Survey of Truth and Reconciliation in South Africa 2000-2001
Household Survey [hh]
The purpose of this study was to explore the relationship between truth acceptance and reconciliation among South Africans during and since the political transition from Apartheid to democracy. The study investigated the extent to which South Africans participated in the Truth and Reconciliation Commission and the degree to which they were "reconciled." The Truth and Reconciliation Commission (TRC) was based on the Promotion of National Unity and Reconciliation Act of 1995. The TRC investigated past gross human rights violations and granted amnesty to individuals in exchange for full and public disclosure of information related to these crimes. The hypothesis that truth acceptance leads to reconciliation was tested in this research. Data were collected through a systematic survey of South Africans. Nearly all relevant segments of the South African population were included in the sample, as well as representative subsamples of at least 250 respondents of most major racial/ethnic/linguistic groups.
Questions about the TRC investigated respondent awareness, knowledge, and approval of the activities of the TRC. Respondents were asked for their opinions on the effectiveness of the TRC in its efforts to provide a true and unbiased account of South Africa's history and in awarding compensation to those who suffered abuses under the Apartheid regime. Other questions included those eliciting respondents views on the importance of revealing the truth about the past and achieving racial reconciliation.
Kind of Data
Sample survey data
Unit of Analysis
v1: Edited, anonymised dataset for licensed distribution
The scope of the Survey of Truth and Reconciliation in South Africa, 2000-2001 includes:
Individuals' characteristics: demographic characteristics (age, marital status), education level, employment status, respondent awareness, knowledge, and approval of the activities of the TRC among other questions about the TRC.
The Survey of Truth and Reconciliation in South Africa, 2000-2001 had national coverage.
The lowest level of geographic aggregations covered by the General Household Survey 2011 is Province.
The universe under investigation included all South Africans aged 18+.
Producers and sponsors
James L. Gibson
Washington University in St. Louis. Department of Political Science
The area probability sample included a primary sample of South Africans of all races and a boost sample of white South Africans. Representative subsamples of at least 250 respondents of most major racial, ethnic, and linguistic groups were also included.
The sampling was divided into two parts – a primary sample, including South Africans of all races, and a boost sample of white South Africans. In the main sample, 3,139 interviews were completed. The Boost Sample was composed only of white South Africans, with a control for language (English versus Afrikaans). A total of 588 additional whites was interviewed.
The overall response rate for the survey was approximately 87 percent (after treating “break-offs” as unsuccessful interviews). The main reason for failing to complete the interview was inability to contact the respondent; refusal to be interviewed accounted for approximately 27 percent of the failed interviews. From the response rate alone, the representativeness of the sample seems assured. Such a high rate of response can be attributed to the general willingness of the South African population to be interviewed, the large number of call-backs we employed, and the use of an incentive for participating in the interview (the incentive was a magnetic torch (flashlight), with which the respondents were quite pleased).
Because the various racial and linguistic groups were not selected proportional to their size in the South African population, it is necessary to weight the data according to the inverse of the probability of selection. In addition, post-stratification weights have been applied to the final data in order to make the sample more representative of the South African population. After correcting for unequal probabilities of selection, the data have been post-stratified according to three variables: (1) size of place of residence; (2) the respondent’s age; and (3) the respondents race. Size of place of residence is a trichotomy: (1) metro areas; (2) other urban areas (cities, large towns, and small towns); and (3) rural areas (villages, farms, and kraals). Age categories were defined as (1) 18-24 years old; (2) 35-44 years old; (3) 45-54 years old; (4) 55-64 years old; and (5) 65 years old and older. The four major racial groups in South Africa were used. After trimming the weight, and adjusting the weights to the actual number of completed interviews (3,727), the post-stratification weight variable ranges from .29 to 2.02.
Two weight variables are included in the dataset. One weight variable (NATWT) should be used when analysis is not conducted by race, and the other (RACEWT) when conducting analyses including respondent race.
Dates of Data Collection
Data Collection Mode
Data Collection Notes
This study was conducted in collaboration with Amanda Gouws (Stellenbosch University, South Africa), Charles Villa-Vicencio (Institute for Justice and Reconciliation, Cape Town, South Africa), and Helen Macdonald (Institute for Justice and Reconciliation, Cape Town, South Africa).
The questionnaire for the Survey of Truth and Reconciliation in South Africa, 2000-2001 includes individual characteristics, respondent awareness, knowledge, and approval of the activities of the TRC, how important it was for respondents to find out the truth about the past and achieve racial reconciliation.
Licensed use files, available for non-commercial use only
Gibson, James L. Survey Of Truth And Reconciliation In South Africa, 2000-2001 [Computer file]. ICPSR version. Johannesburg, South Africa: Decision Surveys International [producer], 2004. Cape Town: DataFirst [distributor], 2013. DOI: https://doi.org/10.25828/7wrp-3281