dedicated family studies have been conducted in resource poor countries in sub-Saharan Africa where household and population-based studies continue to be the main sources of empirical data about families. Even though many household surveys, household panel studies and demographic surveillance systems are conducted in the region, typically few questions are asked about parents and parenting, especially with respect to fathers and fathering.

The dearth of detailed data about family functioning in general, and men's involvement in families in particular, is of concern. At best gaps in knowledge remain unfilled. At worst, a lack of data or poor measures misrepresent fathering and father involvement, opening the way for incautious interpretations about men's absence or limited contribution to family life (O'Laughlin, 1998; Posel & Devey, 2006; Townsend, 1997).

In this brief report, we review empirical data on fathers and father involvement in South Africa collected in national household surveys, household panel surveys, demographic surveillance systems, and a birth cohort study. We consider the extent to which existing data adequately measure the identity and participation of fathers in families. We conclude with suggestions about ways to enhance the contribution of ongoing data collection efforts to promoting research and interventions related to men and families.


South Africa was chosen as a country case study within sub-Saharan Africa for several reasons. There is an established body of multi-disciplinary work on fatherhood research and policy in South Africa (Richter, 2006). Fathers have recently been a focus in high profile NGO and government programmes on HIV prevention, reproductive health and gender-based violence, for example (Brothers for Life, 2010; Sonkhe Gender Justice Network, 2010). South Africa has considerably more resources (financial, scientific, logistics) than most other sub-Saharan African countries for research and data collection. A birth cohort study, one of only two longitudinal family studies in Africa, is conducted in the Soweto area of Greater Johannesburg (Richter et al., 2004). In addition, South Africa has several national and provincial cross-sectional and panel household surveys, as well as three ongoing demographic surveillance systems.

The context for fathers and families in South Africa also illustrates many of the challenges for empirical data collection pertinent to other parts of sub-Saharan Africa; most notably measuring the involvement of non-resident and social fathers. Families in apartheid and post-apartheid South Africa are characterised by very high levels of residential separation between biological fathers and their children (Jones, 1992; Jones, 1993). The widespread pattern of families "dispersing" between different households for reasons including work and schooling, was fostered by highly regulated apartheidera systems of labour migration and bantustan settlements but still continues today (Murray, 1980; Sharp, 1994; Spiegel, 1987). Labour migration compounded by family dispersal and economic insecurity has also contributed to a long-term decline in marriage and increasing non-marital fertility rates (Budlender et al., 2005; Hosegood et al., 2009b; Preston-Whyte, 1978).

Paternity and fathering are important aspects of men's identity in South Africa (Hunter, 2006; Hunter, 2008; Mkhize, 2006; Morrell, 2006). While young children born to unmarried parents will often live with their mothers, most children will be acknowledged by their father and paternal kin (Preston-Whyte, 1974, 1993; Russell, 2003). While studies in rural areas of South Africa have reported that less than half all children are co-resident with their biological fathers at birth (Hosegood et al., 2009a), it is very common for children to be considered part of maternal and paternal households and move between them (Madhavan et al., 2008). Social fatherhood is also widely practiced and important for men and families in South Africa (Mkhize, 2004). Many men are involved in fathering of step-children non-biological children. Social fatherhood in South Africa is very common whether formally through fostering or adoption or more typically informally. Many factors contribute to the extent of involvement by men in the raising non-biological children including cultural practices that emphasise collective responsibility of the extended family in child-rearing, high levels of long-term migration, relationship dissolution and re-partnering, and the high level of rate of paternal orphaning due to the severe H1V epidemic and other causes of premature male mortality (Hosegood et al., 2009b; Hunter, 2006; Jones, 1992; Mkhize, 2006).

Detailed observations of different dimensions of men's involvement in families in South Africa, particularly in relation to children, are scarce (Morrell & Richter, 2006). Co-residence has been shown to have only limited utility as an indicator of father involvement in South Africa given the context of labour migration, household fluidity and non-marital childbearing (Brookes et al., 2004; Moultrie & Timaeus, 2001). The involvement of fathers in financial and material provision is undoubtedly central and a dimension about which most is known (Hunter, 2006; Madhavan & Townsend, 2007; Madhavan et al., 2008; Wilson, 2006). Qualitative studies reporting men's own accounts of their involvement report father involvement in a wide range of traditional and non-traditional roles including intimate physical care, education, play, emotional engagement, organisation, and monitoring (Montgomery et al., 2006; Ramphele, 1993; Swartz & Bhana, 2009).


In this section, we describe data on fathers, fatherhood and fathering collected in South African household surveys, demographic surveillance systems and birth cohort. We relate these data to issues in conceptualising, measuring and collecting data about men and families. Table 1 summarises data about fathers collected in two national cross-sectional household surveys (South African DHS, 2003 and General Household Survey (GHS), 2006); a national household panel survey (National Income and Dynamics Survey (NIDS), 2008/9); two provincial household panel surveys (KwaZulu Income Dynamics Study (KIDS), 1993-2004 and Cape Area Panel Study (CAPS), 2002-2006); a birth cohort study (Birth to Twenty (BT20) study, 1990-ongoing); and two demographic surveillance systems (Africa Centre Demographic Information System (ACDIS), 2000-ongoing and Agincourt Health and Demographic Surveillance System (AHDSS), 1992-ongoing). While varying in terms of their design, purpose and study populations, with the exception of the birth cohort study, the primary sampling and enumeration unit used in the other surveys and surveillance systems is that of a household. All these data sources collect some information about each member of a household and his or her relationship to one or more members of the household.

