The influence of education level on choosing coastal regions as tourist destinations

The main aim of the paper is to investigate the influence of formal education level on decision-making process when choosing a tourist destination based on multivariate analysis of variance (MANOVA). The survey was conducted on the sample of 252 respondents from Bačka region (Vojvodina/Serbia). Also, this study strives to examine the influence of education level on decision-making process including all five phases of decision-making process: need awareness, information search, alternatives estimation, purchase and purchase evaluation, applied to the process of choosing coastal regions as tourist destinations. The study shows that education level is related to four out of five phases of decision-making process (only in case of need awarness there is no statisticaly significant difference). This is especially important for creation of a marketing platform with promotional acitivities adjasted to different market segments diferentiated by education level. Moreover, the study discusses differences in behaviour of diffrent educational groups while choosing travel destinations.


Introduction
Achieving success on travel market depends mainly on the knowledge of the characteristics of potential tourists, their needs and motivations, behaviors in the decisions making process about the choice of tourist destination as well as the factors that affect that behavior.
A research on potential tourist behavior is important for companies in tourism field in order to create value and adequately communicate with users about their tourist products and services.That is why research on tourist behaviour and decision-making is critical to business success of every company on the tourist market and thus the first researches on the topic of decision making represent the cornerstone in marketing and consumer behavior.
One of the most important segments of the research regarding customers behavior in tourism is the process of decision making about a tourist desti-nation (Djeri, et al., 2007).Tourist's decision-making processes are complex; involve a wide array of single decisions from deciding 'where to go' through 'what are we going to do now we're here' and beyond.Many choices are based on contextual 'facts'.Many more are based on perceptions of relatively high-risk decisions judgments, that is, no-one knows how 'good' their holiday is going to be until they are experiencing it (Smallman, Moore, 2009;Marion, 2016).
It is evident that geospace, with its natural and cultural values, is a significant resource in tourism industry (Ćopić, et al., 2014;Kasagranda, et al., 2016).This also refers to coastal regions as important parts of destinations' natural values.This study aims to analyse the influence of formal education level of tourists on the process of choosing coastal regions as tourist destination by applying the classical model of "decisionmaking process".This model treats a potential tourist as a person who, by making a decision about a tour-ist destination, solves his problem, and it involves five phases: need awareness, information search, alternatives estimation, decision about purchasing a tourist product or service and purchasing evaluation (Kotler, Keller, 2006;Mathieson, Wall, 1982).The main hypothesis on which this study is based is: H1.There is a statistically significant difference among respondents of different education level in all five phases of decision-making process when choosing a tourist destination (Figure 1).

Literature review
Due to the specificity of the tourism product, primarily its intangibility, customer's information seeking prior to purchase and alternative estimation, is emphasized in tourism industry like in no other sector (Decrop, 2006;Sirakaya, Woodside 2005;Blešić, et al., 2011, Mc-Cabe, et al., 2016).It is therefore extremely important to understand the processes and phases of decision making on taking a trip to a particular destination.
There is a wide array of different theories concerning the behaviour of potential tourists when it comes to decision making about a tourist destination.The analysis of the decision-making involves researching into how they choose among different tourist destination offers, as well as processes that precede and are a part of the decision-making itself.Moreover, many of the earlier literature on consumer behavior research focuses on the distribution of gender roles (Filiatrault, Ritchie, 1980;Jenkins, 1978;Nichols, Snepenger, 1988;Blešić, et al., 2014) and cultural influences (Lee, Kalen, 2009).Also the economic and non economic factors influence on decision making (Davies, Downward, 2007;Djeri, et al., 2014).Also very popular are the studies about children's influence over family destination decision making (Kozak, 2010;Labrecque, Rich-ard, 2001;Bronner, Hoog, 2008;Borgeman, Van der Poel, 2006;Foxman, et al., 1989;Howard, Madrigal, 1990;Palan, Wilkes, 1997;Thornton, et al., 1997;Beatty, Talpade, 1994;Nickerson, Jurowski, 2001;Wang, et al., 2004) but also those dealing with the influence of education on consumer behavior on the insurance market (Ioncică, et al., 2012).
The classical concept of prescriptive, analytical everyday decision making (Edwards, 1954;Von Neumann, Morgenstern, 1944) claims that people collect and analyze information, eventually selecting an optimal solution from the range of alternatives (the 'choice set').They work on it by evaluating the advantages and disadvantages of each possible outcome, choosing the most appropriate ones to achieve their desired objective.Thus, one of the major factors influencing consumer decisions to purchase a product or service is information about the product or service.Information search or information seeking is the process of consulting various sources before making a purchasing decision (Anderek, 2016).Choosing and buying products involves either dependent or independent decisions and thereby a decision-making process.In particular, the literature entails an increasing number of both conceptual and empirical studies over the last two decades concerning decision making in tourism and leisure choices.These studies relate specifically to the influence of internal and external factors (e.g., income, marital status, gender, e-word of mouth, and motivations) on decision making regarding vacations (Hernández-Méndez, et al., 2015;Adzam, et al., 2016;Um, Crompton, 1990;Woodside, Lysonski, 1989;Nichols, Snepenger, 1988).Um and Crompton (1990) were identifing the role of attitudes in an individual's pleasure travel destination choice process.In the mentioned study potential travelers' awareness sets and evoked sets were defined and identified longitu- However, there is little research addressing the influence of education level on the process of decision making concerning the travel destination.The influence of education level in chosing travel destination is mentioned in only one tourism paper (Wong, et al., 2016), and the results revealed that more educated people engage in international travel more often, while those less educated travel within the state more often.Studies dealing with influence of education on travel decision making are not found in this field.This study is basen on the assumption that education level of potential tourists may not only affect their preferences, but also information seeking prior to purchase, the way of decision making, and selection of specific destinations.

