Abstract
This paper examines data on teenage fertility and patterns of uptake of the Child Support Grant in South Africa from 1998 to 2005, to assess how far this Grant is associated with the trend in teenage childbearing. Teenage fertility was fairly high during the 1980s when state financial assistance to teenage mothers did not adequately serve the majority of South Africans. Since the first half of the 1990s, however, teenage fertility has steadily declined. This trend was already underway when the grant was expanded in 1998 to reach beneficiaries in all sub-groups of the national population. If teenage girls were having children primarily to benefit from the Child Support Grant, then more would be making claims than is in fact the case. The findings of this study do not suggest any significant positive association between the grant and the trend in teenage childbearing in South Africa during the past decade.
1. Introduction
In many parts of the world, teenage childbearing is seen as a social problem, for a number of reasons. First, it has been associated with physiological risks to the mother and child (McGrew & Shore, Citation1991; Hoffman et al., Citation1993; Richter et al., Citation2006). It is therefore helpful to differentiate between pre-teen and very young teen mothers, for whom childbirth presents a high risk of physiological problems, and mothers in their late teens, for whom the risk is lower.
Second, a number of social and economic disadvantages tend to be strongly associated with teenage childbearing (Manzini, Citation2001; Hao et al., Citation2007). Teenage fertility is believed to set the young mother and her child on a trajectory of lifetime poverty (Card & Wise, Citation1978; Makinson, Citation1985; Butler & Burton, Citation1990; Geronimus & Korenman, Citation1992). Poverty may, however, be the result of mediating factors, such as expulsion or exclusion from educational facilities or a lack of material and social support (Ojwang & Maggwa, Citation1991). For instance, although South African Education Policy allows adolescent mothers to attend school during pregnancy and after giving birth, about half of them drop out of school after falling pregnant (Manzini, Citation2001; Kaufman et al., Citation2001).
Third, in many societies today, teenage childbearing takes place mainly outside of marriage – a phenomenon that raises religious, cultural and pragmatic concerns, such as poor support from the father of the child (Gordon, Citation1997; National Campaign to Prevent Teen Pregnancy, Citation2001). Another issue, often raised by demographers, is that early childbearing is generally associated with a higher overall fertility rate in societies with low rate of contraceptive prevalence (see Hobcraft & Little, Citation1984; Bongaarts, Citation1985).
In most low-fertility countries, debates about teenage childbearing are usually related to the patterns of use of the national welfare system by teenagers. For instance, an often quoted trend in the US is that approximately 50 per cent of teenage mothers go on welfare within one year and 77 per cent within five years of having a child (Burt et al., Citation1984; Rodriguez & Moore, Citation1995). Such debates are rare in developing countries, which do not generally have a comprehensive welfare system that targets children. The introduction of the Child Support Grant in South Africa has sparked the debates about welfare and childbearing behaviour that are common in more developed countries with a relatively high level of teenage fertility. This paper considers what insights we can gain from available evidence about the relationship between the Child Support Grant and the pattern of teenage childbearing in South Africa.
The issues raised here are important for debates about the direction of social policy in South Africa. One way to proceed would be to do an analytical review and in the process propose and defend arguments on the basis of the available local and international research. However, given the underdeveloped state of research on the interactions of demographic trends and social policy in South Africa, we take what we consider a more promising approach. We examine the scant empirical information that currently exists in order to get some sense of the arguments it can possibly support. This approach of course limits the scope of the analysis to the available demographic and statistical evidence. This study therefore does not engage in detail with the perspectives and arguments highlighted in the analysis or elsewhere in the local literature.
2. Data and method
The question of whether or not the Child Support Grant has a significant influence on teenage childbearing can be more thoroughly addressed with sets of data that include individual-level information on both variables. Unfortunately, such data do not yet exist for South Africa. Constrained by a lack of detailed information, this study makes use of national administrative statistics on the Child Support Grant and secondary estimates of teenage fertility.
2.1 The Child Support Grant
The Child Support Grant is the successor to the State Maintenance Grant, which was initially designed for whites but later extended to other racial groups. Although theoretically the State Maintenance Grant was available for all races, most beneficiaries were whites and coloureds. In the former homelands, where Africans mostly lived, this Grant was poorly implemented. Even among whites and coloureds, it targeted only single parents.