Counting Fathers

In 2006, Posel and Devey concluded that: "There are currently no data available in South Africa with which we can count and describe all men who are fathers, or identify changes over time." This remains the situation as no paternity histories are available from the South African household surveys and studies. Instead information about fathers is collected indirectly in relation to children in the household. The most widely available indicator about biological fathers is his survival status, and his identity in situations where father and child belong to the same household.

Information about paternal survival status and child-father dyads typically refers to biological fathers. Some questionnaires or training guides do not make this explicit but reports of social fathers cannot be distinguished from biological fathers. Information about social fathers is inconsistently collected in the data sources reviewed. Any detailed data on social fathering is collected from the perspective of children within the household. The Birth to Twenty cohort study has collected information about men described as "father figures." In CAPS, foster-, adopted- or step-fathers are recorded when the social father is co-resident with the child. In other surveys, this information may be imputed for some children indirectly from data collected on intra-household relationships, for example, where the relationship between each member and the head of household is recorded. However, this information cannot be consistently used to identify father-child pairs in cases where the child's father is not the head of household. None of data sources reviewed ask all men whether they were social fathers to children in or outside the household.

Measures of Father Involvement

Very little information about the involvement of fathers with children is collected in South African household surveys. We examine the available data with reference to Lamb et al.'s (1985) well-known model that groups dimensions of father involvement into accessibility, engagement and responsibility (Lamb et al., 1985). Descriptions of father involvement in South African surveys are primarily based on a child's access to his biological father; with access narrowly defined as whether the father is alive, and if so, whether he is co-resident. Given the severity of the HIV epidemic in South Africa, there is considerable interest in understanding the impact of parental deaths on children's health and wellbeing. In 2005, 12 percent of children under 18 years in the ACDIS population were paternal orphans and 2 percent were double orphans (Hill et al., 2008). The survival status of biological fathers is asked in all South African cross-sectional surveys, and updated in some of the sources of longitudinal data.

Measuring the accessibility of fathers to their children is complicated in South Africa by high levels of circular labour migration (Hosegood & Madhavan, 2010). Many households, especially in rural areas, include members who whilst living somewhere else, are nonetheless still considered to be part of the household (Russell, 2003). Fathers of young children will often be non-resident members because they have migrated for work or to look for work. Qualitative research in rural South Africa has shown that non-resident fathers belonging to the same household as their child are more available and engaged than non-resident fathers who are not members of the same household (Madhavan et al., 2008). Whilst all data sources examined collect information about whether fathers are resident or not, the definitions of households vary which influences social representations of fathers and children. Where a co-residential concept of household is used, non-resident fathers will be excluded even when considered by respondents to be members of the household (e.g., DHS). In contrast, if household eligibility is defined by membership rather than co-residency, non-resident fathers can be recorded if reported by the household respondents (e.g., ACDIS and Agincourt DSS). Collecting data about the characteristics and involvement of fathers identified on a survey household roster is considerably more straightforward than for fathers whose are not explicitly identified. For fathers on the rosta, questions can be formulated for his specific circumstances and any data collected about him in different modules or rounds of data collection will also be available. Some South African surveys and cohort studies have sought to collect information about basic socio-economic characteristics about all biological fathers of children in the household. For example, the 1998 South African DHS asked about the education and employment status of fathers for all children.

Beyond survival and co-residence, information about other dimensions of father involvement are limited and inconsistently collected. Where father engagement and responsibility is investigated, most surveys focus on the topic of financial contributions towards children's food, clothing, school fees and school uniforms. In some surveys, questions are asked directly about financial contributions made by the father (biological or social) for each of his children within the household or to children outside the household (e.g., NIDS). However, it is more typical that questions about engagement and responsibility in relation to aspects of children's lives (schooling, meals, intimate personal care, health care) ask only about the person who makes the "largest" contribution or is the "primary" person responsible. Therefore, unless the child's father holds these roles, no information is collected about his involvement. Social and cultural gender norms in South African communities can strongly influence the reporting of father involvement (Montgomery et al., 2006; Swartz & Bhana, 2009). In ACDIS for example, a question is asked about care: "Who is the main person responsible for NAME's day-to-day care?' It would be unusual for families to report the child's father as having primary responsibility for caregiving if a child's mother or another woman was present in the household, particularly with respect to young children. Other than a few one-off data collection exercises (e.g., the Birth to Twenty parental questionnaire administered at 18 years), detailed information about the types and amount of specific activities and roles played by fathers have not been collected.

