Factors of Perception in Novel Food Consumption
gabriel lagunes marti'nez
University of Primorska, Slovenia lagunesgabriel@aim.com
štefan bojnec
University of Primorska, Slovenia stefan.bojnec@fm-kp.si
This paper aims to analyze factors of novel food consumption. The study was conducted in Maribor, Slovenia. Two hundred respondents from two age groups are included in the sample: between 18 and 35 years and between 36 and 55 years. The analysis focuses on quantitative and qualitative factors. The respondent age is the predominant variable to test hypothesis. The Pearson correlation and analysis of variance are used in the empirical analysis. The results show that qualitative factors explained greater statistical differences than quantitative. Key words: novel food, consumption, socio-economic factors, quantitative, qualitative
Introduction
Consumer behaviour regarding food consumption has been studied by economics and behavioural sciences, i. e. psychology, sociology, and anthropology. The analysis of food consumption is a subject of interest to the perspective for economic theory as well as for those who are interested more generically in the behaviour of individuals in their process to make decisions (Muelenber andSteenkamp 1991).
Functional, genetically modified, ethnic, and organic and convenience are just some examples of novel food. The food sector is one of the largest manufacturing sectors in European Union (eu) countries in terms of turnover and employment (Food Drink Europe 2011). The consumers' perceptions of food consumption might change and transform over time (Meulenber and Steenkamp 1991). The ways in which the consumers observe the characteristics of food products could be relative to the learning experience for each product category. Hence, this study aims to explore the consumers' perceptions about the qualitative and quantitative factors of food consumption focusing on the comprehension of consumers' perceptions regarding novel food products consumption.
Backgrounds
According to Krugman (2005), globalisation has made the world more vulnerable. Movements of short-term capital can be a source of instabilities and these have increased with relaxation in the control of financial flows. Financial and economic crises might be a price to be paid for greater openness and integration. On the contrary, Sachs (2005) argued that the global economy is handled in a reasonable manner when facing crises.
This paper focuses on specific types of novel food and its acceptance by local consumers. It is possible that different factors determine a consumer's behaviour such as age, gender and education. The consumers might not accept any changes in their food consumption habits. Many people do not change their habits because they do not know how to change. Therefore, getting consumers to alter their customs can be a process that cannot happen overnight (Ulene and Prochaska 2011).
Consumers might consider combinations of attribute levels when they develop preferences and select food products according to the perception of the value they provide. Therefore, consumers' preferences when making the purchase decision are concerned in determining the contribution of each of the product attributes and their levels (Vázquez 1990). Consumers have a limited capacity to process the information, and because of that, the learning process is important. This process requires a consumer's attention, selection of information and interpretation of perceptions (Bettman, Luce and Payne 1998). Initial experience and expectations predispose learning. Thereby, to have previous knowledge of the characteristics of the product is useful in the buying process. On one hand, it allows the consumer to easily recognize the product they are looking for. On the other hand, it helps to incorporate more information about a new brand (Baker et al. 1986).
Defining the context is important for supply-side and demand-side decisions. The supply-side decision is to define the design for the product/service supply in the market. The demand-side consumer's decision is a selection among several alternatives, subject to checking the availability of budget, time, brand, flavour, regarding the product service that best meets their desires and needs (Ramos 1999).
Product consumption is a part of the life experience of individuals. For this reason, consumers have the ability to associate shapes and colours with certain characteristics that make up categories. The
research regarding the categorisation that makes the consumers' perceptions has focused on how the new products of the category are influenced by pre-existing beliefs and emotions (Loken 2006). A consumer's existing relationship with a brand plays an important role in the brand extension's success. There is ample empirical evidence that well-known brands could benefit from extensions more than brands, which exhibit less success. For example, Smith and Park (1992) argued that stronger brands might have a better ability to reduce perceived risk than weaker ones and showed a positive and significant relationship between brand strength and the brand extension's market share.
Literature focused on marketing and consumer behaviour has proved that geographical origin can contend with price effects, the value of the brand, perceived quality and it supposedly transfers the image of an attitude towards the region where the brand of the promoted products is from (Han 1989; Maheswaran 1994; Gurhan-Canli and Maheswaran 2000). Roth and Romeo (1992) showed that the perceived match between geographical origin and the product could be crucial in consumers' purchasing behaviour.
