|
The aim of the survey questions, which form part of SASAS, was to assess the attitudes of South Africans towards the language of instruction in various levels of the education system. More specifically the study sought to find: Statistically significant differences among key demographic characteristics, namely race, personal monthly income, and highest educational level completed, and the preferred language of instruction in the education system; Trends in preferred language of instruction over four years, from 2003 to 2006.
Collecting the data The study was conducted from 2003 to 2006 among people 16 and older, using a face-to-face administered questionnaire. This includes people living in households, informal dwellings and hostels, but excludes those living in special institutions, hospitals and prisons. The HSRC master sample was used to draw a systematic random sample from ten households in 500 enumeration areas. The sample of respondents that completed the questionnaire was representative of geo-type (urban versus rural), gender, and race. The sample sizes for data used in the analysis ranged from 2 497 in 2003 to 2 904 in 2006. The average proportion of women respondents was higher (59.9%) than that of men, while that of black Africans was 62.8%, coloureds 15.7%, Indians/Asians 9.9% and whites 11.6%. Information was gathered from respondents over the four-year period on the following question: What do you think should be the main language of instruction in grade 1–3, grade 4–9, grade 10–12 and higher education (university, college, technikon)? The following choices were provided: English Home language of the learner Afrikaans Do not know.
Analysing the data A baseline analysis conducted on data gathered in 2003 showed that a majority of respondents favoured English as a language of instruction, even at grade 1–3, except among Afrikaans-speakers. It was found that race, personal monthly income and education level were the main predictors of choice. The results of the study show a complex and cross-cutting pattern of class, ethnicity, education and preferred language of instruction. Regarding race groups, Indians/Asians were more likely than whites to select English as the preferred language of instruction in the formative phase of schooling, followed by Africans. Those with no income were more likely than better-off individuals to select English as the preferred language of instruction in the formative phase of schooling.  This trend decreases with increases in income. Those with lower levels of education were less likely than those with tertiary education to select English. Spotting trends Preliminary analysis showed that, over the four year period, major changes in terms of preferred language of instruction occurred at the foundation phase (grade 1–3). Preference for Afrikaans at the various levels of education was relatively unchanged, therefore we decided to merge Afrikaans with home language and focused on the foundation phase, where trends over the four years were identifiable. The proportions of South Africans who preferred English as the language of instruction in the various levels of the education system are illustrated in Figure 1. It is evident that English remains the dominant preferred language for instruction in the tertiary levels, followed by grade 10–12 and grade 4–9. Despite a decrease in 2005, the preference for English as the language of instruction from grade 4 upward continued to increase over the years. The demand for English in the foundation phase (grade 1–3) has progressively decreased in preference for the home language of the learner. A trend analysis was used to investigate long-term social change. In trend studies, the data of the successive waves refer to different persons who all belong to the same population group. Analysis of the data over the four years showed a statistically significant linear association between time (in years) and preferred language of instruction. The preference for English in grade 1–3 clearly decreased over the four years of the survey while the preference for a home language increased. However, there were no statistically significant changes in the preference for English or home language in grade 10–12 and tertiary education. English remains the preferred language of instruction at these levels. We examined the trends on preferred language of instruction at the foundation phase (grade 1–3) pertaining to the categories of race, highest level of education completed and personal monthly income. Among black Africans there was a linear association between time (in years) and preferred language of instruction, with a decrease in choosing English as the preferred language of instruction over the years. 
Among coloured respondents, there was a curvilinear relationship between time and choice of language. The results showed that there was an increase in the preference for English from 2003 to 2004, followed by a decrease in 2005/06. Among Indians/Asians, English remains the preferred language. And among the white respondents there was a curvilinear relationship between time in years and preferred language of instruction, with a decrease in the preference for English from 2003 to 2005, followed by an increase in 2006. Comparison by highest level of education did not yield significant results except for those with primary education, where the preference for English at the foundation phase decreased. No significant results were found with regard to personal monthly income except for the R1 750.00 income group, where there was a decrease in the preference for English in grade 1–3. Upwardly mobile in English To conclude, English is the language of perceived potential for upward educational mobility among black Africans and Indians and remains the preferred language of instruction in South African schools, except for the foundation phase (grade 1–3). However, Afrikaans maintains some strength among better-off white and coloured respondents. Mixed results are found in the foundation phase where there is a realisation that home language is important during the formative years of schooling. While a linear decrease in commitment to English in grade 1–3 is evident among black Africans, the curvilinear relationships among the coloured and white respondents require further investigation. Special events in the education sector during the period of the study could shade light on some of the trends. Dr Mbithi wa Kivilu is the director and head of the Socio-Economic Surveys at the HSRC, and Mandla Diko is a statistician in the same unit.
|