Cross-Sectional and Longitudinal Data About Fathers

Whilst most nationally representative household surveys in South Africa have been cross-sectional, South Africa has several sources of longitudinal data including household panels and demographic surveillance systems. There are particular disadvantages to using cross-sectional survey data to explore father involvement in South Africa. A static profile of household composition and functioning cannot capture changes in men's involvement with respect to different children and families. Longitudinal data on fathers, if appropriately measured, can reveal much more about the very dynamic social and residential arrangements of men and families. Age can influence the circumstances under which men become fathers and the way in which they are involved in fathering. Marriage, partnering, childbearing, employment, and migration patterns vary by age but there may also be changes in men's values and attitudes with age and experience. Given low rates of marriage and late age at marriage most young fathers will not be married to or be members of the same household as the mothers of their children (Hosegood et al., 2009a). Circumstances in which older men become fathers or are involved in fathering may be quite different to those of younger men. For example, older men may already have established their own independent households. With high levels of union instability, many men will father children with more than one woman with whom the level and quality of contact and communication may differ and change over time influencing their involvement. Furthermore, changing father roles may also reflect social and cultural norms about appropriate involvement depending on the age and sex of each child. Longitudinal data are required to answer questions about the long-term effects of fathers and fathering for children, other family members, and indeed men themselves.

However, the existing approaches used to collect longitudinal data about fathers in the ongoing studies reviewed present challenges for studying men and families. The most important limitation is that little or no prospective data is collected about fathers who are not household members. Changes in these men's social and economic circumstances may be potentially important influences on family involvement, for example, loss of employment, remarriage, becoming a father again. However, such information is not currently collected for fathers who are not part of the index household. The exception is questions asked directly about the survival status of fathers of all children in the household (e.g., in ACDIS) (Hosegood et al., 2007). A absence of prospective information about fathers is also an when men cease to be part of the household, for example, upon divorce or out-migration.

Choice of Respondents

For reasons of logistics and cost, household surveys and demographic surveillance systems primarily rely on key household informants to answer routine questions. A knowledgeable adult member of the household is typically selected to answer questions about household composition, basic characteristics of each member, and events such as deaths or migrations. The consequence is that most data about men collected in these surveys are proxy reports rather than reports by men themselves. In the majority of cases, key household respondents are women.

The extent to which proxy informants can provide reliable and valid data on men's paternity and fathering will depend on several aspects including: their knowledge, the type of information that is collected, sensitivity of the questions, social desirability, and, the quality of their relationship.

Studies examining the reliability data reported by proxy household respondents, in particular, mothers in providing data on fathers suggest that proxy respondents are generally good informants for social and demographic characteristics they are less reliable for health and well-being, and poor judges of emotional states and perceptions (Nelson et al., 1994).

With respect to the types of questions currently asked about fathers in household surveys, those related to financial contributions and care responsibility may be at particularly vulnerable to misreporting by proxy respondents. Information about non-resident fathers, particularly fathers who are not members of the households, may not be known by the respondent or interpreted in a particular light. For example, aspects of men's involvement in families may be under-reported where household respondents perceive that a man has behaved irresponsibly towards or abandoned his family (Montgomery et al., 2006). Some surveys and longitudinal studies reviewed conduct interviews specifically with men; however, data collection is focused on the topic of sexual and reproductive health (DHS and DSS) or employment (GHS). Such Interviews are therefore, missed opportunities for collecting data about paternity and fathering directly from men.

Improving Data Collection in South African Household Studies

Despite the conceptual and methodological shortcomings of household-based survey and study design, they will nonetheless continue to be the most common source of empirical data on fathers in South Africa and in the rest of the sub-Saharan Africa. A need for research and policy addressing how best to promote and support positive involvement of men in families justifies efforts to improve the collection of relevant, detailed, and valid data in large, cross-sectional and longitudinal household surveys in Africa (Desmond & Hosegood, 2011). South Africa is well placed with its multiple, well-resourced repeated surveys and longitudinal studies to develop approaches and instruments that can be adapted for other African settings. While some of these may be particularly dominant features of South African communities, in particular the very large proportion of biological fathers who live apart from their children; these issues are by no means unique. Much can also be learned from efforts to learn more about fathers in other countries, for example, the Fatherhood Research Initiative in the United States (Cabrera et al., 2002). Similarly different approaches to validating and collecting data about and from parents in longitudinal family studies can be explored for their applicability in household-based studies (Tomkins, 2009). In addition, the extensive research exploring how best to measure father involvement provides us with a picture of the diverse circumstances in which men become fathers and the variety of forms men's involvement in families can take.

We suggest that the main priorities are a) to use innovative approaches to collecting data about fathers who are not part of the index household, and b) to identify socially and culturally relevant measures of men's involvement in families; not only the role of biological fathers but all men contributing to child and family health and well-being. In Table 2, we present a list of indicators and measures that could potentially be included in ongoing household surveys or studies in South Africa. The table indicates whether indicators would be collected for all men or with respect to an index child; topics related to biological and social fathers. The suggestions are necessarily generic as their inclusion would require a process of refinement, piloting and validation in a specific study population and research design context. Social fathering for example, unlike biological fathering requires definitions and constructs that are developed and tested locally (Townsend, 2002). We include a suggestion that data could be collected not only about fathers within but also outside the household. As the type of information about fathering becomes more detailed interviews with men are needed. Several ongoing surveys and studies already conduct interviews with men. Furthermore, alternatives to face-to-face interviews successfully in other studies might be usefully explored, for example, telephone interviews (Kirsch, 2002). In a household survey or cohort where a father-child relationship has been openly reported by the household respondents and no paternity tests are used, questions about fathering may not reduce men's willingness to participate in interviews. If information is collected from multiple respondents for example, reports of father-child contact from household respondent and fathers, attention will be needed to examine respondent bias and data validation.