The Child Support Grant was implemented in its current form in 1998 following the recommendation of the Lund Commission (see Lund, Citation2008). A major difference between the old State Maintenance Grant and the Child Support Grant is that the latter is for all poor children from every type of family. In particular, it targets historically neglected communities, especially those in rural areas. The amount paid for each child per month by the Child Support Grant has risen from its original R70 to its current value of R250 (for 2009/11).
Eligibility for the Child Support Grant is determined on the basis of specific demographic and socioeconomic criteria. Currently, more than three million children aged 15 or younger are beneficiaries. However, although it is designed for all children, for administrative, technical and logistic reasons not all currently benefit. In some cases caregivers do not have the requisite documents for the children, such as birth certificates and national identity documents, and this prevents them from accessing the grant.
The data for the Child Support Grant used in this study were obtained from the state electronic database which records all government's welfare transactions. These data (SOCPEN) are available from 1999. A major limitation of this source of data is that information is released only in an aggregate format. For confidentiality and other reasons, data made available for this study are aggregated by age group of recipients and year of benefit. Other demographic and socioeconomic characteristics of children and grant recipients, including the characteristics of the biological mother, would permit more in-depth analysis but are unfortunately not available for public use from SOCPEN.
2.2 Data on teenage childbearing
Teenage childbearing is quantitatively defined as the teenage specific fertility rate or the number of births per 1000 women aged 15–19 in a specified period of time. It is a component of the total fertility rate, i.e. the number of children a hypothetical woman would have during her lifetime if she conforms to the current fertility rates of women in different reproductive age groups.
A feature of reproductive behaviour in South Africa that could be highlighted as a methodological issue in this study is the fact that almost all teenage childbearing takes place outside a recognised marital union. The contribution to teenage fertility by currently married teenage girls is negligible. This is mainly a combined effect of a low prevalence of marriage and a high rate of sexual activity among South African teenagers. By age 19, close to 80 per cent of South African women have had sex, and about 37 per cent have been pregnant (Shisana & Simbayi, Citation2002; Reddy et al., Citation2003; Pettifor et al., Citation2004). Measurement and discussion of teenage fertility for South Africa and a number of other countries in southern Africa are therefore by implication concerned with non-marital childbearing. This is in contrast to the pattern in other sub-regions of Africa where most teenage childbearing takes place within marriage.
Statistics on teenage childbearing are often contested, partly because they are among the most difficult to collect accurately in demographic studies in South Africa (see Moultrie & Timaeus, Citation2003). For this study, estimates of teenage fertility were collected from various sources that include censuses, national surveys and independent estimates by demographers. To improve the quality of the evidence, means of estimates of teenage fertility were computed for each period of interest, on the assumption that individual estimates may not be robust enough to support definitive conclusions about trend. These averages graduated the estimates of teenage specific fertility rates effectively without necessarily introducing significant distortions in the observed trend.
3. Results
3.1 Teenage childbearing in South Africa
Trends in teenage specific fertility in South Africa are shown in . The level was relatively high in the first half of the 1980s when the rate was 103 per 1000. This was followed by a slight increase to a level of 106 per 1000 by 1990, since an exceptionally high rate of teenage specific fertility rate was reported for the second half of the 1980s (Mostert, Citation1990), based on a 1987–89 Demographic and Health Survey. Data from 1991 onwards show a declining trend, as can be seen in . More recent and detailed statistics on the level of teenage childbearing have yet to be made available. Statistics South Africa (2008) produced an estimate of teenage fertility from the 2007 Community Survey which is apparently lower than the 56 per 100 recorded for the 2000–05 period.
Figure 1: Trends in teenage specific fertility rates 1980–2009
3.1.1 Changes in teenage fertility, 1998–2003
shows the percentage change in the teenage specific fertility rate in the five years from 1998 to 2003 for the provinces, population groups and rural and urban areas, using two data sets that encapsulate the study's period of interest fairly. The total South African teenage specific fertility rate recorded a 30 per cent decline during this period. Most provinces experienced a similar scale of decline, with the exception of the Free State, which declined by 7 per cent. The North West Province, however, showed a small increase of 3 per cent. All population groups experienced a decline, and the fertility rate declined in both urban and rural areas (see Moultrie & McGrath, Citation2007, for rural KwaZulu-Natal; Garenne et al., Citation2007, for rural Mpumalanga). In the rural areas, it declined by 33 per cent, from 99 per 1000 in 1998 to 66 per 1000 in 2003. The declining trend in rural fertility is important in this study because there are more Child Support Grant beneficiaries in the rural areas.
