The data used in this research came from various sources, including search engines such as PubMed or EMBASE, using a range of relevant keywords, such as health, HIV, AIDS and policy. Books, journals, and reports regarding the MDGs, their implementation, and of course their results were searched and used.
The inclusion criteria for secondary resources selected included: 1) studies on HIV/AIDS prevalence within the region of interest, particularly South Africa; 2) studies addressing policy recommendations/implementation with regards to HIV/AIDS; 3) peer-reviewed articles as well as conference abstracts that underwent peer-reviewed blinded selection processes; 4) studies not published in peer-reviewed journals but commissioned by governmental agencies; 5) full articles on health public policymaking; and 6) paper informing the research on methodology and health policy focus.
Moreover, the documents studied and analysed here also came from the organisations that are concerned with the MDG, including the United Nations and its specialised agencies such as WHO, UNAIDS and the World Bank, and other international groups whose aims are towards achieving these health-related goals. Furthermore, official documents and reports from the South African government were used. The data collection method is document study. A document refers to all written and/or recorded materials (Lincoln & Guba, 1985) that were prepared for an original purpose or based on the request of the researcher.
Documents are either public records or personal documents (Lincoln & Guba, 1985). Public records refer to the materials made and recorded as accounts or attestation to a particular phenomenon. Public records could be external when sourced from outside of the venue of the evaluation such as government agencies collecting census data and internal when sourced within the venue of the evaluation such as reports by government agencies on department affairs.
Both external and internal sources provide important sources of secondary data. Using both could support the quality of research since external sources are susceptible to recall bias that could arise during the time in between data collection in the field and organisation or analysis in the office. (Lincoln & Guba, 1985) Personal documents refer to accounts of experiences of events in the first person such as journals and works on personal accounts.
Personal documents help provide an understanding of the perspectives of a key person involved a particular phenomenon, which could be valuable in studies requiring a first person perspective. (Lincoln & Guba, 1985) Although both public records and personal documents are accessible by not requiring permission or requiring minimal permission from the source (Fetterman, 1989), public documents are more useful for the present study on the HIV/AIDS policies made by South Africa in the context of the millennium development goals.
The South African government is an artificial entity and its perspectives are captured in public records of its official stand on HIV/AIDS and the direction of policymaking on this health issue. Although there could be personal account by the current and previous presidents and other government officials, these are personal perspectives and may not sufficiently represent the perspective of the South African government as a single political entity.
Search for sources of secondary data would focus on public documents. Conducting a document study has advantages and disadvantages as summarised in the table below. Justifying the use of document study as the approach to data collection would mean addressing the disadvantages by identifying the pertinent data, allocating sufficient time for document search, and thoroughly considering the sources found to come up with data relevant to the research.