Theoretical and Methodological Implications Beyond South Africa

The picture of South African fatherhood that comes to the fore is one of large numbers of children whose biological fathers are not co-resident and the commonplace situation of men taking on father roles and responsibilities for non-biological children, typically without formal legal recognition or rights. Questions may therefore be asked about whether the case of South Africa needs to be understood as exceptional, particularly given some of the legacies of the unique policies and practices linked to Apartheid. Moreover, it might be assumed that collecting data on fathers and fatherhood is considerably more difficult in South Africa than elsewhere and would require specifically tailored study designs. In reality, however, the forms and patterns of fatherhood and father involvement are similar in their diversity of types to those identified in work in Western countries, for example the main permutations of fatherhood described by Marsiglio and colleagues (2000). An exception would be that of polygamous marriage arrangements although these constitute a small and declining proportion of marriages in South Africa. For South African as for international scholars, the need to represent and understand the involvement of non-resident and social fathers is increasingly an important shared issue; one that poses methodological challenges for measurement, conceptualisation, design and analysis across different settings. It is also important to note that as elsewhere, a considerable amount of fathering in South Africa occurs in co-residential two parent families. There is value in comparing empirical findings and theoretical frameworks between different cultural contexts as exemplified in work by Townsend exploring men and families in rural Botswana and suburban U.S.; Roy's comparison of African-American and South African Black fathers using life course theories; and Madhavan and Roy's analysis of how kin support fathering in low income Black communities in the US and South Africa (Madhavan & Roy, 2012; Roy, 2008; Townsend, 2002). While cultural differences are important we need to recognize the similarities that exist across contexts, including marginalization from the labor market and racism. With social inequality increasingly becoming a defining characteristic of so many countries, Black men in South Africa are not unique in facing challenges to positive involvement in families as fathers.

One distinctive feature of fatherhood studies in South Africa and even more so in the are the region are the limited resources for data collection which influences the types of study designs with which we work. Without support for the kind of detailed, longitudinal data on fathers and family processes available in the U.S. and Europe, we are substantially constrained in our ability to identify the determinants of father involvement and outcomes for children and other family members. When relying on data available from household surveys that have been primarily collected for other purposes, researchers must critically review the data from these sources whilst, as this article demonstrates, identifying ways that these surveys can improve the definitions and data on fathers and families.

In conclusion, the dearth of the data about fathers in South Africa has resulted in a gap in knowledge about the role of fathers at best and a misrepresentation of what they do (or do not do) at worst. Throughout the region, household surveys can be enhanced to improve the data collected about the involvement and impact of fathers, mothers and other people on child and family health and well-being.

DOI: 10.3149/fth.1003.257

The authors thank the anonymous reviewers for their helpful comments and suggestions. All correspondence regarding to article should be addressed to Dr. Vicky Hosegood, School of Social Sciences, Murray Building (58), University of Southampton, Southampton, SO17 1B J, UK. Email: Hosegood acknowledges funding support for this work from the Wellcome Trust, UK (#WT082599MA).


Batty, G.D. (2009). Family-based life course studies in low- and middle-income countries. In D.A. Lawlor & G.D. Mishra (Eds.), Family matters: Designing, analysing, and understanding family-based studies in life course epidemiology (pp. 129-150). Oxford: Oxford University Press.

Brookes, H., Shisana, O., & Richter, L.M. (2004). The national household HIV-prevalence and risk survey of South African children. Pretoria: HSRC Press.

Brothers for Life. (2010). Brothers for Life website. Retrieved from

Budlender, D., Chobokoane, N., & Simelane, S. (2005). Marriage patterns in South Africa: methodological and substantive issues. Southern African Journal of Demography, 9, 1-25.

Cabrera, N., Brooks-Gunn, J., Moore, K., West, J., Boller, K., & Tamis-LeMonda, C.S. (2002). Bridging research and policy: Including fathers of young Children in national studies. In C.S. Tamis-LeMonda & N. Cabrera (Eds.), Handbook o f father involvement: Multidisciplinary perspectives (pp. 489-523). New Jersey: Lawrence Erlbaum Associates, Inc., Publishers.

Desmond, C., & Hosegood, V. (2011). Men, families and HIV/AIDS. Men in families and family policy in a changing worm (pp. 163-201). New York: United Nations Department of Economic and Social Affairs.

Eggebeen, D.J. (2002). Sociological perspectives on fatherhood: What do we know about fatherhood from social surveys. In C.S. Tamis-LeMonda & N. Cabrera (Eds.), Handbook of father involvement. Multidisciplinary perspectives (pp. 189-209). Mahweh, NJ: Lawrence Erlbaum Associates, Inc.

Engle, EL. (1997). The role of men in families: Achieving gender equity and supporting children. Gender and Development, 5, 31-40.

Hill, C., Hosegood, V., & Newell, M. (2008). Children's care and living arrangements in a high HIV prevalence area in rural South Africa. Vulnerable Children and Youth Studies, 3, 165-77.

