Marginalized Knowledge: An Agenda for Indigenous Knowledge Development and Integration with Other Forms of Knowledge

2. Research methodology
Two research methods were deployed in order to gather data from the respondents. First, the case study method was used to collect data from key informants in four organizations that work with ICTs and rural women. Using frame lists from both South Africa (Sangonet, 2005:1) and Kenya (EUFORIC, 2006), a total of four organizations were purposefully selected to include Womensnet and National Community Radio Forum (NCRF), and AfriAfya and the Arid Lands information Network (ALIN-EA) respectively from South Africa and Kenya In the second method, survey research was used, and women aged between 16 and 60 were targeted for interviews. Data was collected from a cross section of female inhabitants in the region of the study in-order to determine the relationship between the type of ICT resources accessed and used. The survey research design was found suitable, given the large number of respondents. However, the research design took into account the purpose of the research, the accuracy of the results, and the cost, time and labour involved.

In Kenya, survey data was obtained from the sub-divisions of the Kaplamai Division in Trans-Nzoia district, i.e. Kimoson, Sinyerere, Sitatunga and Makutano. The sampling frame included: small-scale traders (68; 34%); housewives (29; 14.5%); educators/teachers (27; 13.5%); farmers (26; 13.0%); students (11; 5.5%); domestic workers (10; 5.0%); preachers (10; 5.0%); farm workers (6; 3.0%); large-scale entrepreneurs (5; 2.5%); nurses (4; 2.0%); clerical workers (2; 1.0%); and community development workers (2; 1.0%).

In South Africa, a similar sampling frame included: small-scale traders (58; 29.0%); housewives/homemakers (48; 24.0%); farm employees (25; 12.5%); domestic workers (18; 9.0%); educators/teachers (16; 8.0%); students (15; 7.5%), entrepreneurs managing large-scale enterprises (3; 1.5%); clerical workers (9; 4.5%); community development workers (6; 3.0%) and two preachers (2; 1.0%). Sampling data was obtained from census household data from the magisterial districts of Umlalazi, i.e. Eshowe, Amatikulu, Gigindlovu and Mtunzini.
A sample size of 400 was deemed suitable, given that the study population in both countries was beyond 5000. This selection was informed by Gay’s guidelines (1996:125), which state that: (i) the larger the population size, the smaller the percentage of the population needed to get a representative sample; (ii) for smaller populations, i.e. where N<100, there is little point in sampling; (iii) if the population size is around 1500, 20% should be sampled; and (iv) beyond 5000, the population size is irrelevant and a sample size of 400 is adequate.
By using the snowball technique, women respondents who were directly and indirectly linked were identified and consequently interviewed.
3. Results

This section responds to the research objectives outlined in section 1. Thus section 3.1, responds to the question -What problems do rural women experience when accessing and using ICTs? section 3.2 responds to the second research question, “What are the views/comments on the use and availability of ICTs in the rural areas of KZN and Trans-Nzoia district?, section 3.3 responds to the question, Which ICTs serve rural women’s information needs in rural KZN (South Africa) and Trans-Nzoia (Kenya)? And lastly, section 3.4 responds to the question “how do ICTs enhance rural women’s social welfare and quality of life in the rural areas of KZN (South Africa) and Trans-Nzoia District (Kenya)?