||This study commences with an overview of the concept “quality of life” as perceived from a developmental point of view. The study focuses on the current measures of the improvements in quality of life which operate at different measurement levels. Most of the measures are economic in nature like household income and income per capita, gross domestic product (GDP) and Gross national product (GNP) (Todaro, 1997). Other quantitative measures considered by the current study include measures of wealth particularly the Living standards measurement (LSM) by the South African advertising and research foundation (SAARF), Consumer confidence index (CCI), Index of economic well-being and the Human Development index (HDI) among others (Hagerty et al., 2001). A household-based measure using nominal level data, the LSM in particular tracks improvements in household wealth (as opposed to household income) through changes in household possession of durable items. Subjective measures of quality of life and changes in life satisfaction are looked at by the current study, including studies by Erikson (1993), Moller (1987, 1996, 1997) and, the wellbeing measures by Diener and Suh (1997) amongst others. Quality of life however, is not just about money as economics might have it portrayed. It is not just about how individuals feel because, according to Diener and Suh (1997), feelings are in most cases a response to external influences. Quality of life is a complex and multidimensional phenomenon which needs to be viewed holistically. As a result this study embarked on developing a measure of quality of life (a quality of life index) using household data pertaining to socio-economic aspects. The level of measurement for the data is ordinal. Operatinalised at household level, the measure was intended to analyse changes in household quality of life (QOL) between 1996 and 1999. Data for October household surveys for the period 1999-1996 was used in the study. The analysis focused on changes in household access to selected indicators of quality of life. The study applied cluster analysis to group households accessing similar QOL indicators into QOL groups. Identifying the indicator or indicators which differentiate the QOL conditions among QOL groups was achieved through the use of discriminant function analysis. The entire array of QOL groups or clusters from a particular set of data (OHS 1996-OHS1999) constituted the QOL index. The main findings of this study are that broadly, there has been an improvement in household quality of life (QOL), basing on the developed measure of quality of life. This is revealed by an increase in the number of clusters of households or QOL groups from five in 1996 to eight in 1999. The study attributes the increase in QOL groups to an increase in households’ ability to access the selected QOL indicators. In spite of the increase in the number of QOL clusters, the study finds that proportionally fewer households are found in the QOL groups with better material living conditions (i.e. measurable QOL) than otherwise. This is contrary to the expected pattern in development terms based on empirical evidence in South Africa (see SAARF, 2002; SAARF, 2004; Stats SA, 1996; Stats SA, 2001; Stats SA, 2004). The study also finds that female headed households are generally predominant in groups with poor QOL. Discriminant function analysis results highlight access to toilet, refuse disposal services and water source as discriminant indicators in addition to Highest level of education completed by a household head and, Employment status of household head, among others. The latter consistently differentiate between groups of households throughout the reference period except in 1999. Findings relating to the influence of household material conditions on perceived quality of life show that proportionately more households in groups with the better access to the selected QOL indicators being satisfied with life than otherwise. A point worthy noting is the consistency in the proportions of households which felt that things had not changed after all, irrespective of the groups’ ranks, throughout the reference period. The key conclusion drawn from these findings is that low levels of education and employment status among household heads strongly influence household quality of life. These two indicators have been found to consistently differentiate the QOL conditions among the QOL groups that emerged. Groups on the poor side of the QOL index are characterized by high unemployment, illiteracy and dysfunctional levels of education for most household heads therein. Most households belonging to the poorest QOL groups are rural-based (found in Eastern Cape, Limpopo, Kwazulu Natal and Mpumalanga), with poor access to basic services identified under discriminant function analysis. The situation is likely to be complicated by the existence of substantial proportions of households headed by people aged 15-19 identified in this study. This needs to be taken seriously particularly in the current era of the HIV/AIDS pandemic (see HSRC, 2002; Rosa, 2003). The study’s findings have revealed that poor QOL among households is not related to the sex of the household head. Although female headed households are predominant in groups of households with poor QOL conditions, adjacent to such groups are households in groups with almost equally poor living conditions the majority of which are males-headed. What is needed therefore is a holistic focus on the factors that impede households’ ability to sustain better living conditions. Most of the study’s recommendations reinforce initiatives which are being undertaken in the development agenda. For instance the need to improve people’s level of education does not need any more emphasis given the study’s results. Sustaining improved household QOL will require households to have a capability of meeting their needs. Successful completion of education – tertiary as opposed to functional literacy- opens channels for households to lead a better life. Achieving this level of education requires time, which from a demographic point of view, most of the currently uneducated household heads may not have. While much has been done in enabling households to access basic services like housing, electricity and water, payment for such services remains the responsibility of individual households. Inability to pay for services – due to unemployment and lack of education-will just perpetuate household dependency on social grants. It is also recommended that in-depth qualitative studies be undertaken to establish the apparent consistent gap between objective living conditions and subjective life satisfaction among households if realistic policy objectives are to be achieved. The study recommends a further application of the formulated QOL index particularly on current data with similar indicators. A more rigorous thinking around the weighting of individual QOL indicators will iron out the inconsistencies observed in the study’s results. This will provide an opportunity to standardise the indicators, update the results of the QOL index while enhancing triangulation at the same time.
||Kironji, E 2008, Measuring quality of life in South Africa : a household-based development index approach, DPhil Thesis, University of Pretoria, Pretoria, viewed yymmdd <http://hdl.handle.net/2263/25060>