Study on the consumption of honey in Aragon


In order to know the commercialization of honey in Aragon and the main distribution channels in the honey market , as well as the preferences, attitudes and purchase motivations that Aragon consumers have towards honey, an exploratory investigation was carried out using primary sources of information based on the survey formulation and for the design of the choice experiment and the virtual supermarket. However, the collection, application and analysis of the data have been different in each case, based on the accessibility, the proximity of the sources and the type of information desired.

Considering the differences in the collection and analysis of information, this chapter describes the methodology that has been used in each of the cases.Firstly, in order to provide a general idea of the methods used, a description of the fieldwork is made, describing the procedures followed to carry out eachthe survey, the virtual supermarket explaining in each case the steps taken to select the sample, the detailed description of the questionnaires and the procedure of application and collection of the information. Subsequently, the scales of measures used in the codification process are discussed, as well as the techniques and methods of analysis carried out. Finally, the univariate, bivariate and multivariate analyses that will be used will beexplained in detail.

Field work design

Once the review of the secondary sources and the descriptive study of the marketing of honey was carried out, we proceeded to investigate the honey market and the motivations, attitudes and habits of purchase of honeythat the Spanish consumer has.Considering that the information that was intended to be used in this study was not identified in other studies, the design of the research was defined on information of primary character. To obtain this type of information, it was decided to use surveys, a choice experiment and virtual reality.

Subsequently it was resolved that the application and collection of information would take the following order: first the information would be requested to the distribution chains and finally to the consumers. The details of content and methodological procedures followed in the survey, choice experiement and virtual reality, are described in the different sections of this chapter.

Procedure ofcollection of information in the distribution channels

To design the survey, the choice experiment and the virtual reality, using the adequate attributes and prices in order to investigate the consumer behaviour towards honey, itwas necessary to have the relevant information about the product current state in

the market as the different types, prices, packaging, sale formats and origins of honey commercialised in the distribution channels.

The procedure to obtain the information was carried out in three steps: 1) An establishment of every retail channel was selected, considering the accessibility and proximity to the source of information;2) Afterwards, the establishment was approached and the objective of the investigation was explained to the person in charge in the establishment and their authorization was requested to carry out the tour; 3) Once the consent was obtained, itwasproceeded to locate the line of honey in the different areas of the establishment and to collect the data of honey mentioned previously.

After collecting the information of all the supermarkets, hypermarkets and specialized stores located in Zaragoza, it was proceeded the analisis ofthe collecteddata through the descriptive statistics.In our case, a table was designed for the collection of information on the characteristics of the honeys that are marketed in Zaragoza through the large distribution and its positioning, taking into account variables such as: Brand name; Name of the distribution chain; Brand type (Producer, distributor); Type of honey (Monofloral, Multifloral,Processed honey); Floral origin (Thym, Rosemary, Multifloral, Orange,Eucalyptus); Geographical origin (EU, Spain, Mix of original and non-original honeys from EU, Aragon); Certification (PGI, PDO); Ecological Type of packaging material (glass, plastic); Price Quantity.

Consumer survey

In order to identify if the current consumption of honey, is due to the preferences like the quality, packaging, visual appearance, organoleptic properties, price, origin , it was considered convenient to conduct a survey to consumers to explore and know some characteristics of their shopping comportment. The details of selection and application of methodology are described in the following sections.

Selection of the consumer sample

Considering that Zaragoza is the most important province of Aragon since it has the highest concentration of the population in the community, it was considered appropriate place to collect the sample.To calculate the sample, the population of Zaragoza was considered infinite since, although it is known that there are 954,811 inhabitants, the amount of the population that consumes honey is unknown.In this way, the sample to study was estimated based on:N= 4 * p * q / ɛ2 = 200N is the total sample size;p = 0.5 for a maximum sample size;q = 1 – p = 0.5Ɛ is the sampling errorIn this way, simple random sampling, with a sample of 200 individuals and an error probability of 7%, was considered convenient and representative for an exploratory research.

Description of questionnaire to consumers

Of the different ways for surveying consumers, the face-to-face surveying have been chosen, because the advantages presented , such as the high percentage of responses obtained, the possibility of explaining the questions, seeing the reaction of the respondent and catch their attention. The questionnaires begin with an introduction to inform about the scope of the research and the objectives, request collaboration, thank the possible interviewee and guarantee confidentiality and anonymity.

