Abstract

BISSCHOFF, Christo. A generalised model to measure brand loyalty. Tydskr. geesteswet. [online]. 2020, vol.60, n.4-2, pp.1280-1294. ISSN 2224-7912.  http://dx.doi.org/10.17159/2224-7912/2020/v60n4-2a10.

The value of brand loyalty not only resides in the rebuy intentions of customers, but is also prevalent in the higher prices loyal customers pay; the brand is therefore less price sensitive. Price is no longer the dominant consumer decision-making criterion in South Africa and brands and brand loyalty now strongly influence the decision-making process. However, how to manage brand loyalty is more challenging, specifically because few organisations are knowledgeable about just how loyal their customers are towards their brands. In addition, specifically which of the antecedents contribute towards brand loyalty are seldom known because this is very seldom measured and qualified. This article analyses a variety of brand loyalty industries with the aim to determine if there are some of the brand loyalty antecedents that can be generalised across industries to measure brand loyalty. The study used a selection of eight brand loyalty studies across six industries, namely, soft drinks to children, agricultural buying, wholesale pharmaceuticals, chicken to consumers, pet food and fast-moving consumer goods. These studies all used the model to measure brand loyalty which was developed by Moolla in 2010 and later refined by Bisschoff and Moolla in 2015. The model finalised twelve brand loyalty antecedents which were measured in a questionnaire consisting of 50 questions. These antecedents are categorised in attitudinal antecedents (brand trust, relationship proneness, commitment, brand affect, brand relevance and culture), behavioural antecedents (repeat purchase, involvement, switching cost and brand performance) and then other antecedents (customer satisfaction and perceived value). This article aims to isolate some of these brand loyalty antecedents that can be used to measure customer behaviour across all the industries (thus generic antecedents in all the studies), and those antecedents which are only relevant to one specific industry. It does so by analysing the original data and also the results of these eight brand loyalty studies. The original eight studies all used the same questionnaire to collect data from the industry-stratified samples to measure the twelve brand loyalty antecedents of the specific industries. Respondents recorded their perceptions regarding the antecedents by answering 50 questions on a 5-point Likert-scale. A total of 2035 responses were captured and analysed. The data pertaining to each brand loyalty construct was subjected to reliability testing, and the results show that all the antecedents have satisfactory Cronbach alpha coefficients (α > .68); they are therefore deemed to be reliable antecedents to measure brand loyalty. The data was further scrutinised to ensure that no multicollinearity exists between the antecedents. This test used Slovin's tolerance level and the variance inflation factor (VIF) to analyse levels of multicollinearity; it was confidently concluded that multicollinearity did not pose any threat to the analysis. In addition, the importance of each antecedent was also determined by measuring the variance explained using exploratory factor analysis. As expected, there are some important common brand loyalty antecedents (σ > 50%) that should be included in any measurement irrespective of the industry. They are brand affect, repeat purchase, and brand trust. The other nine antecedents are all regarded as industry-specific antecedents because they fail to show importance in all of the industries involved in this study. Although the results are valuable to managers, researchers and academia aiming to measure and manage brand loyalty, an obvious drawback is that, despite the model's success in a variety of industries, most of the data originated from South African based consumers in the various industries. Further research using confirmatory factor analysis or structural equation modelling to confirm these results could yield valuable insights with regard to the generalized model. Specifically, quantifying how well the generalized antecedents fit into the model could determine the practical usefulness of the model. Country-specific influences and different consumer behavioural patterns may have played a role in the model construction, and although there is evidence to support this supposition, future users of the model should factor in the information when attempting to migrate the use of the model outside South Africa's borders.

Keywords : brand loyalty; antecedents; model; measurement; factor analysis; brand; management; trust; buying behaviour.

        · abstract in Afrikaans     · text in Afrikaans     · Afrikaans ( pdf )