Assessing the drivers of traditional and local products 2 consumption in Europe . Are there any contradictions ? 3

This study assesses attitudes of young adults' (18-30 years old) consumption on local and 12 traditional products in7 European countries. A clustered sample (n=836) from natives of Greece, 13 Bulgaria, Romania, Slovenia, Croatia, Denmark and France was collected, by distributing 14 questionnaires through social media and university mail services. Sample was examined by 15 implementing Principal Component Analysis (PCA) in three different samples; overall and two 16 subgroups, Eastern and Western European countries. Six major factors revealed: consumer 17 behavior, health issues, cost, influence from media and close environment and availability on store. 18 As a result, young adults have a positive attitude to local and traditional food products but they 19 express insecurity for health issues. Cost factor influences less people from Eastern European 20 countries than those from the overall sample (3rd and 5th factor accordingly). Influence of close 21 environment is a different factor in Eastern countries comparing to Western ones that it common 22 with influence from media. Females and older people (25-30 years old) doubt less about TFPs, 23 while media have high influence on consumers’ decisions. Aim of this survey is to create consumer 24 profiles of young adults and create different promotion strategies of local and traditional products 25 among the two groups of countries. 26


Introduction
It was no longer than 1919 when French agronomists requested in an intense way from their government to protect by law the quality of their wines from Bordeaux region [1] and set the basis for later European Union adequate legal framework.Since July 14 th 1992 EU has established and implemented a protective agenda for products with locality characteristics, regulating (R2081/92 and R2082/92) for Protected Designation of Origin (PDO) Protected Geographical Indication (PGI) or Traditional Specialities Guaranteed (TSG) for food products.Moreover, European Union's (EU) bilateral agreements with Canada, USA and China secured those products in these markets, by expanding the aforementioned legal framework, confirming on practice that locality characteristics can be an economic growth factor for these specific production areas [2].On July 6 th 2017, European Commission (EC) reached an agreement with Japanese government for exports of more than 200 European Geographical Indication products, proving the dynamics of such production methods even on in regions with completely different culture.As agreements and increasing number of submitted geographical indication products per year indicate, there is a rising universal interest for 2 of 22 products with certain sensory characteristics, a unique identity on production process accompanied with a tradition on how and when people taste it.Until today, 1403 products are submitted on EU as PDOs, PGIs and TSGs, the majority of them belonging to Southern Europe countries (Italy(293), France(244), Spain(194), Portugal(138),Greece(105)) and only few of those products originate from Northern Europe countries (Finland (10), Sweden (8), Denmark ( 6)) [3] .
In an attempt to clarify definition of Traditional Food Products (TFPs) for European consumers,Vanhonacker's et al survey underlines the importance of cultural and territorial identity, transferability from one generation to another, processing and sensory characteristics and describes different attitudes between countries [4].TFPs concept has been analysed through personal interviews resulting in the same definition as the previous survey [5].In another survey about locality index consumers have positive feelings both for consuming local products and creating value for the local community but consumers express negative attitudes for cost and difficulties on use [6].European Union's definition about TFPs introduces the sustainability term about local environments and underlines the importance of labelling [7].Euro Food Information Resource (FIR) consortium definition emphasizes on raw materials and production process while simultaneously there is an effort for transcribing local recipes and their nutritional value in order to maintain tradition [8].