Hosegood, V., Floyd, S., Marston, M., Hill, C., McGrath, N., Isingo, R., Crampin, A., & Zaba, B. (2007). The effects of high HIV prevalence on orphanhood and living arrangements of children in Malawi, Tanzania, and South Africa. Population Studies (Camb), 61, 327-36.

Hosegood, V., & Madhavan, S. (2010). Data availability on men's involvement in families in sub-Saharan Africa to inform family-centred programmes for children affected by HIV and AIDS. Journal of International AIDS Society, 13(Suppl. 2), S5.

Hosegood, V., McGrath, N., Bland, R., & Newell, M. (2009a). Fathers' involvement during infancy: Factors influencing social and residential dynamics in rural South Africa. Paper presented at the 1USSP XXVI International Population Conference. Marrakech.

Hosegood, V., McGrath, N.M., & Moultrie, T.A. (2009b). Dispensing with marriage: Marital and partnership trends in rural KwaZulu-Natal, South Africa 2000-2006. Demographic Research, 20,279-312.

Hunter, M. (2006). Fathers without amandla: Zulu-speaking men and fatherhood. In R. Morrell & L. Richter (Eds.), Baba: Men and fatherhood in South Africa (pp. 99-117). Cape Town: HSRC Press.

Hunter, M. (2008). IsiZulu-speaking men and changing households: From providers within marriage to providers outside marriage. In B. Carton, J. Laband, & J. Sithole (Eds.), Zulu identities: Being Zulu, past and present (pp. 566-572). Durban: UKZN Press.

Jones, S. (1992). Children on the move: Parenting, mobility, and birth-status among migrants. In S. Burman & E. Preston-Whyte (Eds.), Questionable issue: Illegitimacy in South Africa (pp. 247-281). Cape Town: Oxford University Press.

Jones, S. (1993). Assaulting childhood. Children's experiences of migrancy and hostel life in South Africa. Johannesburg: Witwatersrand University Press.

Kirsch, S.E.D. (2002). Telephone interviewing: A method to reach fathers in family research. Journal of Family Nursing, 8, 73-84.

Lamb, M.E., Pleck, J .H., Charnov, E .L., & Levine, J.A. (1985). The role of the father in child development: The effects of increased paternal involvement. In B .B. Lahey & A.E. Kazdin (Eds.), Advances in clinical child psychology (pp. 229-266). New York: Plenum.

Madhavan, S., & Roy, K. (2012). Securing fatherhood through kinwork: A comparison of black fathers in South Africa and the U.S. Journal of Family Issues, 33, 801-822.

Madhavan, S., & Townsend, N. (2007) The social context of children's nutritional status in rural South Africa, Scandinavian Journal of Public Health, 35(Suppl. 69), 107-117.

Madhavan, S., Townsend, N., & Garey, A.I. (2008). "Absent breadwinners": Father-child connections and paternal support in rural South Africa. Journal of Southern African Studies, 34, 647-663.

Marsiglio, W., Day, R.D., & Lamb, M.E. (2000). Exploring fatherhood diversity: Implications for conceptualizing father involvement Marriage and Family Review, 29, 269-293.

Mbivzo, M.T., & B assett, M.T. (1996). Reproductive health and AIDS prevention in Sub-Saharan Africa: The case for increased participation, Health Policy and Planning, 11, 84-92.

Mkhize, N. (2004, July/August). Who is a father? Children First, 3-8.

Mkhize, N. (2006). African traditions and the social, economic and moral dimensions of fatherhood. In R. Morrell & L. Richter (Eds.), Baba: Men and fatherhood in South Africa (pp. 183-198). Cape Town: HSRC Press.

Montgomery, C.M., Hosegood, V., Busza, J., & Timaeus, I.M. (2006). Men's involvement in the South African family: Engendering change in the AIDS era. Social Science Medicine, 62, 2411-2419.

Morrell, R. (2006). Fathers, fatherhood and masculinity in South Africa. In R. Morrell & L. Richter (Eds.), Baba: Men and fatherhood in South Africa (pp. 13-25). Cape Town: HSRC Press.

Morrell, R., & Richter, L. (2006). Introduction. In R. Morrell & L. Richter (Eds.), Baba: Men and fatherhood in South Africa (pp. 1-9). Cape Town: HSRC Press.

Moultrie, T.A., & Timaeus, I. (2001). Fertility and living arrangements in South Africa. Journal of Southern African Studies, 27, 207-223.

Murray, C. (1980). Migrant labour and changing family structure in the rural periphery of Southern Africa. Journal of Southern African Studies, 6, 139-156.

Nelson, L.M., Longstreth, W.T., Koepsell, T.D., Checkoway, H., & van Belle, G. (1994). Completeness and accuracy of interview data from proxy respondents: Demographic, medical and life-style factors. Epidemiology, 5,204-217.

O'Laughlin, B. (1998). Missing men? The debate over rural poverty and women-headed households in southern Africa. The Journal of Peasant Studies, 25, 1-48.

Posel, D., & Devey, R. (2006). The demographic of fatherhood in South Africa: An analysis of survey data, 1993-2002. In R. Morrell & L. Richter (Eds.), Baba: Men and Fatherhood in South Africa (pp. 38-52). Cape Town: HSRC Press.