Following from the exposed literature, the following two hypotheses are proposed:
h 1 The perceptions towards novel food product consumption are influenced by food consumption experiences and economic factors (quantitative).
h2 The perceptions towards novel food product consumption (between female, male, younger and older consumers) are influenced by information on novel food and marketing-promotional activities (qualitative).
It is important to mention that age is the primary variable in the correlation analysis, which helped us to take a decision regarding the tested hypothesis. The gender variable is also studied to show the differences of perception-opinion. The hypotheses are tested in the case of novel food consumption in the town of Maribor, Slovenia.
Methodology
survey design
Slovenia is a relatively small country with a population a little more than 2 million (Statistical Office of the Republic of Slovenia 2013). The survey research was limited to Maribor, the second biggest town in Slovenia. Maribor had 93,847 inhabitants according to the 2012 census (Statistical Office of the Republic of Slovenia 2012). The town
is located in the Northeast of the country, in the Podravska statistical region, 23 km from the Austrian border. Maribor is divided into 33 residential areas. A total of 200 surveys were successfully conducted to make the analyses. The survey's structure considered the age and gender of the respondents. Two age groups of the respondents were considered. The first age group was between 18 and 35 years. This group was composed of 100 people: 50 female and 50 male. The second age group was between 36 and 55 years, and was also composed of 100 people: 50 female and 50 male. The survey's research on novel food was carried out between February and May 2013. It was explained to the respondents that novel foods are considered as functional, genetically modified, ethnic, and organic and convenience food. The respondents were asked to respond to four questions: two questions contained quantitative perceptions and the other two qualitative perceptions. The quantitative perceptions were evaluated by the following two questions: First, please indicate from the following aspects which ones are more important for you when buying food: price, nutritional facts, product quality, and freshness (expiration date). Second, please indicate from the information that appears on the food label which one you read: nutrients, vitamins and minerals, and calories.
The qualitative perceptions were evaluated by the following two questions: First, please indicate from the following aspects which ones are more important for you when buying food: the package (how the product looks), geographical origin, taste, already know the product, and the brand. Second, please indicate from the information that appears on the food label which one you read: list of ingredients and geographical origin.
The division into quantitative and qualitative factors of perceptions was made in order to determine which ones are the most important for the consumers when they purchase a particular product. For example, it can be seen as a hierarchical comparison between price (quantitative value) and taste (qualitative value).
The four questions used a five-point Likert-type scale to investigate how respondents perceived the effects of habit consumption: 1 indicated strongly disagree, 2 disagree, 3 neither agree nor disagree, 4 agree and 5 strongly agree (Likert 1932). For the second and fourth questions the importance was measured through the times the respondent read the information labels: 1 never, 2 rarely, 3 sometimes, 4 often, and 5 always. Two statistical methods are used to test hypothesis: Pearson correlation coefficient and analysis of variance (anova).
We want to test if the set hypothesis is likely to be true with two possible outcomes: to reject the set hypothesis because of insufficient statistical evidence to support the set hypothesis, and cannot reject the set hypothesis because of sufficient statistical evidence in the sample in favour of the set hypothesis.
correlation analysis
The Pearson correlation coefficient between the two variables is used as a measure of linear association between two normally distributed interval variables (Rodgers and Nicewander 1988). It tests whether there are significant differences between two variables using a correlation matrix, where any value less than or equal to 0.05 [(Sig.) < 0.05] would be considered as significant.
analysis of variance The analysis of variance (anova) is a statistical method, which finds if there are statistically significant differences between mean values. This technique allows for the analysing of differences between more than two means. Typical values for a are between 0.05 and 0.01. These values correspond to the probability of observing such an extreme value by chance (Tangren 2002). The anova is used for two subsamples by two age groups to test whether the groups are statistically different.
Empirical Results
To clarify the h1, a comparison of the attitude effects towards the quantitative variables- price, nutritional facts, quality, and freshness with the corresponding age - is tested. Table 1 presents the correlation results between age and the quantitative peers. As can be seen, only in one pair is there a statistically significant difference, between the exposed correlations.