Literature review describes an overall image about European population attitudes towards
TFPs.This survey gives an insight on young consumers' beliefs about traditional products.
Adolescents with a higher education level are considered as a health-sensitive group [9].Young adults present a more snack-related food behaviour comparing to overall population [10].Even from human physiology aspect young people receive more sensory characteristics than older ones [11].It is noteworthy that in this age majority of people leave their parental homes, creating new food behaviour.In this research, a questionnaire based survey was conducted by implementing a merged model based on the Theory of Planned Behavior (TPB) and the Health Belief Model (HBM) so as to investigate attitudes that are connected with traditional food consumption and perceived risk for young adults' health.Final goal is to determine a marketing strategy that fulfills young consumer needs.
Theory of Planned Behavior (TPB) [12,13] is an extension of Theory of Reasoned Action (TRA) that was developed in 1980 by, Ajzenand & Fishbein [14].TPB examines attitudes in three levels: (a) Attitude towards the behavior, (b) Subjective norm and (c) Perceived behavioral control.
• (a) Attitude towards behavior is the degree of approval or disapproval of a certain behavior by a person.A rising interest for products with unique characteristics and an increasing consumption rate has created a positive image of TFP [15].Therefore, there is a negative attitude for health issues about production and manufacturing procedures, especially from people of higher education [16].
• (b) Subjective norm is the social attitude about a person's behavior.(b) Norms derive from close environment (family, friends) and media.Family is responsible for creating food and eco-friendly attitudes to a person [17,18], while an information from a friend's mouth can severely influence a person's behavior [19].Increase of cooking related shows is not random, while there is an effort from industry's side for greater consumption [20].European Union presents many differences on food behavior across different cultures, that sometimes is a barrier for consuming certain products [21].
TFPs can be a sign of identity and a link to their roots for people that live away from their home countries, differing them from inhabitants [22].
• (c) Perceived behavioural control is the level of difficulty for a person to perform a certain behavior.
Safety standards, longer self-life and consumer friendly appearance demand from traditional food products to be modernized.Innovation on TFPs brings new technologies on safety issues during production process, nutritional value and packaging.Quality maintenance is the crucial point for acceptance or not of an innovation from consumers.New technologies can help certain TFPs to be accepted easier from higher ages (low fat, less salt) [23].Additionally, innovation is indispensable for TFPs in order to be accepted from food supply chain.TRUEFOOD Integrated project (2006)(2007)(2008)(2009)(2010) supported by European Commission communicates a series of promotion tools for TFPs based on modern surveys according with current market needs [24].The above factors determine the intention of a person to perform a behavior.Strong intensions are more likely to be expressed as a behavior than weak ones.
Health Belief Model (HBM) is being used in the food sector in order to measure individuals' perception for certain products and their opinions about effects to their health [25].Socioeconomic characteristics like age, gender, ethnicity, level of education, income and type of job are also estimated in this model.According to Pechey et al, 60% of people with low socioeconomic status buy food with hazardous effect for their health [26].Men are found more spontaneous and enthusiasts while women focus to morals that will receive, when purchasing a certain product [27].South Europe presents a higher consumption level of TFPs than north, meaning that ethnicity do affect relationship with TFPs [28].Pieniak et al in their study among six European countries found a general difficulty on consumption of TFPs as well as strong doubts about health issues [29].There is though a quite important issue regarding the etymology of the word Ethnicity for food consumption, because it influences performed behavior of a person not only in a national level but also in a regional one [30].In the same survey, men found to spend more on organic food than women.
Cost can be a significant factor on final purchasing decisions, as there are groups of people that are described as hyper-sensitive especially on online markets [31].Sales of healthy products increased by 10% when they were on discount, independently of education level [32].Cost should be considered in accordance with quality.In survey of Di Vita et al, assessing consumer purchasing decisions on different types of olive oil (conventional, PDO and organic) resulted that highest influence factor was price for people from rural areas but for urban population area of origin was the most important one [33], while on another survey for purchasing local jams, quality had the major role [34].Moreover cost, availability and preparation time are limiting factors for TFPs consumption [35].From consumers' view organic products are considered of high quality ones and more expensive than conventional.On the other hand, TFPs preserve quality element but they are not accompanied with high price beliefs [36].Willingness to Pay (WTP) was investigated for TFPs because they are considered as premium products and thus they should have a different pricing policy [37].
Experience on store should be "positive","pleasant" and "reliant" as survey of Walsh et al indicates [38].Trust between supplier and seller, strength of habit and personal preferences are the most significant factors of meat purchasing behavior [39].Same issue of trust about local market is confirmed by Migliore et al, mentioning again health issues [40].Consumption pleasure rates are higher when people are informed about products that are going to taste, especially for young consumers [41].Local people are habitually connected with traditional products, making their consumption a necessity for them [42].Positive images of using TFPs from celebrations and special occasions accompany them in their adulthood [43].There is a rising interest of people caring not only for quality of products but also for animals' welfare [44].Small and medium enterprises (SMEs) should take advantage of this personal communication with local people and promote themselves [45].Special identity of TFPs characterizes them unfavorable for new markets and they should be promoted appropriately to gain entrance in a wider range [46].