Table 1: Decline in teenage fertility, 1998–2003
3.1.2 Teenage childbearing and the national total fertility rate
The technical relationship between teenage fertility and the overall total fertility can easily be misunderstood, especially in a superficial analysis of trends. The observed declining trend in teenage specific fertility rate took place in the context of a declining national total fertility rate. However, total fertility did not experience any slight increase in the early 1990s. It has declined consistently from about 5.0 in the early 1980s to a current level of 2.8. However, there is no guarantee that all other childbearing age groups are experiencing a similar monotonic decline in fertility, or that their speed of fertility decline is the same as that of teenage girls. A full demographic picture of the relationship between teenage childbearing and the total fertility rate would require a more advanced demographic analysis. This is outside the scope of this study.
3.2 Use of the Child Support Grant
shows the estimated percentages of children who received the Child Support Grant in each province in 1999 and 2005, calculated from information about the number of children who received the grant and the projected number of all potential recipients in South Africa in each year. The estimates for 1999 were calculated using the projected number of eligible children aged 0–7, and those for 2005 using the projected number of all eligible children (0–14).
Table 2: Estimated percentages of children receiving the Child Support Grant in each province, 1999 and 2005
About 45 per cent of all children in South Africa were estimated to be receiving the grant by March 2005. This was a dramatic improvement in coverage from a low level of only 0.27 per cent in 1999. The percentages of recipients differed significantly according to province. Most lived in rural areas. Limpopo, Mpumalanga and Eastern Cape had relatively high percentages of recipients, and the lowest percentages were in the metropolitan areas of the Western Cape and Gauteng.
3.2.1 Age distribution of beneficiaries
shows the estimated age distribution of Child Support Grant beneficiaries in 1999 and 2005. The slight increase for the 15–19 age group from 1.64 per cent to 2.69 per cent may well be a function of improved coverage and access. A similar increase is observed for mothers in their twenties. The percentage of women aged 35–49 and 50 and older who received the grant in 2005 was lower than in 1999.
Table 3: Estimated age distribution of Child Support Grant beneficiaries in March 1999 and March 2005
3.2.2 Child Support Grant beneficiaries and contributions to the level of fertility
shows that since 1999 fewer than 3 per cent of recipients of the grant have been teenagers, yet their contribution to the total fertility rate in 2005 was as much as 15 per cent. It is noteworthy that below the age of 30 there were fewer beneficiaries than women who gave birth in that year, whereas the reverse was the case for those aged 30 and older. These statistics agree with the patterns of child care in South Africa where, in a good number of cases, older persons care for the children of teenage mothers. It is also a regulation of the Child Support Grant that older persons who are the main caregivers have a right to access the grant, instead of their biological mothers.
4. Discussion
Do the available data suggest a positive association between the introduction of the Child Support Grant and the trend in teenage childbearing in South Africa? Specific evidence analysed in this study supports two straightforward arguments. First, some studies recorded a slight increase in the teenage specific fertility rate for the early 1990s, before the introduction of the grant, whereas data for subsequent periods show a declining trend. This declining trend occurred in the context of a consistent decline in the national total fertility rate. There is no visible evidence that the grant had a stalling effect on the trend of teenage fertility.
Secondly, the administrative data examined here do not suggest that the grant is the primary reason for childbearing by teenage girls. In the period analysed, teenage mothers who benefited directly from the grant were fewer than 3 per cent, compared to their 15 per cent contribution to the national total fertility rate. If teenagers planned to have children primarily to claim the grant, one would expect the percentage of those doing so would be closer to the actual percentage of teenage mothers, after allowing for non-eligibility and other factors. Even if we take into consideration the fact that some of the beneficiaries are not the biological mothers, but instead grandmothers or foster mothers, the number of recipients in these categories is not large enough to make up for the wide differences between the percentage of teenage mothers receiving benefits and their proportional contribution to the total fertility rate.