Preston-Whyte, E. (1974). Kinship and marriage. In W. Hammond-Tooke (Ed.), The Bantu-speaking peoples of southern Africa (pp. 177-210). London & Boston: Routledge and Kegan Paul. Preston-Whyte, E. (1978). Families without marriage: A Zulu case study. In J. Argyle & E. Preston-Whyte (Eds.), Social system and tradition in southern Africa: Essays in honour of Eileen Krige (pp. 55-85). Cape Town: Oxford University Press.

Preston-Whyte, E. (1993). Women who are not married: Fertility, 'illegitimacy', and the nature of households and domestic groups among single African women in Durban. South African Journal of Sociology, 24(3), 63-71.

Ramphele, M. (1993). A bed called home: Life in the migrant labour hostels of Cape Town. Cape Town & Johannesburg: David Phillip.

Richter, L. (2006). The importance of fathering for children. In R. Morrell & L. Richter (Eds.), Baba: Men and fatherhood in South Africa (pp. 53-69). Cape Town: HSRC Press.

Richter, L.M., Norris, S.A., & De Wet, T. (2004). Transition from birth to ten to birth to twenty: The South African cohort reaches 13 years of age. Paediatric and Perinatal Epidemiology, 18, 290-301.

Richter, L.M., Sherr, L., Adato, M., Belsey, M.A., Chandan, U., et al. (2009). Strengthening families to support children affected by HIV and AIDS. AIDS Care, 21, 3-12.

Roy, K. (2008). A life course perspective on fatherhood and family policies in the United States and South Africa. Fathering: A Journal of Theory, Research and Practice about Men as Fathers, 6, 92-112.

Russell, M. (2003). Understanding black households: The problem. Social Dynamics 29, 5-47. Sharp, J. (1994). A world turned upside down. Households and differentiation in a South African bantustan in the 1980s. African Studies, 53, 71-88.

Sonkhe Gender Justice Network. (2010). Sonkhe Gender Justice Network website. Retrieved from

Spiegel, A. (1987). Dispersing dependents: A response to the exigencies of labour migration in rural Transkei. In J. Eades (Ed.), Migrants, workers and the social order (pp. 113-129). London: Tavistock.

Swartz, S., & Bhana, A. (2009) Teenage tata. Voices of young fathers in South Africa. Cape Town: HSRC Press.

Tomkins, S. (2009). Using available family members as proxies to provide information on other family members who are difficult to reach. In D.A. Lawlor & G.D. Mishra (Eds.), Family matters. Designing, analysing, and understanding family-based studies in life course epidemiology (pp. 149-179). Oxford: Oxford University Press.

Townsend, N.W. (1997). Men, migration, and households in Botswana: An exploration of con- nections over time and space. Journal of Southern African Studies, 23,405-420.

Townsend, N. (2002). Cultural contexts of father involvement. In C.S. Tamis-LeMonda & N. Cabrera (Eds.), Handbook of father involvement: Multidisciplinary perspectives (pp. 249-277). Mahweh, NJ: Lawrence Erlbaum Associates, Inc.

Wilson, F. (2006). On being a father and poor in southern Africa today. In R. Morrell & L. Richter (Eds.), Baba: Men and fatherhood in South Africa (pp. 26-37). Cape Town: HSRC Press.


(1) Faculty of Human and Social Sciences, University of Southampton.

(2) Africa Centre for Health and Population Studies, University of KwaZulu-Natal.

(3) African American Studies Department, University of Maryland.

(4) MRC/Wits Rural Public Health and Health Transitions Research Unit, School of Public Health, University of the Witwatersrand.
Table 1

Data Collected about Fathers in Selected South African National
Household Surveys, Birth Cohort Studies, Household Panel Surveys and
Demographic Surveillance Systems (1)

Name Birth to Twenty

Year 1992-ongoing
Design Birth cohort of 3273 children born in
 the metropolitan area of Johannesburg-
 Soweto in 1990.

Data collected Child all ages: Identity of father-
about fathers figure whether father-figure is a
 resident or non-resident household

 Children aged 18 years: Identity of
 biological father, survival status of
 biological father, financial support by
 biological father and amount of support,
 contact with biological father, co-
 residence with biological father,
 biological father responsibilities with
 regards to child rearing.

 Children aged 15 years: Feelings towards
 being a father

 Paternity: Information about
 childbearing is collected prospectively.
 Individuals in the birth cohort who
 become young biologi cal fathers are
 eligible for recruitment into a separate
 study of young fathers.

Further info

Name South Africa Demographic and Health
 Survey (2)

Year 1998
Design Nationally representative household
 survey of 12,000 women between the ages
 of 15 and 49 years.

Data collected The survey administers three
about fathers questionnaires: a Household
 Questionnaire, a Woman's Questionnaire
 and an Adult Health Ques tionnaire.
 Households surveyed include at least one
 eligible woman aged 15-49 years. Only
 resident household members and vis itors
 on night prior to survey are recorded.
 In every second household, all men and
 women aged 15 and above were selected to
 receive an Adult Health Questionnaire.

 Child 15 years and younger: Survival
 status of biological father, identity of
 biological father where father is a
 household mem ber.