For the first one it can be seen that the correlation coefficient between variables age and good price of the product is negative and weak r = -0.301. A p-value is equal to 0.000, which signifies that is highly statistically significant (p < 0.01). This significant effect explains that this first correlation could be seen in reality. The negative level of association means that variables are moving in opposite directions, the younger the people are, the more concern they have about the importance of the price. When analysing the rest of the peers between age and nutritional facts, product quality and freshness (expiration date) there are not significant effects in their compositions.
Gabriel Lagunes Martínez and Štefan Bojnec
table 1 Correlation Matrix between Quantitative Variables Price, Nutritional Facts, Quality, and Freshness, and Variables Age and Gender
Variable		(1)	(2)	(3)	(4)	(5)	(6)
(1) Age	(a)	l	0.000	-0.30l**	0.ll4	0.024	0.084
	(b)		l.000	0.000	0.l09	0.732	0.237
	n		200	200	200	200	200
(2) Gender	(a)	0.000	l	0.0l8	-0.09l	0.057	-0.101
	(b)	l.000		0.803	0.20l	0.424	0.l56
	n	200		200	200	200	200
(3) Price	(a)	-0.30l**	0.0l8	l	0.092	0.03l	0.046
	(b)	0.000	0.803		0.l95	0.667	0.522
	n	200	200		200	200	200
(4) Nutritio-	(a)	0.ll4	-0.091	0.092	l	0.3l7**	0.295*
nal facts	(b)	0.l09	0.20l	0.l95		0.000	0.000
	n	200	200	200		200	200
(5) Quality	(a)	0.024	0.057	0.03l	0.3l7**	l	0.459*
	(b)	0.732	0.424	0.667	0.000		0.000
	n	200	200	200	200		200
(6) Freshness	(a)	0.084	-0.101	0.046	0.295**	0.459**	l
	(b)	0.237	0.l56	0.522	0.000	0.000	
	n	200	200	200	200	200	
notes (a) Pearson correlation, (b) sig. (2-tailed). **Correlation is significant at the 0.01 level (2-tailed).
Between gender and price, the empirical results are different. One does not find statistically significant differences as the p-value is equal to 0.803 > 0.05, and a correlation coefficient r = 0.018, which is very close to 0. This means that there is almost no correlation between the analysed variables. It is remarkable how the events can change from one variable to another, while age presented highly significant results within the variable price, while gender on the other hand showed opposite results. The similar effect as it was with age can be appreciated for gender and the rest of the combinations. One out of the four peers presented a significant effect, the one between age and price.
Table 2 presents the results of the two-way analysis of variance (anova). Age does not have a significant effect on the importance of quantitative perceptions, since the p-value is equal to 0.388 > 0.05. In addition, the quantitative factors have a significant effect due to p = 0.000 < 0.01. Finally, the interaction effect of both age and quantitative factors showed a significant effect, 0.000 < 0.01. In other words, a difference was not found in the mean importance of age, but there is a difference in the mean importance within quantitative factors as well as in the interaction term of them.
table 2 anova Results between Variables Age, Quantitative Factors and the Interaction of Them
Source	(1)	(2)	(3)	(4)	(5)
Age	0.405	1	0.405	0.746	0.388
Quantitative factors	150.285	3	50.095	92.320	0.000
Age*quantitative factors	15.145	3	5.048	9.304	0.000
notes Column headings are as follows: (1) sum of squares, (2) df, (3) mean square, (4) F-test, (5) sig.
The second question of the survey referred to how often the respondents read the information shown on product labels. The study variables were nutritional composition, vitamins-mineral, and calories with the correspondent variable age. Table 3 presents the results of the correlation coefficients. The variable age did not present any significant difference with the other analysed variables. The pair age and nutritional composition (nutrients) had a very high nonsignificant effect as well as calories, 0.703 and 0.861, respectively. Vitamins and minerals were close to being significant, but the results are above the requested alpha level 0.078 > 0.05.
