Data for the TSS was collected from six South African townships. The townships chosen were Eldorado Park, Ivory Park, Khayelitsha, Mitchell’s Plain, Thabong and Umlazi. 2453 interviews of smokers took place over two rounds between October and November 2017 and again between July and August 2018. The sample includes two of South Africa’s four population groups: African and mixed race (locally referred to as “Coloured”). This is because very few Whites and Asians live in townships. Households were selected at random by enumerators, and one smoker per household was interviewed (if a household contained at least one available smoker).
Kind of Data
Sample survey data
Unit of Analysis
Households and individuals
v1.0: Edited, anonymised dataset for public distribution
The TSS interviewed 2453 smokers. Respondents were asked about cigarette and tobacco product consumption, perceptions of illicit cigarette trade, and select other socio-economic questions.
Data was collected in Gauteng (Eldorado Park and Ivory Park), Western Cape (Khayelitsha and Mitchell’s Plain), Free State (Thabong) and KwaZulu-Natal (Umlazi).
Producers and sponsors
Research Unit on the Economics of Excisable Products
University of Cape Town
International Development Research Centre
Cancer Research UK
Bill & Melinda Gates Foundation (via African Capacity Building Foundation)
Enumerators walked through townships and selected houses/dwelligns based on a randomized process. The enumerator was required to enter into a digital form how many directions it was possible to take at a street intersection so that the system could randomly select the direction to take. It also selected the side of the road to walk along and the house from which to start interviews. This procedure was repeated at every intersection and at every apartment building. One adult smoker per household was interviewed (if a household contained at least one available smoker). If there was more than one available smoker, a randomization procedure was again used to select a smoker from the household.
Dates of Data Collection
Data Collection Mode
There were two questionnaires used, one for each round of data collection. The questionnaire was improved between the two rounds. Some variables were dropped, some were added and some were modified. If the question differed in any regard over the two rounds this is noted in the microdata.
Data was captured using electronic tablets, with a questionnaire provided by UCT operationalised on the SurveyCTO platform.
Reports from the data collection teams often noted dangerous these townships are. Enumerators could choose the areas they covered. So it is likely that the worst areas were avoided, which could effect measurement. For example, the prevalence of illegal cigarettes.
There is possible re-interviewing of the sample people over the two rounds, although the probability is quite low due to the size of the areas in question.
University of Cape Town
Public use files, available to all
Research Unit on the Economics of Excisable Products. Township Smoking Study 2017-2018. [dataset]. Version 1. Cape Town: Research Unit on the Economics of Excisable Products (REEP) [producer], 2020. Cape Town: DataFirst [distributor], 2020. DOI: https://doi.org/10.25828/47cd-z315