|The aim of the research project is firstly to determine the variables that have an influence on academic achievement and
secondly to determine the specific influence of intelligence on academic achievement.
To achieve this aim a literature study was undertaken, followed by an empirical investigation. It emerged from the first literature chapter (see chapter 2) that intelligence is such a complex construct that it can be studied from different angles. Because the psychometric, developmental and information processing theories of
intelligence proved to be the most influential ones, these were specifically studied. Despite the differences between these approaches to intelligence they also proved to have much in common. They agree for example that reasoning is a fundamental aspect of
intelligence and it seems possible to identify a general intelligence factor within any of the three approaches (see chapter 2). In the second literature study chapter (see chapter 3) it was clear that academic achievement is influenced by variables that can be categorised as cognitive, non-cognitive and milieu related. The cognitive variables proved to be the most influential, with
intelligence (a cognitive variable) as the single best traditional predictor of academic achievement. The empirical investigation made use of data gathered in five primary schools in Rustenburg during 1988 (see chapter 4). All the standard five pupils who received education through the medium of
Afrikaans formed the population of the research program. Different measuring instruments of which the "Nuwe Suid-Afrikaanse groeptoets" (NSAGT) as the single most important measuring instrument were used on a sample (see chapter four). A variety of independent variables that influence academic achievement were identified. The SAS-computer and BMDP-computer programs were used to process the information. The different statistical techniques are described in paragraph 4.8. A factor analysis was carried out to group the
different control variables according to their correlation coefficients with the standard five average results as well as results in Afrikaans and Mathematics respectively (see table 5.1).
The independent variables were grouped into six different factors. A predictor variable was identified within every factor and grouped with intelligence (see table 5.2) to determine the contribution to R2 in each of the dependent variables, namely average
percentages, Afrikaans and Mathematics. Because sex variables proved to be a significant statistical predictor variable in both Afrikaans and average percentage, the sample was divided into boys and girls and the multiple regression analysis was rerun (see tables 5.3 and 5.4). Intelligence was finally divided into verbal and non-verbal intelligence and the multiple regression program was run again on the sample boys and girls (see tables 5.5, 5.6 and 5.7). The results of the empirical research can be summarised as follows: (1) Academic achievement is influenced by a large number of different variables of which aptitude and phlegmatic proved to be the most influential ones. (2) The influence of intelligence on academic achievement proved to be of no significance when the sexes are studied together. When the sexes are separated intelligence proved to be an academic
predictor on average percentage and Afrikaans for boys but not for girls.