This dissertation/thesis does not contain the writings of other persons, unless specifically acknowledged as a source by other researchers. Trends show a significant decrease in nuclear-type families (families containing only close family members) while an increase in extended-type families (families containing close family members plus 'other' . relatives). Within families of the extended type, there is an increase in non-standard "skipped generation" and "complicated but related".
While extended-type households are consistently poorer than core-type households across plausible equivalence scales, poverty rankings among specific households change significantly when different equivalence scales are used. Over time, the headcount ratio for core-type households is seen to decline, while the number of people-count ratio for extended-type households appears to rise.
Introduction
The validity of Goode's convergence theory is examined extensively, with particular attention paid to South Africa. Using nationally representative data from the October Household Survey as well as the General Household Survey, household types are identified and trends are estimated by race. Special attention is drawn to the heterogeneity of household characteristics among different types of households by employing the head of the household as a proxy for the status of the household in general.
Attention is paid to the correct choice of the equivalence scale with regard to correctly correcting for child costs and economies of scale. The robustness of the relative poverty rankings between household types is evaluated based on the choice of equivalence scale.
A literature review of household structure and
Furthermore, this increase in the proportion of extended families holds for all racial groups. A trend worth noting in the literature is the rise of single-person households in South Africa. Therefore, an increase in the proportion of single-person households may provide support for the convergence theory.
Second, Wittenberg and Collinson (2007) note that an undersample of migrant hostels in the 1995 OHS (each person in these hostels would have been classified as a single-person household) may be responsible for the spurious increase in the share of single-person households. within the national datasets. In fact, it appears that there is a resurgence in the proportion of extended family households in South Africa.
Mapping changes in household composition in South Africa…
While the share of nuclear-type households decreased from 67.1% of all household types in 1995 to 62.7% of all household types in 2006, the share of extended households increased from. More specifically, there is a statistically significant increase in skip generation households from 3.7% of all household types in 1995 to. However, their low proportion of all household types makes this decline largely insignificant for extended type households in general.
Alongside the decline in nuclear-type families is a statistically significant increase in extended-type families (for Africans and whites). Among nuclear-type families, there is a significant increase in the proportion of single-person households.
A univariate and multivariate investigation of household
In fact, the heads of households of the extended type are the oldest, followed by households of the nuclear type and finally by households of the other type. Compared to nuclear-type households, all extended-type households have substantially older household heads (with the exception of complex but related households, which had an average age of household heads of 37 years in 2006). Extended-type households have more than double the percentage of household heads who are economically inactive compared to nuclear-type households in 1995.
Heads of households in core-type households have significantly higher levels of education than heads of households in extended-type households. The table shows a significant decrease in the proportion of household heads with no schooling for both core and extended type households. This is accompanied by a significant increase in the proportion of core and extended type household heads with a matric education.
One can therefore conclude that working household heads of nuclear-type households do earn higher incomes than. Consequently, the income differential between working household heads of core and extended type households becomes fractionally larger between 1995 and 2006. In general, both core and extended type households are worse off in terms of their household heads' income between 1995 and 2006.
Second, household heads of extended type households are more likely to be African, Colored or Indian than White compared to households of core type households. In terms of geographic location, household heads of extended type households are more likely to live in non-metropolitan areas than household heads of core type households (although this result was only significant at the 88% confidence level). Furthermore, in terms of provincial location, household heads of extended type households are more likely to live in all seven provinces (Western Cape, Eastern Cape, Northern Cape, Free State, KwaZulu-Natal, North West, Mpumalanga and Northern Province) in comparison with Gauteng (although Western Cape was statistically insignificant at the 95% confidence level).
Moreover, it appears that heads of households from extended households are more likely to have demographic characteristics that would reduce their access to financial resources. The first multinomial probit is estimated using a restricted sample of nuclear-type household heads (only single persons, couples, nuclear families and single parents) and the second multinomial probit is estimated using a restricted sample of extended-type household heads (three generations, skip over generation, multigenerational and complex but related heads of household).
A literature review of measuring household well-being
Poverty estimates can be generated using two common measures of poverty – namely, the poverty headcount ratio and the poverty gap ratio. 65 | P a g e A second measure of poverty that can be used alongside the poverty headcount ratio to calculate the depth of poverty is the poverty gap ratio. For the current analysis, the poverty headcount ratio and the poverty gap are given together to provide a more accurate picture of poverty.
In this regard, estimating per capita poverty simply divides total household income by the total number of household members. Therefore, when estimating poverty per capita, an economies of scale parameter of θ=1.0 and a weighting of child costs of α=1.0 are implicitly chosen. This highlights the fact that making implicit assumptions about economies of scale (namely that θ = 1.0 in per capita poverty estimates) has a substantial impact on poverty estimates.
Holding per capita expenditure constant, the budget share devoted to food falls with household size, meaning that larger households experience negative economies of scale. Similar to economies of scale, the costs of children need to be accounted for in order to make more accurate estimates of poverty. Engel's proposition (as discussed in section 4.2.1. 'economies of scale' above) was that the proportion of the budget devoted to food accurately identifies a household's well-being.
Woolard and Leibbrandt (2001) conduct a notable study of the impact of equivalence scales on estimates of poverty in South Africa. The poverty employment ratio and the poverty gap ratio are further chosen as two appropriate measures of poverty. In this regard, estimates of poverty per per capita distort the poverty profile for certain groups given the implicit assumption that there are no economies of scale in consumption for larger households and equal costs for children compared to adults.
Household composition and household well-being in
However, the poverty gap ratio for the R447 poverty line shows that nuclear-type households are further from the poverty line than extended-type households (the depth of poverty). However, the poverty gap ratio at the lower R242 poverty line increases for nuclear-type households, while it decreases for extended-type households compared to the R447 poverty line. Overall, these results suggest that extended-type households are more likely to fall below the poverty line than nuclear-type households.
Nuclear-type households experience little to no change in the share of households below the poverty line as the value of θ decreases (and thus the effect of economies of scale increases). However, extended-type households experience a consistent decline in their poverty rate as the value of θ is the same. All extended-type households experience a decrease in their workforce when the value of θ is decreased.
Specifically, extended-type households become relatively less poor compared to nuclear-type households when the weighting of economies of scale increases. In this regard, the poverty rate for nuclear and extended households decreases by 6.02% and 6.35%, respectively. 88 | P a g e Figure 6.7: Epanečnikov's core estimate - Sensitivity of household equivalent income per adult to child costs (α) for extended-type households.
In this regard, extended-type households have consistently higher workforces than nuclear-type households. Nevertheless, the poverty rankings of broad household types are robust to changes in the equivalence scale—namely, extended-type households have consistently higher poverty rates than nuclear-type households. 12 This is with the exception of equivalence scale 3, where both nuclear and extended types of households experience an increase in their workforce.
In 1995, nuclear and extended households appear to have relatively similar depths or intensity of poverty. Over time (between 1995 and 2006), the share of households below the poverty line increased for extended-type households, while it decreased for nuclear-type households.
Conclusion
Therefore, the composition of households in South Africa appears to be undergoing a unique change in the post-apartheid period. However, a deeper understanding of the structure of households in South Africa requires an analysis of the demographic characteristics of households of different composition. A common thread when analyzing household composition in South Africa is the heterogeneity of household types both between broad household types and within broad household types.
Overall, the composition of households in South Africa is seen to undergo significant changes over the period. A gender analysis of the burden of care for family and voluntary caregivers in Uganda and South Africa. The structure of urban black households: new research evidence from a colored and African community on the Cape Flats of the Western Cape of South Africa African Sociological Review, 12:22-40.
In Amoateng, A en Heaton, T (red.) Families en huishoudens in post-apartheid Zuid-Afrika: sociaal-demografische perspectieven. Apartheid Changes in South African Poverty and Inequality to key Data Imputations, Centre for Social Science Research, Working Paper 106. Education and Employment in Post-Apartheid South Africa: A case study in the Western Cape University Leipzig, Small Enterprise Promotion and Training Working Paper 2000:4.
In Amoateng, A and Heaton, T (eds.) Families and Households in Post-Apartheid South Africa: Socio-demographic perspectives. Intra-household resource allocation and their impact on expenditure patterns: comparative evidence from South Africa and Pakistan. The impact of multiple imputation of raw data on estimates of the. working poor in South Africa.
