This content requires Adobe Flash Player version
or later.
Either you do not have Adobe Flash Player installed,
or your version is too old,
or there is a problem with your Flash installation and we were unable to detect it.
The race profile of the respondents in this study is in keeping with the overall race demographics of South Africa. Ensuring an equitable representation of all race groups once again certified that the effects of culture in brand loyalty were considered in keeping with the research results by Ha, Janda and Park (2009: 211) and Lam (2007: 16) that confirm that culture and sub culture have a definite bearing on brand choice.
Quantitative analysis
Each of the 12 influences was analysed by means of factor analyses to determine if the three different consumer products (toothpaste, coffee and bread) do yield a similar factor or factors. This comparison was conducted to determine whether all three products could be treated and analysed as a single entity. Comparisons per response are represented in Tables 2 to 13. A three-point test was conducted to determine the similarities between influences. The three-point test examined for similarities in the statistics loaded on the component matrix, the variance explained and the correlation coefficients.
Each table provides a component matrix representing values for each product in each question and explains the percentage variance for each product factor per product category. Each product is then compared with the other by means of Pearson correlation coefficient to determine the similarity between them.
The Kaiser-Meyer-Olkin (KMO) measure is utilized to examine the appropriateness of factor analysis in this study. Judgements are based on the recommendations by Hutcheson and Sofroniou (2009: 233) who indicate that a value close to 1 indicates that patterns of correlations are relatively compact and so factor analysis should yield distinct and reliable factors. Values between 0.5 and 0.7 are mediocre and values between 0.7 and 0.8 are good, values between 0.8 and 0.9 are great and values above 0.9 are superb (Field, 2007: 66). The Bartlett test of sphericity is used to determine if each set correlates perfectly with itself (r = 1) or has no correlation with the other variables (r = 0). Finally, Cronbach alpha is used to test the reliability of data for subsequent analytical scrutiny where the recommendations of Field (2007: 668) is used in that an alpha coefficient of 0.7 is sufficient subsequent analytical scrutiny even though Moss et al. (1998: 180) suggest that an alpha score of 0.6 is generally acceptable.
Tables 2 to 13 reveal the component matrix for each FMCG product based on the factors from the questionnaire. The correlation coefficient is also indicated which reveals the nature of similarity between the products. Finally, the tables provide results for adequacy and reliability.
A COMPARATIVE ANALYSIS OF BRAND LOYALTY OF COMMON HOUSEHOLD PRODUCTS 398