|Country Report South Africa - ICRC Worldwide Consultation on the Rules of War (International Committee of the Red Cross , 1999, 70 p.)|
The findings in this report are based on a consultation carried out by the ICRC in South Africa under the supervision of Greenberg Research and with the help of a local opinion research partner, Markinor, based in Randburg. The aim was to assess the impact of armed conflict on peoples lives and to allow people to share their personal experiences and opinions on a range of issues, from the armed conflict itself and the limits of warfare, to the impact of international law, including the Geneva Conventions.
The South African consultation consists of three elements:
· Nine focus groups (FG) recruited by the ICRC, the South African Red Cross and Markinor. Moderators were provided by Markinor, working under the supervision of Greenberg Research. The groups were held between 4 and 13 March 1999 in three different regions of the country: KwaZulu Natal, the Cape Town area and Guateng (the main urban centre encompassing Johannesburg and Pretoria and the surrounding townships, including Soweto). Sessions were held with the following groups: medical personnel from Baragwanath Hospital in Soweto township; members of ANC Self Defense Units (SDU) and Inkatha Freedom Party Self Protection Units (SPU) in Thokoza township, Gauteng; widows in Bambaay township near Durban; teachers from Ndaleni township near Richmond, KwaZulu Natal; and in the Cape Town area, former members of the South African Defense Force (SADF), displaced coloured persons from District 6, former MK members detained under the apartheid regime on Robben Island, and mothers of youth living in Mannenberg township, which is known for its gang activity.
· Twenty-one in-depth-interviews (IDI) conducted mainly by ICRC and South African Red Cross staff between 23 March and 14 June 1999. Interviewees included a farmer from Pietermaritzburg, staff from local NGOs, a religious leader in Pretoria, a teacher and an elder in KwaZulu Natal.
· A national quantitative survey conducted among the South African population. A total of 1,500 respondents were surveyed, of at least 18 years of age and stratified geographically according to population. The sub-sample for KwaZulu Natal was boosted to enable an analysis of that region and was weighted down to its correct percentage in the overall sample. The survey was conducted by Markinor, under the supervision of Greenberg Research. The questionnaire was translated from English into five languages: Zulu, Xhosa, Northern Sotho, Southern Sotho and Tswana. The survey took place between 14 April and 21 May 1999. Total percentages reported here are subject to a sampling error of +/- 3.5 percentage points (at a 95 in 100 confidence level). Results in smaller segments, such as the 293 interviews in the KwaZulu Natal area, are subject to an error of +/- 7.1 percentage points. 9
9 These estimates are based on population values of 50 per cent. Obviously many reported percentages are lower or higher than that; high percentages would have a smaller sampling error. For example, a reported percentage of 90 per cent for the total population would have a sampling error of +/- 2.4 percentage points.
During the focus groups, research observers listened to the discussions from a separate area with the aid of a simultaneous translator. Discussions were held in English, Afrikaans and Zulu, often a mixture of several languages and their township versions. They were tape-recorded and then translated into English for analytical purposes.
The in-depth interviews were conducted in either English or Zulu. Each interview was tape-recorded and then transcribed and translated into English, when necessary, by the local partner.