Materials and Methods
Literature review about consumer behavioral theories was conducted in order to identify the most important factors that influence the European consumers' perceptions about local and traditional products.More specifically, social, demographic and psychological factors are of major importance on influencing decision making process on TFPs consumption.European Union presents a high heterogeneity as it comes to cultural aspects making it an ideal area for such surveys [47].In addition, shop selection and attitudes towards a more environmentally friendly consumer profile, have been searched.As it was aforementioned, there is a general knowledge about European perceptions on TFPs but there is a need to broaden the knowledge about consumer behavior of the young consumer community, which is in fact the dynamic purchasing market for the next 30years.
For this reason, a field, questionnaire based, research was conducted on a European level, during the period September -October 2016.An Overall stratified sample of 836 respondents has been collected.Sample was composed from 295 males and 541 females of a mean age 23.3 (St.Dev =3.34).
A higher rate for female respondents has been observed from other papers as well [48].Educational level is pretty high while 57.68% of participants had already a degree.Low income was another characteristic of this dataset with 58.49% receiving a monthly income lower than 500€per month.
Unemployment ratio was a little lower than EUROSTAT (7.7%) for 2016 [49] compared to overall sample (5.98%).The questionnaire was based on the above mentioned Theory of Planned Behavior (TPB) and Health Belief Model (HBM).
In the first part of the questionnaire there are questions about the social and demographic data of the respondents, such as gender, age, income and level of education.The second part introduces to consumers a group of proposals.More specifically, this part consists of four groups of questions examining perceptions about TFPs: consumer behavior (group 1), health issues (group 2), cost (group 3) and degree of influence from different factors (group 4).The Third part focused on re-examining the respondents' opinions about the previous referred groups and also there were 2 questions referring to the availability of products in stores (group 5).For these questions a Likert scale was used from 1 (Absolutely disagree) -5 (Totally agree).
The questionnaire was initially tested before distribution in eight (8) European countries in which it was to be distributed (Greece, Denmark, England, France, Slovenia, Croatia, Bulgaria, and Romania) to verify that the questions were adequately written and being understandable.
In order to evaluate reliability of results, two tests were used: • The Kaiser-Meyer-Olkin index (KMO) • Barlett's sphearisity test The Kaiser-Meyer-Olkin index (KMO) explores the suitability of the sample to be analyzed.In particular, it examines the relative magnitude of correlation coefficients in relation to the partial correlation coefficients.The higher correlation, the better is for sample to be analyzed.When KMO value is too low, less than 0.5, means that the factorial analysis will not conclude to satisfactory results.Values above 0.7 are acceptable for analysis, and 0.8 are great.
With Barlett's sphearisity test, it is examined whether the observed correlation of the data table, let R = (rij) (p x p), differ statistically from its actual identity.Therefore, the zero hypothesis (H0) is that the data is arranged in a rectangular form.This assumption is controlled at a 5% materiality level.
Principal Component Analysis (PCA) was developed by Charles Spearman.Implementation of the method is clear in areas such as psychology, market research, the labor market and human resource management, where quantitative measurements produce results that help in making critical decisions.It is a method that is being used to analyze large datasets, and in this try it eliminates information in order to form groups (components).Two main elements of PCA are: factor loading and sum of squared loadings.Factor loading interprets whether a set of data affects the factor that has been formed.If the participation ratio is fairly low, below 0.4 then it is assumed that its influence is too small and for this reason it is rejected, from 0.5 -0.7 the degree of participation is considered satisfactory and from 0.7 and above it can to explain to a great extent the factor.Sum of squared loadings describes the amount of variance of a factor.PCA was performed with SPSS 23 statistical software package.

Demographics
Table 1 gives an overview of characteristics of Overall sample (n=836), which was derived from questionnaires completed by 18-30 years old people from different European countries (Greece, Bulgaria, Romania, Slovenia, Croatia, Denmark, France, and England) with stratified sample to produce statistically significant results, consisted of 35.3% of men and 64.7% of women.About 70% of the participants are from 18 to 24, while about 30% are from 24-30.This is mainly due to the way in which questionnaires were distributed, because distribution has taken place over the internet and to a large extent through university networks.That explains the lower unemployment rates of the sample, compared with those of the EU.Moreover, educational level of Overall sample is rather high, with 35% holding a postgraduate degree.In terms of income criteria, about 60% have an income of <500 €/month, while 22% are in the income class of 500-1000 €/month, covering a total of 82% of the distribution.
Table 1 Further analysis was inducted, dividing the overall sample in two subgroups; Eastern European Countries (Greece, Bulgaria, Romania, Slovenia, Croatia) and Western ones (Denmark, France, England) in order to further investigate possible significant differences.For this reason, Table 2 presents an overview of the sample characteristics resulting from Eastern European questionnaires (n=569), which consists of 34.6% of men and 65.4% of women.About 75% of participants are from 18-24 while about 25% are from 24-30.The educational level of the sample is quite high since 32% hold a master's degree, while about 75% have incomes <500 €/month, and 14% have income of 500-1000€/month.Table 3 To sum up, it is noted that the overall sample refers to young people with an average age of 23 years, a high educational level, while the economic criteria differ for the individual categories, with Western countries having higher monthly incomes.