It is to be expected that some non-teenage mothers continue as recipients so long as the children they had as teenagers remain eligible for the grant. The highest possible age for a claimant in 2005 who had a child as a teenager would be 34. However, it is quite unlikely that many teenage recipients would fall into this category. It is common for a woman to have a first child early (possibly as a teenager) and then to allow a very long interval before the next child, by which time she and her children are likely to fail the means test for receiving the grant. The point here is that a recipient aged 35 or older is unlikely to have been a biological teenage mother. This further corroborates the statistical gap between the percentage of teenage mothers who claim the grant and their contribution to the total fertility rate.
At the same time, the data suggest that women aged 35 and older are more likely to be direct beneficiaries of the grant than teenage mothers. According to the Child Support Grant guidelines, the child's main caregiver should be the recipient of the grant. In many cases in South Africa, grandparents are the main caregivers of children born to teenage mothers (see for instance Makiwane et al., Citation2004). The trend in uptake of the grant confirms this pattern. The proportion of teenage mothers who actually receive the grant is lower than the estimated percentage of all teenage mothers. On the other hand, the data show that older women, including some past childbearing age, are more likely to be recipients of the grant than teenage mothers.
Another piece of evidence that supports the argument put forward by this study is the increasing number of pregnancies that are unplanned and unwanted by women. shows a consistently growing number of pregnancies terminated by women of all age groups in state health institutions (from 46 188 in 1999 to 87 036 in 2003). Statistics are not available about abortions conducted in private institutions, but a report released by Marie Stopes (the largest private institution providing abortion) for 2007 shows an increase in the number of abortions performed on women aged 18 and younger (V Trudy, Coordinator for Marie Stopes clinics, personal communication, 29 January 2008).
Figure 2: Overall fertility trends in South Africa
Figure 3: Number of abortions conducted in public institutions per year
Statistics provided by the National Department of Health show that 10.4 per cent of all abortions performed in state institutions in 2003 were for women younger than 18. This is considerably higher than contribution of mothers younger than 18 to the total fertility rate in the same period. Since termination of pregnancy is free in state institutions, we would expect the statistics here to reflect largely the trend among poorer sections of South African women. These are the same sub-group of women who could easily keep the pregnancies and begin to claim the Child Support Grant when the child is born. In effect, if the grant encourages childbearing among teenagers, one would expect this to be reflected in a reduction, not an increase, in the number of pregnancies terminated by poorer and younger women.
4.1 The debate about teenage childbearing in South Africa
Most researchers and policy-makers do not find any defensible rationale for actively encouraging non-marital teenage childbearing. This is true even in countries with a very low level of fertility that have generous incentives to raise the level. It is also true for high-fertility countries that use the potentially negative consequences of non-marital teenage childbearing as grounds for anti-natalist population programmes.
Chimere-Dan Citation(1993) identifies economic, cultural and reproductive health perspectives in the debate about non-marital teenage childbearing in South Africa. First is the economic perspective, i.e. the economic costs to society of personal reproductive behaviour. It emphasises a concern that cash transfers to unmarried teenage mothers would be an unproductive investment, and could serve as an incentive for them to have children that they are unable to support. Since for many decades in the past there was no comprehensive welfare assistance programme that targeted the majority of the disenfranchised population, this perspective was carefully subsumed under general arguments about the negative consequences to society of a high level of fertility and was seen as part of the rationale for an interventionist population policy that was implemented by South Africa until recently.
In South Africa, as in many other countries, childbearing out of wedlock is frowned upon on cultural and moral grounds. Against this background, Preston-Whyte (Citation1990, Citation1993) and Preston-Whyte and Zondi Citation(1992) introduce a cultural perspective in an attempt to understand the interplay of culture, morality and reproductive behaviour in an apartheid society. They draw attention to emerging factors that appeared to weaken the cultural and moral stigma attached to childbearing outside marriage in South Africa, arguing that for an African household in the apartheid era, it was not the end of the world if an unmarried teenager became pregnant. They extend this thesis by arguing that non-marital teenage childbearing could in fact be a calculated sociocultural, if not necessarily economic, survival strategy for the teenager in an oppressive (apartheid) society. To an extent, this perspective was persuasive in the context of political and socioeconomic policies and practices that discriminated against some sub-groups of the population and favoured others. The past decade has witnessed fundamental transformations that promise very good socioeconomic opportunities to previously disadvantaged girls and women in general. If this viewpoint is to retain its explanatory power in South Africa today, it will have to be reformulated to accommodate the actual and potential impacts of these transformations.