Gender gives some interesting results. The three peers presented significance differences. Gender and nutritional composition show that p-value is equal to 0.030 at the alpha level of 5%, which makes it significant. The correlation coefficient is very weak and negative, r = -0.154, which signifies that both variables are moving into opposite directions. The remaining pairs gender and vitamins and minerals and gender calories show the same negative correlation coefficients which are equal to r = -0.194 and r = -170 and significant differences 0.006 and 0.016 respectively, at two different alpha levels. This is a good example of some greater differences among the studied variables, where age and gender indicate opposite results. In the comparison between age and gender, three out of the six variables are significant.
Using the same data, a two-way analysis of anova was executed to analyse the quantitative variables previously exposed. Table 4 shows that age did not have a significant effect on the number of times the respondents look at the quantitative perceptions on labels; the p-value was above the alpha level 0.05, as it was equal to 0.182. The quantitative factors had a significant effect equal to 0.023 at 5%. Finally, the interaction effect of them was not significant according to the 0.472 > 0.05.
Based on the results obtained by the Pearson Correlation coefficients and the analysis of variance, the set h1 can be rejected in
table 3 Correlation Matrix between Quantitative Variables Nutrients, Vitamins and Minerals, and Calories, and Variables Age and Gender
Variable		(1)	(2)	(3)	(4)	(5)
(1) Age	(a)	1	0.000	0.027	0.125	0.012
	(b)		1.000	0.703	0.078	0.861
	N		200	200	200	200
(2) Gender	(a)	0.000	1	-0.154*	-0.194**	-0.170*
	(b)	1.000		0.030	0.006	0.016
	N	200		200	200	200
(3) Nutrients	(a)	0.027	-0.154*	1	0.673**	0.669*
	(b)	0.703	0.030		0.000	0.000
	N	200	200		200	200
(4) Vitamins	(a)	0.125	-0.194**	0.673**	1	0.613*
and	(b)	0.078	0.006	0.000		0.000
minerals	N	200	200	200		200
(5) Calories	(a)	0.012	-0.170*	0.669**	0.613**	1
	(b)	0.861	0.361	0.000	0.000	
	N	200	200	200	200	
notes (a) Pearson correlation, (b) sig. (2-tailed). **Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).
table 4 anova Results between Variables Age, Quantitative Factors (Labels) and the Interaction of Them
Source	(1)	(2)	(3)	(4)	(5)
Age	2.407	1	2.407	1.789	0.182
Quantitative factors	10.210	2	5.105	3.789	0.023
Age*quantitative factors	2.023	2	1.012	0.752	0.472
notes Column headings are as follows: (1) sum of squares, (2) df, (3) mean square, (4) F-test, (5) sig.
favour of the null hypothesis. There are not enough statistically significant differences to reject the null hypothesis.
To clarify the set h2 the effects of attitude towards the qualitative variables are compared: Package (how the product looks), geographical location (where the product comes from), and the taste, already know the product, the brand name and the variable age.
The empirical results from the correlation analysis can be seen in table 5. The first pair between age and how the package looks exhibits a non-significant effect. The p-value is equal to 0.075 ^ 0.05. The next pair was faced to geographical origin. In this case it can be detected that they have a significance effect between variables (p = 0.000 < 0.01) at the alpha level of 1%, and a low positive correlation equal to 0.296. This evidence indicates that both younger and older people once think that product origin is important. The
next pair regarding age and taste gives a low negative correlation r = -0.167 with a significant effect equal to 0.018 or the significance level less than 5%. Younger people have different perceptions regarding taste than older people, as for the former group it might be important while the latter group does not pay much attention to this variable. The fourth pair, age and already know the product demonstrates that the level of significance was above the 5% (0.879 ^ 0.05); the coefficient of correlation is close to zero (r = 0.011). This finding proves that there is a very weak correlation between the variables. Finally, the last peer reveals the interaction effect with the brand, which is statistically significant at less than 5% (0.027 < 0.05), and a relatively low correlation coefficient equal to 0.156. Resuming, three out of five peers between age and the qualitative variables presented significant effects at different levels. Campbell (1996) explains that the experience seems to be the principal motivation to learn about new products on the market. Therefore, the consumers' contacts with the product category generates experience and affects the ability to recognize the products and brands (Foxman, Muehling and Berger 1990). Nevertheless, in our experiment the variable age does not have a significant correlation with the variable 'I already know the product' while there is a significant effect with the variable brand name.