Overall sample
For Principal Component analysis of the overall sample (n = 836), KMO index (= 0.793) showed a high degree of consistency between variables.Analysis carried out revealed 5 main factors reflecting 61% of total variance.Finally, by comparing the importance of the components and their rating according to the socio-economic characteristics (gender, age, education and income), some differences can be found, which are summarized in the following table.Overall Sample 1st Component: Consuming TFPs is a good and healthy consumer behavior.In this component it appears that women tend to give even greater importance to hygienic behavior (p-value <0.05), confirming that there are differences in gender consumption behavior, as described in the literature review.
2nd Component: Emphasis is given on questions about insecurity on consumption of TFPs.It is worth noting that this feeling is negatively correlated with age, while the higher the level of education, the less insecurity and fear.
The first two components refer to two different perceptions but the first component (positive perception) is almost two times more important than the second one (negative perception).As in the bibliographic review, it was emphasized that consumers are rather skeptical about their use due to the lack of food safety criteria that are being observed during their production process.
3rd Component: The role of the economic dimension does not overshadow the first component that highlights positive factors such as health and good consumer behavior.This is perhaps due to the fact that even if local products are sometimes more expensive than conventional ones, they provide consumers an additional degree of satisfaction.This additional satisfaction, in comparison with conventional products, provide hints for extra loyalty between producers and consumers of TFPs.
4th Component: Explains the influence of the media and the surrounding environment.The role of media and friends increases with age (positive correlation).Influence is on average higher for women than men.
5th Component: The last component is the degree of difficulty in finding and purchasing local traditional products.Consumers do not have a particular problem with finding local products, since both variables have a high degree of participation (> 0.700).

Eastern European Countries
The main difference with respect to the overall sample is the economic benefit resulting from the The first component is particularly strong consisting of 9 variables.Positive consumer behavior is directly related to the opinion of the immediate environment as well as the economic benefit, which is not a separate dimension.As a consequence, the economic dimension for young people in the Western world is not as important as the previous group and is not, by itself, a dominant factor.The second component refers to the impact of traditional products on health with a large contribution of individual factors (> 0.750), which once again confirms the lack of trust for TFPs.The third component expresses the degree of influence from media, which appears to be a component in itself.
Finally, in the fourth component there is a difficulty on finding and identifying local products, as many declare that it is difficult to find them, while others verify that traditional products are available in supermarkets.
Having in mind the finding of this survey, it is noteworthy to clarify promotion methods for each market segment.In particular, two clear consumer profiles are presented οn Table 8.Five main dimensions can be considered are significant pillars for introducing and implementing a promotion plan for local and traditional products.
Table 8 1 st Dimension: It is very positive that the consumption of local and traditional products is a good consumer behavior for young consumers in both Eastern and Western European countries.
Therefore, since the perception of these products is already positive, there is no need to further insist on this issue.
2 nd Dimension: TFPs should inspire safety to consumers both in terms of production process and packaging.Safety issue is extremely important as almost all research findings verify this, with this particular field research confirming it as well.It is therefore of major importance that TFPs should have all corresponding certifications in order to eliminate insecurity from consumers.The same sense of security should be depicted for packaging, by using the appropriate technology and aesthetics.
3 rd Dimension: Cost dimension was one of the factors introduced in the questionnaire but it was quite ambiguous the way it affects consumer behavior and more precisely younger audience with lower incomes.So it appeared that cost comes as 5th Component for the Eastern countries compared with the overall sample where it appears as 3rd component.This was quite surprising, as eastern European countries, facing greater financial difficulties, seem to be willing to pay more for purchasing quality products and more precisely TFPs.Therefore, for firms producing or marketing such products, there is a potential for greater profitability in Eastern European countries.
4 th Dimension: According to the Theory of Planned behavior, individuals appear to be affected by subjective norms.It seems that the opinion of close environment is, for the sample of the Eastern European countries, a fourth component in contrast to the young people of Western European countries, which, in that case, is not a separate component.This information can be immediately utilized as it is fully understood that young people in Eastern countries are more easily affected by their friends or close related people.In addition, thematic events could be organized referring to such products and their uses so as to give an overall approach of their special characteristics to the public.In this way, they have the potential to influence both participants and individuals from their nearby environment, influencing them positively about their products.
5 th Dimension: Availability dimension, which is a separate component for all three samples, appears to be of great significance.It is worth mentioning that interpretation of results is a model of analysis of perceptions rather than events.In other words, consumers believe that these products are not in hypermarkets even if they are there.This is a fairly serious problem, as if consumers have difficulty finding them, then they are not able to buy and consume them.For this reason, it is proposed either to place TFPs is separate places in supermarkets, or for companies distributing their TFPs through hypermarkets to create special stands for their products under a clear thematic approach, so as to be easily recognizable from consumers.It could also be part of an advertising campaign, focusing, among others, that these products can be found in specific chain stores.