The questionnaire addressed to consumers consists of 20 questions (see corresponding annex) of closed and open type, using 2 Likert-type scales of 5 points for the first ones. The global set of the questionnaire has been divided into 7 blocks. These blocks have been the following:Block 1: Shopping habits(questions 1 to 5) deals with consumers’ buying habits of honey and also helps to reveal the consumption frequency of purchase (Q4), with options: Once a year,two or three times a year ,four to six times a year, once or twice a month, once a week; the place of purchasing honey (Q3) with options :Supermarket, Hypermarket, Directly from the Beekeeper, Specialty shop, Traditional products fairs.

In question (Q5) consumers where asked about their perception of some characteristics of honey during the purchase, aLikert scale has been applied with 5 answers, from Does not influence at all until it totally influences.Block 2: Consumption habits: (questions from 6 to 10): deals with the habit of consumption of honey by extracting information about the frequency of consumption of honey by the household (Q7) , the types of honey consumed (Q8),the quantity consumed during a month(Q9) and the different uses of honey by the consumers (Q10).Block 3: Motivations (questions 11 to 12), deals with the motivation of consumption of honey.

This is done with question directed on the reasons of consuming honey using the likert table (Q11) and on Consumer evaluation of some attributes to determine the quality of the honey (Q12).Block 4:Lifetstyledeals with lifestyle ofhoney consumers (Q13);a questionwith the Likert scale has been applied with 5 answers, from totally disagree to fully agree in order to evaluate some habits according the lifestyle criteria of the consumer.Block 5:Attitudes:deals with attitude criteria of the consumer towards honey from Aragon with respect to health and environmental issues (Q14).Block 6: Intention of purchase(Q15 to Q16) try to assess the consumer intention of purchase of honey from Aragon.Block 7:Contains information about the Socio Demographic Characteristics (Q17 to Q20) of the people surveyed.

Description of the analysis techniques used

The information collected through the surveys will be managed through the statistical program SPSS. The entire questionnaire questions were coded for further analysis, generating a table with all the answers obtained, resulting in 20 rows and columns corresponding to the number of variables. The data was entered in SPSS 20.0. In the first stage of the analysis, the univariant analysis was performed. Subsequently, a bivariate analysis was carried out to identify possible correlation factors among some factors.

Univariate analysis

Univariate analysis is the simplest form of quantitative analysis, which provides a short abstract of the study results. Univariate analysis was used in the first descriptive stage of this research, prior to conducting more advanced inferential bivariate or multivariate analysis. The univariate analysis can differ according to the nature of the data (nominal or metric). In this way, if variables are nominal, the frequency and the percentage were calculated and then presented in charts or tables. As for the metric variables, the mean and the standard deviation were calculated (Grande et al, 2009).

Bivariate analysis

In this case, we have applied bivariate techniques to investigate the association between pairs of some variables. This is the case, fundamentally, when it comes to identifying the possible characteristics (mainly sociodemographic) that differentiate consumer segments that differ in some specific behaviours, such as the frequency of consumption of honey or the places of purchase of honey. Specifically, those that classify the segments and those that describe the profile are the qualitative variables, and Pearson’s Chi-square test is applied.

By means of this statistic, the null hypothesis of independence of the analysed variables is contrasted, that is, if this hypothesis is rejected, it can be affirmed that there is an association between the classification variable (segmentation) and the treatment or descriptor variable (normally, social demographic). When the descriptive variable is quantitative, then the analysis of variance or ANOVA is applied. In this case, the null hypothesis is of equality of means of classification groups.

Multivariate analysis

Multivariate analysis is the third type of statistical analysis used. It refers to all statistical techniques that simultaneously analyze multiple measurements on individuals or objects under investigation. Thus, any simultaneous analysis of more than two variables can be considered multivariate analysis (Hair et al, 2010).

Factorial analysis

The objective of the factor analysis is to identify the number of factors or underlying dimensions that summarize the information provided by a greater number of variables and interpret what they represent. The factor analysis combines the original variables looking for common elements among them. The group of variables for which this technique was applied were the consumer’s perception of the attributes of Olive Oil.The basic hypothesis to be confirmed with KMO and the Bartlett test are:• H0: the data is not correlated• H1: the data is correlatedIf the statistical significance <0.05, H0 will be rejected, which will confirm that the variables are correlated to the significance level of 5%.

The number of factors is obtained by evaluating the criteria of -Eigen value‖ (> 1), and also refers to the “Scree test plot”Communality is the proportion of the variance of a specific variable explained by all the derived factors.For h2 <5 => the variable is weak to explain the factorFor h2 = 5 => variable is moderate in the explanation of the factorFor h2 5-6 => variable is respectable in the explanation of the factorFor h2> 7 => variable is strong to explain the factor.

