| Summary: | Honey is considered the only food of animal origin that can be consumed without being processed (Pocol and Teselios, 2012). The literature presents several reasons associated with the consumption of honey, namely, it is a natural and healthy product; its dietary, nutritional and medicinal characteristics; the product quality; the geographical region of production; the information available on the products’ label, the brand's reputation; as well as, the variety, texture, taste, aroma, appearance, packaging and price (Yeow et al., 2013, Ismaiel et al., 2014; Wu et al., 2015, Ribeiro and Fernandes, 2017). The present research intends to identify the intrinsic and extrinsic attributes of honey on which consumers base their purchasing decision. Thus, a cross-sectional study was developed based on a non-probabilistic sample composed by 474 individuals, from which 399 were honey consumers. Data collection took place from March to May 2016 and was based on a questionnaire developed by Ribeiro et al. (2009), which was applied directly to consumers in the city of Bragança. Later, data was treated using the SPSS 23.0 software. The data analysis includes univariate and multivariate analysis. The first consists on a univariate descriptive analysis, namely, the calculation of relative and absolute frequencies in qualitative variables and the calculation of measures of central tendency and dispersion in quantitative variables; and the multivariate analysis includes the estimation of a binary logistic regression in order to identify the product attributes determinants of honey consumption. In the logistic regression model, the stepwise method was used to choose the factors. The overall validity of the model was tested using likelihood ratio (LR), as well as, the significance of each estimated parameter, with hypotheses H0: βj = 0 and H1: βj ≠ 0, at a significance level of 1%. Plus, the adjustment quality of the model was tested using the Nagelkerke R2, a coefficient that reveals the proportion of variation explained in the model of logistic regression. The majority of the surveyed honey consumers were aged between 25 and 64 years old (61.2%), were female (58.1%), employee (45.6%), had secondary school level (36, 6%) and higher education (40.6%). Additionally, they lived in households of 3 (21.6%) and 4 people (36.3%), with a monthly income of up to 999 euros (50.5%) and in the urban area (56.1%). The honey attributes most valued by the respondents, in order of importance, were: taste (71.4%); aroma and crystalline appearance (57.1%); colour (51.5%); viscosity (47.2%); geographic origin (42.3%); certification label (42.4%) and price (41.2%). The output of the logistic regression model estimation is shown in Table 1. The estimated model is statistically significant (significance = 0.000). The results show that the statistically significant parameters are taste, colour, origin and certification label, at a level of significance of 1%. These characteristics accounted for 68.9% of the consumer's decision to buy honey. It is noteworthy that the certification label is considered important by non-honey consumers, probably when they buy the product to offer. While taste, colour and origin are important characteristics that honey consumers value in the purchasing decision process of this product.
|
|---|