The variable gender reveals some intriguing differences. Between the five variables, just the brand name presented a statistical significant effect (0.005 < 0.01), while the rest of variables remained nonsignificant. The difference of the empirical results between age and gender could give another perspective to the research, depending on the perceptions between these two groups of consumers. An interesting to see the results for the variable I already know the product, which contradicts what Vargas (2003) said: Familiarity with the product predisposes the purchase; it does not allow that a misinformed purchase can be made, and finally leads to a trustworthy purchase.
Table 6 shows the results of the anova performance for the qualitative variables. It can be seen that age, qualitative factors and the interaction of them were significant at the 1% level. In other words, there is a significant difference in the mean 'importance of the qualitative perceptions' based on age, qualitative factors and the interaction of them.
The last part of the research was focused on the food label information for the consumers. The investigated variables are ingredients and geographical origin. Table 7 confirms that the pair between age and ingredients is not statistically significant due to p = 0.395 ^ 0.05.
Gabriel Lagunes Martínez and Štefan Bojnec
table 5 Correlation Matrix between Qualitative Variables Package, Origin, Taste, Know the Product, and Brand Name, and Variables age and Gender
Variable		(1)	(2)	(3)	(4)	(5)	(6)	(7)
(1) Age	(a)	l	0.000	0.l26	0.296**	-o.l6y*	-0.0ll	0.l56*
	(b)		l.000	0.0y5	0.000	0.0l8	0.879	0.027
	N		200	200	200	200	200	200
(2) Gender	(a)	0.000	l	-0.023	-0.02y	-0.06l	0.ll9	0.l98*
	(b)	l.000		0.y4y	0.705	0.393	0.093	0.005
	N	200		200	200	200	200	200
(3) Package	(a)	0.l26	-0.023	l	o.l75*	-0.029	0.26l**	0.3l9*
	(b)	0.0y5	0.y4y		0.0l3	0.686	0.000	0.000
	N	200	200		200	200	200	200
(4) Origin	(a)	0.296**	-0.02y	0.ly5*	l	0.l53*	0.l30	0.l33
	(b)	0.000	0.y05	0.0l3		0.03l	0.066	0.060
	N	200	200	200		200	200	200
(5) Taste	(a)	-o.l6y*	-0.06l	-0.029	0.l53*	l	0.027	-o.loy
	(b)	0.0l8	0.393	0.686	0.03l		0.699	0.l32
	N	200	200	200	200		200	200
(6) Known	(a)	-0.0ll	0.ll9	0.26l**	0.l30	0.027	l	0.326*
product	(b)	0.8y9	0.093	0.000	0.066	0.699		0.00
	N	200	200	200	200	200		200
(7) Brand	(a)	0.l56**	0.l98**	0.3l9**	0.l33	-0.l07	0.326**	l
	(b)	0.02y	0.005	0.000	0.060	0.l32	0.00	
	N	200	200	200	200	200	200	
notes (a) Pearson correlation, (b) sig. (2-tailed). **Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).
table 6 anova Results between Variables Age, Qualitative Factors and the Interaction of Them
Source	(1)	(2)	(3)	(4)	(5)
Age	8.836	l	8.836	l0.809	0.00l
Qualitative factors	5l2.824	4	l28.206	l56.836	0.000
Age*qualitative factors	22.304	4	5.576	6.821	0.000
notes Column headings are as follows: (1) sum of squares, (2) df, (3) mean square, (4) F-test, (5) sig.
There is almost no correlation between them since the correlation coefficient is equal to r = 0.060. On the other hand, age and geographical origin displays a statistically significant correlation at the 1% level as p = 0.000 < 0.01. However, the correlation coefficient r = 0.251 is relatively low. It can be seen that the results for gender are the opposite. Ingredients have significant differences while geographical origin does not. There is a significant correlation effect in the first pair as p = 0.001 < 0.01 with a negative correlation coefficient equal to -0.242, while the second pair also presents a negative correlation
table 7 Correlation Matrix between Qualitative Variables Ingredients and Geographical Origin, and Variables Age and Gender
Variable
(1)
(2)
(3)
(4)
(1) Age
(a)
(b) N
0.000 1.000 200
0.060
0.395
0.251* 0.000 200 200
(2) Gender (a) (b) N
0.000	1 -0.242** -0.078
1.000	0.001 0.270
200 200 200
(3) Ingre- (a) dients (b) N
-0.054 -0.242* 0.447 0.001 200 200
1
0.412* 0.000 200
(4) Origin (a) (b) N
-0.190** -0.078 0.412* 0.007 0.270 0.000 200 200 200
1
notes (a) Pearson correlation, (b) sig. (2-tailed). **Correlation is significant at the 0.01 level (2-tailed).
1
table 8 anova Results between Variables Age, Qualitative Factors (Labels) and the Interaction of Them
Source	(1)	(2)	(3)	(4)	(5)
Age	13.323	1	13.323	11.175	0.001
Qualitative factors	7.023	1	7.023	5.891	0.016
Age*qualitative factors	6.003	1	6.003	5.035	0.025
notes Column headings are as follows: (1) sum of squares, (2) df, (3) mean square, (4) F-test, (5) sig.
coefficient (r = -0.078) but with no significant correlation coefficient according to p = 0.270 > 0.05.
Finally, the anova was performed. Table 8 suggests that age has a significant effect as p = 0.001 < 0.01. The qualitative variables also have a significant effect (p = 0.016 < 0.05) as well as the interaction of them (p = 0.025 < 0.05). The results suggest that age, qualitative factors and the interaction of them have statistically significant differences in the mean how often consumers see the labels of the products they buy.
Taking into account the results obtained by the Pearson Correlation coefficients and the anova, the set h2 cannot be rejected: The perceptions towards novel food product consumption are influenced by information on novel foods and marketing-promotional activities.
Conclusions
The empirical analysis confirmed that there are significance differences in consumer perceptions when buying novel food, regarding
age. The comparison between the variables age and gender showed that the preferences-perceptions regarding the novel food could vary in a significant way. The correlation analysis confirmed the importance of quantitative and qualitative perceptions. The correlation coefficients were more significant for age and the qualitative perceptions. The anova confirmed the most significant results in the qualitative variables and the age variables, and the interaction of them.
The perceptions of younger and older consumers can be perceived in a very different way when making the statistical analysis. According to the set h2, consumers can perceive towards the brand name and the product origin. This suggests that in the case of new products, there is a greater probability of acceptance by the consumer when there is a similarity of the product either with a category previously related to, or with a specific known brand (Barone, Miniard and Romero 2000) or it is supported by the strong promotional campaign made by marketing companies and social networks (Subramani and Rajagopalan 2003).
It is possible that consumers' experiences can be transcendental when buying products and services. Moreover, as argued by Erdem and Keane (1996), consumers can react adversely to the variation in product attributes. Therefore, a change in product attributes can cause a lack of credibility in content or distrust. For future research, it would be interesting to further investigate whether Slovenian consumers would be willing to accept a change in their consumption habits and thus distrust in food consumption.
References
Baker, W., W. Hutchinson, D. Moore, and P. Nedungady. 1986. 'Brand Familiarity and Advertising: Effects on the Evoked Set and Brand Preference.' In Advances in Consumer Research, vol. 13, edited by Richard J. Lutz, 637-42. Provo, ut: Association for Consumer Research.
Barone, M. J., P. W. Miniard, and J. B. Romero. 2000. 'The Influence of Positive Mood on Brand Extension Evaluation.' Journal of Consumer Research 26 (4): 386-400.
Bettman, J. R., M. F. Luce, and J. W. Payne. 1998. 'Constructive Consumer Processes.' Journal of Consumer Research 25 (3): 187-217.
Campbell, J. 1996. 'Shape Properties, Experience of Shape and Shape Concepts.' Philosophical Issues 7:351-63.
Erdem, T., and M. Keane. 1996. 'Decision Making under Uncertainty: Capturing Dynamic Brand Choice Processes in Turbulent Consumer Goods Markets.' Marketing Science 15 (1): 1-20.