 Identified resident fathers: Demographic
 information collected during the DHS
 household members questionnaire. A sub-
 sample of identified fathers will be
 among the men selected for to
 participate in an adult health
Further info

Name General Household Survey (GHS) (3)
Year 2002-2007

Design Annual national household survey
 conducted by Statistics South Africa.
 The 2002 GHS included 26,287 households
 in which only individuals who were
 resident at least four nights a week
 were included.

Data collected All individuals: Survival status of
aboutfathers father (biological or social not
 specified), identity of father where
 father is a resident co-mem her of

 Identified resident fathers: Demographic
 and socio-economic information collected
 on individuals in the GHS.

Further info http://www/statssa/gov/za/additional
 services/quest archive social.asp

Name Africa Centre Demographic Information
 System (ACDIS)

Year 2000-ongoing
Design Demographic surveillance system of
 approximately 11,000 households resident
 in an area of northern KwaZulu-Natal

 Longitudinal bi-annual follow-up of
 approximately 89,000 resident and non-
 resident household members. Routine data
 collection focuses on demographic and

 health information. Special surveys are
 conducted periodically including socio-
 economic, HIV and sexual behaviour

Data collected Individuals of all ages: Survival status
about fathers of biological fathers, identity of
 biological father where father is a
 resident or non-resident member of

 For children 18 years and younger:
 Person with primary responsibility for
 school fees (option to report father)
 Person responsibility for day-to-day
 care (option to report father)

 Identified co-member fathers:
 Longitudinal demographic and socio-
 economic information collected by ACDIS.
 A sub-sample of fathers 15 years and
 older will have been interviewed once or
 more in special HIV and sexual behaviour
 surveys. In the 2003-4 sexual behaviour
 survey information about paternity
 (number of children ever born) was asked
 of men 15-54 years.
Further info

Name Agincourt DSS (AHDSS)
Year 1992-ongoing
Design Demographic surveillance system of
 approximately 82,000 members of
 households in the Agincourt sub-
 district of the Bush buckridge district
 of Mpumalanga province. Longitudinal
 annual follow-up of households with
 routine demographic and health data
 collection. Special data collection
 modules are included in the annual
 survey including modules on adult
 health, child grants, and fathering.

Data collected Individuals all ages: Identity of
about fathers biological father where father is
 resident or non-resident member of
 household and is the head of the

 For children aged 0-17 years: In the
 annual census conducted in 2007 and
 2008, information was collected about
 fathers including whether the father is
 biological or social, the survival
 status of fathers, current and previous
 locations of father, contact with father
 in past 30 days, father support for
 school fees in past 30 days, existence
 of legal order mandating father to pay
 child sup port.

 Identified co-member fathers: Collected
Further info

Name KwaZulu-Natal Income Dynamics Study

Year 1993, 1998 and 2004
Design Household panel survey of households in
 the province of KwaZulu-Natal. Fieldwork
 has been conducted in 1993, 1998 and
 2004, with respondents interviewed in
 865 households in most recent wave. The
 focus of study is to measure trends in
 poverty and changes in the socio-
 economic condition of households.
 Resident and non-resident household
 members are recorded in KIDS.

Data collected All household members: Identity of
about fathers father where father is a household
 member. Survival status of father.

 Resident child under 21 years:
 Responsibility for day-to-day care
 (option to report father). All
 residents: Would their father would
 assist if household experienced
 financial trouble, fathers' employment
 and education sta tus, financial and
 material contributions of household by
 amount and type (option to report
 information about fathers within or out
 side the household that contribute to
 any member of the household). Questions
 about contributions by fathers do not
 specify whether the term 'father'
 applies only to biological fathers or
 includes adopted, foster and step-
Further info,0,0,0,0

Name Cape Area Panel Study (CAPS)
Year 2002,2003,2005,2006
Design Panel study of youths and young adults
 in metropolitan Cape Town. In 2002, the
 first wave conducted individual and
 household interviews with 4,800 young
 people aged 14-22 years. Subsequent
 waves have continued to follow the youth
 sample, as well as conduct interviews
 with their parents and older adults
 within their households. CAPS collects a
 wide range of information in cluding
 schooling, employment, health and family

Data collected All household members: Identity of
about fathers biological / step-father / adoptive /
 foster father where father is co-
 resident household member, survival
 status of biological father. This
 information is updated at re-interviews
 of the household.

 All household members aged 14 years and
 older: Number of biological children
 alive and living elsewhere

 Youth sample only: Information is
 updated at each interview about the
 presence of biological and social
 fathers within the housh old. If
 biological or social fathers previously
 identified are no longer co-members,
 information is collected about the
 reasons for departure and their current
 employment status. Similarly information
 about the arrival of a youth's father in
 household is recorded. Person who helps
 young person with homework (option to
 record father), person with primary
 responsibility for school fees (op tion
 to record father).