The Total Variation Explained is the combined contribution to the total variance of the set of all derived factors. After identifying the factors explained by the variables, an interpretation of these has been made, in terms of the associations with original variables in descending order of importance. Also the derived factors were used in order to profile the study participants using Cluster analysis (cluster analysis).

Cluster Analysis

The objective of this technique is to regroup the subjects (consumers) so that these within a group are very similar to each other and very different from the subjects of the other groups. Therefore, the main concerns of cluster analysis are:decide the number of groups (groups),identify the membership of each group, andprofile the characteristics of each group in terms of behavior, attitudes or characteristics. Conglomerate analysis helped to profile consumers according to the perceptions of the consumption of honey.

For this purpose, the optimization technique was used to identify clusters. This technique groups the objects into a predefined number of clusters in relation to a grouping goal. In order to demonstrate that the scores of the grouping centers are not equal, an ANOVA table was generated using the SPSS 20 program and analyzed using the 5% significance level. The objective variables were the factors derived from the factorial analysis. After having confirmed that the scores of the grouping centers were not equal, an interpretation was made according to the average scores of the factors.

Virtual reality

As the advancement of science suggest the introduction of new models that allow rigorous experimental testing of causal relationships (Siegrist et al.,2018),so, the need to develop new models in the domain of consumer behaviour that allow, in a simpler way than existing models, the examination of consumer decisions in a realistic environment is relevant . (Ung et al., 2018). For this reason, Virtual environments have flourished in the last decade as a way to elicit new and exciting consumer experiences as technological innovation allowed marketers to use such advancements in commercial applications. (Loureiro et al.,2018)


Virtual Reality can be described as the environment “in which the participant-observer is totally immersed in a completely synthetic world, which may or may not mimic the properties of a real-world environment”. (Milgram et al., 1994).

It might be considered also the technology that could enable the development of reliable models that allow efficient and rigorous research in the field of consumer research. (Siegrist et al., 2018)In fact, the use of immersive virtual reality techniques to simulate a grocery store presents exciting opportunities for research into consumer behavior.Through the use of virtual reality, a 3D store environment can be simulated in a realistic and cost-efficient way (Berneburg, 2007; Bressoud, 2013; Ruppert, 2011).

Description of the main experiment

Virtual reality increases the realism of the choice environment compared to lab experiments using textual or pictorial stimuli. This added realism should increase feelings of being present in the virtual reality environment: participants feel as though they are in real store rather than the lab. (Van Herpen et al.,2016)


100 volunteer participants were recruited from different places ( Campus of Aula Dei , Veterinary School of Zaragoza and the home for the elderly of Casablanca ) in order to get a stratified sample. Only individuals from regular honey consuming household were involved.Eligible participants who consented to participate have been randomly assigned to one of four treatment in the virtual supermarket. Each treatment have been assigned to 25 participants.

Design of the experiment

The virtual experiment was constructed in a manner similar to Lusk and Schroeder (2004) and Hudson et al., (2012), whereby respondents faced a series of choice on product origin_ in this case , Mix of honey from EU and Non-EU countries, Mix ofEU honey, honey from Spain , honey from Aragon, honey from Teruel where only the price of the product was allowed to vary.

As the most common format ofselling honey according to the data collected from the distribution channels was 500g containers, we have considered the product as bottle of honey of 500 gand the stated price usedin the CE and VR experiment was on 500g basis. The price for honey in VR were consistent with the prices found in supermarket and hypermarket. To ensure comparability between CE and VR, athe “none of these option” included in the CE was replaced by a bottle of honey with the same label.As shown in table .., the price of honey of mixfrom EU and Non-EU countries ranges from 1,75€ to 5,05€ in 1,1€ increments.For the 4 other honeys, the price ranges from 3,95€ to 7,25€ in 1,1€ increments .


After responding to the survey that we have mentioned in the previous part, consumer were assigned to the virtual reality task. Participants received an oral explanation of the nature of the task and the most adequate way to use the equipment’s (Oculus and joystick). Before the main experiment was started, we made sure that the participant felt confident at performing the task with offering some practice trials.

The virtual environment simulate a supermarket as in thereal life , where consumer can move freely, six shelves of different products were presented and the last one in the extreme right was the honey shelve where consumer where asked to go and choose one bottle of honey of the bottles presented there and add it in there shopping cart .The movement and the choice of the item in the virtual supermarket where assured by the joystick.