Segmenting food consumption in the Free State province of South Africa

In this paper the nature of diversity that exists with regard to food consumption in the Free State Province is investigated. Furthermore, this paper also investigates the factors (quality attributes and household characteristics) differentiating these market segments from one another. Four separate steps were followed in an integrated manner in the investigation. Firstly, a representative sample was drawn for the Free State Province food consumers. Households are then segmented into smaller submarkets. The second step involved the use of principal component analysis to correct for the strong and prevalent underlying structure with respect to household preferences. In the final step, a binary logit model was applied to analyse the determinants of the market segments' food consumption. Five market segments are prevalent in the Free State Province, namely: Cluster 1 (34.7%) - inferior-product consumers; Cluster 2 (23.6%) - at-home (basics) consumers; Cluster 3 (26.6%) - balanced consumers; Cluster 4 (12.3%) - value-added-oriented consumers, and Cluster 5 (2.6%) - high-frequency consumers (broad product range). The final logit model that was applied consisted broadly of economic and non-economic variables. The results show that significant differences prevail amongst the five market segments, for both economic and non-economic variables included in the model. Producers, marketers and other participants in the food value chain can use this as a basis from which appropriate marketing strategies could be developed, formalised and implemented.