 Fathers of youth sample: In wave 3, a
 sub-sample of parents (fathers or
 mothers) of youth sample were asked to
 participate in an interview. Data was
 collected about their quality of life,
 attitudes to education, their
 expectations about the index adolescents
 fu ture education, employment and
 marriage, the quality of the parent and
 index youth relationship, important
 influences in their own life, and HIV-
 AIDS. In addition to questions about the
 index youth (17-25 years), parents were
 asked about whether they had children 7-
 16 years and if so, questions were asked
 about one child in this age group.
Further info

Name National Income Dynamics Survey (NIDS) (4)
Year 2008
Design NIDS is a national household panel
 survey. The first wave was conducted in
 2008 with interviews with respondents in
 7,300 households. Resident and non-
 resident household members are recorded.
 Data collection will occur every two
 years. A primary focus of NIDS is
 household income and expenditure and
 includes information about intra-and
 inter-household transfers includ ing
 those by non-resident fathers.

Data collected Individuals all ages: Identity of
about fathers biological father where father is a
 resident or non-resident household
 member, survival status of biological
 father, if father is dead / year of

 Children 15 years and younger: Contact
 with biological father, whether
 financial support given by biological
 father, status of re lationship between
 child's biological parents. Person with
 primary responsibility for child (option
 to report father), other people involved
 in care of the child (option to report
 father). The NIDS questionnaire
 specifically notes that for the previous
 two items the term 'father' would be
 valid for biological, adoptive, foster
 or stepfathers.

 Adult household members: Amount and type
 of financial contributions made for a
 child within or outside the household.

Further info

General notes: In longitudinal studies the data collected about
fathers may differ from wave to wave and we note this where possible.
We have not included information that is collected using questions
phrased in relation to "parents" more generally, for example in the
case of a question 'Who chose the school you are currently attending?'
where the option is 'My parents' without distinction between mother or

In some surveys, the questionnaires, published papers or available
supporting documents do not describe the way in which the term
"father" was explained by the fieldworker to the respondent. We note
this in the table whilst recognising that those involved in
administering the questionnaire may have used a specific definition in
training and fieldwork.

(1) A third DSS, the Dikgale DSS is conducted by the University of the
North in the Northern Province. We have not included a review the
Dikgale DSS given the strong focus on nutritional research in
published studies that use data collected in the Dikgale DSS.
Furthermore, given the highly selected households and children
enrolled we also do not include studies based on surveys or cohorts
that were established with a very speci fic research focus, for
example, the Transitions to Adulthood in the Context of HIV-AIDS or
the KwaZulu-Natal Longitudinal Study of Orphan Welfare.

(2) The second South African Demographic and Health Survey was
conducted in 2003. This sur vey included an interview with men in
which questions about their paternity histories were asked. However,
because the 2003 survey data has not been available we have not
included the survey in our summary table.

(3) Two other repeated national surveys have been conducted by
Statistics South Africa: the an nual October Household Surveys (OHS)
(1993-1999) and the biannual Labour Force Surveys (LFS) (2000-ongoing).
The Project for Statistics on Living Standards and
Development (PSLSD) was another national household survey conducted by
SALDRU in 1993. The information avai lable on fathers and the approach
to defining households are generally similar in all these sur veys to
that collected in the General Household Survey i.e. the biological
father is identified if a member of the household and some limited
information was collected about financial contri butions of fathers to
the household. A difference between the PSLSD and the OHS, LFS and GHS
is the inclusion of resident and non-resident household members. We
have included only the GHS in the table as being broadly
representative of these types of national household surveys.

(4) Another panel study (PSLSD) conducted in 1993 is not included in
the table.

Table 2

Measures of Father Involvement for Improved Data Collection in
Household Surveys Related to a) Fathering and Fatherhood for
Men in Study Households, and b) Father Identity and Involvement
for Index Children in Study Households


Paternity history

History of adoptive, foster, step-father relationships
with children


Identification of biological or social children within
the household

Characteristics of biological or social children outside
the household

Relationship history with child's biological mother
and/or primary caregiver

Measures of accessibility

e.g., Frequency of contact with the child

Measures of engagement

e.g., Shared activities (meals together, visiting
relatives together)

Measures of responsibility

e.g., Role in decision-making (health care, schooling);
financial support


Identity of biological or social father within the
household (2)

Survival status of biological and social fathers (date of
death, cause of death)

Characteristics of fathers outside the household (age,
marital status, education, employment, health, place of
residence) (3)

Father relationship history with child's biological mother
and/or primary caregiver

Measures of accessibility

e.g., Frequency of contact with the father

Measures of engagement

e.g., Shared activities (meals together, visiting
relatives together)

Measures of responsibility

e.g., Fathers' role in decision-making (health care,
schooling); financial support

(1) Given the current approaches used in household
surveys, this would be asked only of fathers who are members
of the index household. However, as we discuss in the paper,
information about fathers outside the household is also be

(2) South African surveys differ in the inclusion of resident
and non-resident household members. In sources such as ACDIS,
members of the household include resident and non-resident men.
Thus, providing that a non-resident father is a member of
the same household as the child, information
recorded about his characteristics will be available.

(3) Similar data do not need to be collected for fathers who
are members of the household and linked to the child unless
not collected as part of the main survey or study.
COPYRIGHT 2012 Men's Studies Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Hosegood, Victoria; Madhavan, Sangeetha
Geographic Code:60SUB
Date:Sep 22, 2012
Previous Article:Disadvantaged, urban fathers' health and risky health behaviors: the role of relationship status and fathering practice.
Next Article:"Why aren't you at work?": negotiating economic models of fathering identity.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters