| Daughters of Sysiphus |
|Overview of findings and recommendations|
The first important finding concerned the prevalence of different kinds of households among the 677 low-income households surveyed. Only 16 per cent of the households proved to be male-headed (headed by a male adult without a resident partner), 42 per cent were joint-headed (with a head who had a resident contributing partner) and 41 per cent proved to be female-headed (headed by a woman who had sole responsibility for major decisions concerning the household).
The distribution of different types of households, however, was by no means uniform. High rates of female headship occurred in areas normally identified as being especially deprived, particularly within the inner city, and female-headship rates were lower in the more affluent of the 42 low-income areas surveyed, suggesting a close relationship between form of headship and poverty.
The fact that the age of the head of household did not vary significantly between different kinds of household suggests that female-headed households may not constitute a particular stage in low-income household formation that will eventually lead to the "normality" of the joint-headed household. On the contrary, it appears that the female-headed household may well constitute a long-term and permanent household type.
1. Further work is required on the methodology of household classification. The traditional notions of male- and female-headed households are clearly inadequate. Yet the concept of female-headship has its own methodological problems and should not be assumed to offer an easy or totally satisfactory alternative. The notion of headship itself is controversial with definitions varying from those based on purely economic criteria to those that focus on issues of household power.
Statisticians and planners within the shelter sector should become sensitive to the issues involved in household classification and explore the manner in which their own data collection and analytical systems could be adjusted to resect more adequately the realities of household dynamics.
2. National household data should be collected and disaggregated in a manner that allows gender-related variations to be traced more clearly. Until a more adequate system is developed, it is suggested that disaggregation of data on the basis of female headship, male headship and joint headship, as defined in this study, be carried out.
3. The relationship between female headship and poverty is being increasingly demonstrated in the literature. (See, for example, the proceedings of the seminars hosted by the Population Council of New York and the International Centre for Research on Women). This relationship requires particular attention from policy-makers and practitioners within the shelter sector in order to ensure that solutions targeted at low-income households do, in fact, as well as theory, benefit the people they were intended for.