Air pollution is one of many issues that have a direct impact upon the economy and the well-being of society in South Africa. Domestic coal combustion contributes significantly to the air pollution problem in the country. Both qualitative and quantitative methods of data collection have been employed in this study. A questionnaire survey was conducted in 100 households in Doornkop (Soweto) and 100 households in KwaGuqa (Witbank). The observations were carried out simultaneously while the questionnaire surveys were being administered in both study areas. Interviews were also held with Eskom (the Electricity Supply Commission of South Africa) officials. Finally, the indoor concentrations and elemental composition of respirable particulate matter (PM7) were measured in three different types of households: electrified without coal burning, electrified with coal burning, and un-electrified with coal burning. The results show that township households, whether electrified or not, continue to burn coal. In both study areas, 80 % of electrified households burn coal for space heating and cooking. Although the major obstacles preventing people from discontinuing domestic coal combustion are poverty and the ready availability and social acceptability of coal, the social value of a fire inside township households cannot be underrated. Previously developed coal-supply networks still exist in the townships and makes coal utilisation very convenient. The findings also point to use of multiple fuels in the communities studied. The key fuels used for domestic energy supply are coal, electricity and paraffin. Emergent patterns of domestic coal combustion, driven in part by various societal dimensions, are also observed. Further, despite the previously observed increase in respiratory ailments in winter, township residents do not think that such increases are linked to domestic coal combustion. The study, as shown here, is in line with theories of the energy ladder which posits that as people’s financial situations improve, their energy-use patterns change. Indoor aerosol concentrations followed the same trends in all selected households with morning and evening peaks. These peaks are directly related to the making of coal fires. The highest aerosol levels, reaching a maximum of 2344.89 μg.m-3, are recorded in the un-electrified coal-burning household. Aerosol concentrations are slightly lower, averaging 1854.07 μg.m-3, in the electrified coal-burning household, implying a slight decrease in the amount of coal burnt. The lowest aerosol concentrations, averaging 478.74 μg.m-3, are recorded in the electrified household with no coal-burning. Elemental analysis reveals that the biggest contributor to respirable particles in KwaGuqa is soil dust followed by coal smoke, and then emissions from neighbouring steel smelters, whilst traffic emissions are the lowest contributor. Most importantly, it is people’s activities that determine the type and levels of respirable aerosols that they are exposed to as compared to the fuel-use patterns and types of fuels used in their household. In conclusion, electrification might phase out domestic coal combustion in the long term but only if the economic status of coal users improves. Alternatively, there is an opportunity to reduce emissions by introducing a low-smoke solid fuel, however, households will only use it if it is priced competitively and its heating and ignition properties are similar to, or better than, those of coal.