Food Drink Europe. 2011. 'Data & Trends of the European Food and Drink Industry 2011.' http://www.fooddrinkeurope.eu/uploads/ publications_documents/Final_Data_Trends_304.2012.pdf
Foxman, E. R., D. D. Muehling, and P. W. Berger. 1990. 'An Investigation of Factors Contributing to Consumer Brand Confusion.' Journal of Consumers Affairs 24 (1): 170-89.
Gürhan-Canli, Z., and D. Maheswaran. 2000. 'Cultural Variations in Country of Origin Effects.' Journal of Marketing 37 (3): 309-17.
Han, M. 1989. 'Country Image: Halo Or Summary Construct?' Journal of Marketing Research 26 (2): 222-9.
Krugman, P. 2005. 'Crises: The Price of Globalisation?' In Economics of Globalisation, edited by P. Gangopadhyay and M. Chatterji, 163-80. Aldershot: Ashgate.
Likert, R. 1932. 'A Technique for the Measurement of Attitudes.' Archives of Psychology 22 (140): 5-55.
Loken, B. 2006. 'Consumer Psychology: Categorization, Inferences, Affect and Persuasion.' Annual Review of Psychology 57:453-85.
Maheswaran, D. 1994. 'Country of Origin as a Stereotype: Effects of Consumer Expertise and Attribute Strength on Product Evaluations.' Journal of Consumer Research 21 (2): 354-65.
Meulenberg, M. T. G., and J. E. B. M. Steenkamp. 1991. 'El analisis del consumo de alimentos y la eleccion de los productos alimenticios.' Revista de estudios agrosociales 157:125-51.
Ramos, A. M. 1999. Análisis de las preferencias del turista mediante análisis conjunto: el caso de Tenerife. Tenerife: Departamento de Economía de las Instituciones, Estadística Económica y Econometría.
Rodgers, J. L., and A. Nicewander. 1988. 'Thirteen Ways to Look at the Correlation Coefficient.' The American Statistician 42 (1): 59-66.
Roth, M. S., and J. B. Romeo. 1992. 'Matching Product Category and Country Image Perceptions: A Framework for Managing Country of Origin Effects.' Journal of International Business Studies 23 (3): 47797.
Sachs, J. 2005. 'Preface.' In Economics ofGlobalisation, edited by P. Gan-gopadhyay and M. Chatterji, xv-xviii. Aldershot: Ashgate.
Smith, D. C., and C. Whan Park. 1992. 'The Effects of Brand Extensions on Market Share and Advertising Efficiency.' Journal of Marketing Research 29 (9): 296-313.
Statistical Office of the Republic of Slovenia. 2012. 'Population in the City of Maribor in the Year 2012.' http://www.stat.si/krajevnaimena/ default.asp?txtIme=MARIBOR&selNacin=celo&selTip=naselja&ID =2620.
-. 2013. 'Population by Settlements, Detailed Data, Slovenia, 1 Jan-
uary 2013 - Final Data.' http://www.stat.si/eng/novica_prikazi.aspx? ID=5530.
Subramini, M. R., and B. Rajagopalan. 2003. 'Knowledge-Sharing and Influence in Online Social Networks via Viral Marketing.' Communications of the acm 46 (12): 300-7. Tangren, J. 2002. 'Basic Info: What Is an anova?' Washington State University, Tree Fruit Research and Extension Center, Washington, dc. http://www.tfrec.wsu.edu/anova/basic.html Ulene, V., and J. Prochaska. 2011. 'Why Are Unhealthy People So Reluctant to Change Their Lifestyles?' http://articles.latimes.com/2011/ may/23/health/la-he-the-md-change-illness-20ii0523. Vargas, L. 2003. 'Procesamiento de información y familiaridad de marca.' Analisi Quaderns de Comunicació i Cultura 30 (3): 225-40. Vázquez, R. 1990. Investigación de las preferencias del consumidor mediante análisis de conjunto: importancia para el diseño de nuevos productos. Oviedo: Facultad de Ciencias Económicas.

This paper is published under the terms of the Attribution-NonCommercial-NoDerivatives 4.0 International (cc by-nc-nd 4.0) License (http://creativecommons.org/licenses/by-nc-nd/4.Q/).