A third perspective understands the challenges as an expression of unmet reproductive health needs. It sees non-marital teenage childbearing as evidence of inadequate access to effective reproductive health facilities for teenagers. Researchers taking this perspective argue that most childbearing by unmarried teenage girls is unplanned and unwanted. Richter et al. Citation(2005) show that as much as 70 per cent of teenage pregnancy is unplanned. Support for this perspective is high, especially in the context of HIV/AIDS prevention and management programmes among young people in South Africa.
To understand the state of the debate about teenage childbearing in South Africa, we need to separate popular concerns from other considered perspectives that may have empirical bases. There is a widespread public perception based on moral and cultural concerns which argues that teenage fertility has increased as a result of the introduction of the Child Support Grant. Moral aversion to out-of-wedlock fertility underpins this perception and most negative sentiments towards the grant in South Africa. Welfare grants are seen by some as an incentive for female teenagers to have children that they are unable to support otherwise (see Sawhill, Citation2000; Evans, Citation2006). This perspective is also routinely reported in the popular media in South Africa. An example of this outcry appeared in the Cape Argus in 2008, citing community workers who said:
Lazy young mothers went on drinking sprees, bought clothing and gambled with the [Child Support Grant] money … There is a dramatic increase in young mothers misusing the child support grant, sometimes known as ‘womb fee’ in the township … (Cape Argus, Citation2008:6)
This popular outcry against the grant also feeds on an empirically untested argument that the grant does not benefit children because teenagers deliberately give birth in order to access the money for their personal use. A logical outcome of the above claim, if true, would be a dramatic increase in teenage fertility. A community member was quoted as saying ‘more babies will be born as long as the child grants do not stop’ (Matsie, Citation2009). Teen mothers are supposed to be benefiting from the grant themselves. However, there are immediate and long-term social costs of motherhood which far outweigh any benefit they might receive from it.
Some researchers suggest that even if no such perverse incentive is found, some form of punitive withdrawal of welfare from unwed teenage mothers is recommended as a signal of disapproval of teenage and out-of-wedlock fertility. For example, in both the US and the UK there had been strong proposals to exclude sections of teenage mothers or to withdraw their grants on these grounds (Geronimus, Citation1997; Bullen & Kenway, Citation2000). In the US, a law was passed in 1996 restricting welfare grants, as a way of reducing teenage and out of wedlock childbearing. These restrictions include a condition that teenage mothers under the age of 18 must live with their parents and remain in school and denial of welfare to any mother of any age who bears a child while already receiving welfare (Hao et al., Citation2007). The issues raised by the moral argument as highlighted here are important but perhaps need not run ahead of availability of scientific data. Until detailed data become available, many cardinal issues in the understanding of personal demographic behaviour that have important costs to society may not be fully addressed in the South African context.
5. Conclusion
The aim of this study was to examine existing data on the interaction of demographic trends and social policy as a way of laying a more scientific foundation for research and debates about the issues as better data become available. The data examined here do not show a positive association between the introduction and use of the Child Support Grant and the trend in teenage childbearing. An observed slight increase in teenage fertility in the late 1980s predated the introduction of the grant. From the first half of the 1990s teenage fertility has been on a declining trend, before and after the introduction of the grant. There is no apparent positive association between the grant and the empirical trend in teenage fertility. While national debates about social policy issues need to be informed by international experience and best practice, it may not be very helpful to import or enforce preconceptions derived from other societies without careful attention to the South African social context, and, perhaps in this case more importantly, the prevailing empirical situation and statistical trends. This paper contributes an important note of caution to debates about demographic trends and social policy in South Africa.
Notes
References
- Bongaarts , J . 1985 . A framework for analysing the proximate determinants of fertility in a population with natural fertility . Population and Development Review , 4 : 105 – 31 .