Conclusions
This study focuses on the consumers' attitudes and perceptions on TFPs, trying to describe the factors driving their final consumption choices.Through this research, the profile of new consumers, who will be the most important market segment of the next decade, is described.Main research findings are summarized as follows: • Positive image for TFPs among young adults • Health issues are highly important and thus TFPs should inspire safety with their certifications and packaging • Cost is a factor that affects less young adults from Eastern European countries than those from Western countries • Media affect more adolescents from Western countries while influence from close environment is more significant for people from Eastern European countries • Availability of TFPs is an issue for young consumers and certain actions from suppliers should be taken.
Further research is essential to be made by focusing on both specific markets and products.As it was already mentioned, Europe is an area with many different cultures and each case should be examined separately.This new knowledge is essential for formulating more efficient promoting strategies towards increasing market shares of TFPs globally, by incorporating the values of European culture into dietary habits.2.8 I intend to increase consumption of local and traditional products.
2.9 I want, from now and on, to consume local and traditional products.
Part III: Please rate the following sentences (1-5 scale) 3.1 Health is better than wealth.
3.2 Consuming local and traditional products is dangerous for my health.
3.3 I am afraid of jeopardizing my health by consuming local and traditional food products.
3.4 Consuming local and traditional products can cause irreversible damage to my health.
3.5 Consuming local and traditional products is economically beneficial.
3.6 Consuming local and traditional products is beneficial for my health.
3.7 It's hard to find local and traditional products.
3.8 I buy local and traditional food products from small local shops.
3.9 I buy local and traditional food products from supermarkets.
3.10 Media persuade me so as to consume local and traditional products.
3.11 Media persuade me so as to consume healthy products.
3.12 My friends influence me so as to consume local and traditional products.

Table 3 ,
which consists of 36.7% of men and 63.3% of women.About 60% of participants are from 18-24 while about 40% are from 24-30.The educational level of the sample is quite high, with 41% holding a postgraduate degree.There are noticeable differences with respect to Table2concerning income criteria, since approximately 22% have an income of<500€/month, while 37% are in the income class of 500-1000€/month, covering a total of 82% distribution.

Table 4
presents the factor loadings, which are quite satisfactory (> 0.500) to very satisfactory (> 0.700), except of two cases where the index value is marginal (0.400-0.500).[Q2.4People, that their opinion is important for me, approve buying and use of local and traditional products.(0.470),Q3.12Myfriendsinfluencemeso as to consume local and traditional products(0.453)]Moreover 4 questions removed due to their very low loadings (<0.200)Q2.6Itdepends on me if I will consume or not local and traditional productsQ2.7 I don't feel well when other people see me buying local and traditional productsQ3.1 Health is better than wealth Q3.8I buy local and traditional food products from small local shops.

Table 2 :
Demographics of East European countries sample (n=569 )

Table 3 :
Demographics of West European countries sample (n=267)

Table 7 :
Linear regression of socioeconomic status with factors of two subgroups, East and West European countries accordingly.

Table 8 :
Eastern and Western European young adults' profile Preprints (www.