- Bullen , E and Kenway , T . 2000 . New Labour, social exclusion and educational risk management: The case of gymslip mums . British Educational Research Journal , 26 ( 4 ) : 441 – 56 .
- Burt , M , Kimmich , M , Goldmuntz , J and Sonenstein , F . 1984 . “ Helping pregnant adolescents: Outcomes and costs of service delivery ” . In Final Report on the Evaluation of Adolescent Pregnancy Programs , Washington, DC : The Urban Institute .
- Butler , J and Burton , L . 1990 . Rethinking teenage childbearing: Is sexual abuse a missing link? . Family Relations , 30 : 73 – 80 .
- Cape Argus . 2008 . Teenage mothers abuse state child grant . 29 September : 6
- Card , J and Wise , L . 1978 . Teenage mothers and teenage fathers: The impact of early childbearing on the parents' personal and professional lives . Family Planning Perspectives , 10 ( 4 ) : 100 – 205 .
- Chimere-Dan , O . Racial patterns of fertility decline in South Africa . Proceedings of the 22nd International Conference of the International Union for Scientific Study of Population (IUSSP), Montreal . December 10–11 . pp. 43 – 51 .
- Chimere-Dan , O . 2008 . Teenage childbearing in South Africa: Trend and explanations . Africa Strategic Research , A document prepared for the Africa Indicator Research Group
- DoH (Department of Health) . 2001 . South African Demographic and Health Survey 1998 , Pretoria : Government Printers .
- DoH (Department of Health) . 2006 . South African Demographic and Health Survey 2003 , Pretoria : Government Printers .
- Evans , D . 2006 . “ The impact of child support enforcement policy on non-marital childbearing ” . Evans School Working Papers Series Seattle, WA : Evans School of Public Affairs, University of Washington .
- Garenne , M , Tollman , S , Collinson , M and Kahn , K . 2007 . Fertility trends and net reproduction in Agincourt, rural South Africa. 1992–2004 . Scandinavian Journal of Public Heath , 35 ( 3 ) : 68 – 76 .
- Geronimus , T . 1997 . Teenage childbearing and personal responsibility: An alternative view . Political Science Quarterly , 112 ( 3 ) : 405 – 30 .
- Geronimus , T and Korenman , S . 1992 . The socio-economic consequences of teen childbearing reconsidered . Quarterly Journal of Economics , 107 ( 4 ) : 1187 – 214 .
- Gordon , L . 1997 . “ Teenage pregnancy and out-of-wedlock birth: Morals, moralism, experts ” . In Morality and Health , Edited by: Brandt , A and Rosen , P . New York : Routledge .
- Hao , A , Stone , N and Cherlin , A . 2007 . Effects of child support and welfare policies on non-marital teenage childbearing and motherhood . Population Research and Policy Review , 26 : 235 – 57 .
- Hobcraft , J and Little , A . 1984 . Fertility exposure analysis: A new method of assessing the contribution of proximate determinants to fertility differentials . Population Studies , 38 ( 1 ) : 21 – 46 .
- Hoffman , S , Foster , E and Furstenberg , E . 1993 . Re-evaluating the costs of teenage childbearing . Demography , 30 ( 1 ) : 1 – 13 .
- Kaufman , C , De Wet , T and Stadler , J . 2001 . Adolescent pregnancy and parenthood in South Africa . Studies in Family Planning , 32 ( 2 ) : 147 – 60 .
- Lund , F . 2008 . The Child Support Grant in South Africa , Pretoria : Human Sciences Research Council (HSRC) .
- Makinson , C . 1985 . The health consequences of teenage fertility . Family Planning Perspectives , 17 ( 3 ) : 132 – 9 .
- Makiwane , M , Schneider , M and Gopane , M . 2004 . Experiences and needs of older persons in Mpumalanga . A report compiled by the Human Sciences Research Council (HSRC) on behalf the Mpumalanga Provincial Government
- Manzini , N . 2001 . “ Sexual initiation and childbearing among adolescent girls in KwaZulu-Natal, South Africa ” . In Reproductive Health Matters: By and For Women and Men Vol. 9(17) , 44 – 5 .
- Matsie , L . 2009 . Child grants in South Africa: Incentive to have more children? . The African: Our Voices, Our Vision, Our Culture , http://www.africamag.com/FORUM-809-design004 Accessed 31 August 2009
- McGrew , M and Shore , W . 1991 . The problem of teenage pregnancy . Journal of Family Practice , 32 ( 1 ) : 17 – 25 .
- Mostert , W . 1990 . “ Recent trends in fertility ” . In South Africa's Demographic Future , Edited by: Mostert , W and Lotter , J . Pretoria : Human Sciences Research Council (HSRC) .
- Moultrie , T and McGrath , N . 2007 . Teenage fertility rates falling in South Africa . South African Medical Journal , 97 ( 6 ) : 442 – 3 .
- Moultrie , T and Timaeus , I . 2003 . The South African fertility decline: Evidence from two censuses and a Demographic and Health Survey . Population Studies , 57 ( 3 ) : 265 – 83 .
- National Campaign to Prevent Teen Pregnancy . 2001 . “ Halfway there: A prescription for continued progress in preventing teen pregnancy ” . Washington, DC : National Campaign to Prevent Teen Pregnancy . http://www.teenpregnancy.org Accessed 31 August 2009
- Ojwang , S and Maggwa , A . 1991 . Adolescent sexuality in Kenya . East African Medical Journal , 68 ( 2 ) : 74 – 80 .
- Pettifor , A , Rees , H , Steffenson , A , Hlongwa-Madikizela , L , Macphail , C , Vermaak , K and Kleinschmidt , I . 2004 . “ HIV and Sexual Behaviour among South Africans: A National Survey of 15–24 Year Olds ” . In Reproductive Health Research Unit , Johannesburg : University of the Witwatersrand .
- Preston-Whyte , E . 1990 . “ Qualitative perspectives on fertility trends among African teenagers ” . In South Africa's Demographic Future , Edited by: Mostert , W and Lotter , J . Pretoria : Human Sciences Research Council (HSRC) .
- Preston-Whyte , E . 1993 . Women who are not married: Fertility, illegitimacy and the nature of households and women in groups among single African women in Durban . South African Journal of Sociology , 24 ( 3 ) : 63 – 71 .
- Preston-Whyte , E and Zondi , M . 1992 . “ African teenage pregnancy: Whose problem? ” . In Questionable Issue: Illegitimacy in South Africa , Edited by: Burman , S and Preston-Whyte , E . Cape Town : Oxford University Press .
- Reddy , S , Panday , P , Swart , D , Jinabhai , C , Amosun , S , James , S and Monyeki , K . 2003 . ‘Umthenthe uthaba usimila’: The South African Youth Risk Behaviour Survey 2002 , Cape Town : South African Medical Research Council .
- Richter , L , Emmet , T , Makiwane , M , Du Toit , R , Brookes , H , Potgieter , C , Altman , M and Makhura , P . 2005 . Status of the Youth Report: Produced on Commission to the Umsobomvu Youth Fund , Pretoria : Human Sciences Research Council (HSRC) .
- Richter , L , Norris , S and Ginsburg , C . 2006 . The silent truth of teenage pregnancies: Birth to twenty cohort's next generation . South African Medical Journal , 96 ( 2 ) : 122 – 4 .
- Rodriguez , C and Moore , N . 1995 . Perceptions of pregnant/parenting teens: Reframing issues for an integrated approach to pregnancy problems . Adolescence , 30 : 685 – 706 .
- Sawhill , I . 2000 . Welfare reform and reducing teen pregnancy . http://findarticles.com/p/articles/mi_m0377/is_2000_Wntr/ai_58672899/ Accessed 30 August 2009
- Shisana , O and Simbayi , L . 2002 . “ Nelson Mandela/HSRC study of HIV/AIDS ” . Cape Town : Human Sciences Research Council . South African national HIV prevalence, behavioural risks and mass media household survey 2002
- Statistics South Africa, 2007. Statistical Release P0301. Statistics South Africa, Pretoria.
- Udjo , E . 2003 . A re-examination of levels and differential in fertility in South Africa from recent evidence . Journal of Biosocial Science , 35 ( 3 ) : 413 – 31 .