Sample
The sample consists of 252 respondents whose residence is in Bačka region (the region of Vojvodina/Serbia).There was a slightly higher number of female respondents, and the most frequent age category was 26-35 (36.11% of respondents).More than half of the sample are married respondents, and they mostly are highly educated.Sample characteristics are shown in Table 1.

Questionnare design
The research was conducted through a questionnaire which consisted of two parts.The first part of the questionnaire referred to the analysis of social demographic characteristics of the respondents (gender, age, education, montly income and employment status).The second part of the questionnaire consisted of questions referring to all five phases in the decision making process when choosing a tourist destination (see Tables 2, 4, 6, 7,  9).The number of questions differs according to the features of the phases they refer to.

Procedure
The research was conducted among the population of Bačka region (the region of Vojvodina/Serbia), and the sample which represents the basic span consists of individuals who consume the services of selected travel agents in -Novi Sad (18 agencies), Subotica (6 agencies), Sombor (4 agencies), Backa Topola (3 agencies), and Vrbas (4 agencies) were interviewed over the phone and asked to participate in the research.From the total of 35 travel agencies, 27 participated in the research.
The selected agencies are located in the largest cities and physically are placed on the different parts of the territory of Bačka region in Vojvodina/Serbia.Travel agents sent the questionnaires to the costumers by e-mail, after their return from the holiday.The sample was random and the survey was conducted from January till December 2012.About 600 questionnaires were sent but only 252 were returned.The questionnaires reflect the tourists opinion upon the process of decision making after selecting a tourist destination (in this case coastal regions).Questionnaires have been analyzed according to answers of 252 participants, but what lacks is the estimation of frequency variation within different phases.Indicators for all five phases of decisionmaking process were identified in order to measure the influence of the level of education on different phases of decision-making process when purchasing a tourist product.Indicators are described in precisely determined phases of decision-making process.

Data Analysis
Univariate and multivariate methods have been used in the mathematical-statistical procedures of data processing: multivariate analysis (MANOVA), dis- criminative analysis, coefficient of discrimination (Hadžić, 1989).Application of methods, based on obtained measures, throws a new light on research.The features defining specificity of groups and also those to be excluded from further revealed research by calculating the discrimination coefficient.
Multivariate analysis of variance (MANOVA) is a generalized form of univariate analysis of variance (ANOVA).In order to address the statistical differences between and in the tested groups Wilks' Lambda (χ) and Roy's Largest Root (R) tests were used.The features defining specificity of groups and also those to be excluded from further research have been revealed by calculating the discrimination coefficient (Pallant, 2013).
Discrimination coefficient is a correlation coefficient between the results obtained from a single question and from the whole questionnaire.This parameter may have values between +1 and -1 and is calculated as follows: xy) denotes a sum of the deviation products for the results of single questions and overall questionnaire results, N is a number of responds to a single question obtained from the respondents, sx is a standard deviation of a partial sum for that question and sy is a standard deviation of the overall questionnaire results.Significant differences between the variables are marked with sig.For the value of sig.more 0.05, there is no difference between the observed groups (place of residence, age structure, level of education attained, etc.) while for the values of sig.less than 0.05 there is a significant variation between the observed groups (Pallant, 2013).

Analysis of the first phase of decision makingneed awareness
In this part of the research we aim to confirm or reject the claim of the existence of significant differences between opinions and attitudes of potential tourists with different education level (secondary school education, higher and university degree, or Master of Science degree or PhD) in relation to the need awareness (10 indicators that define the first phase (n)) as a first phase in the decision making process about the tourist destination choice.
A total of 65 variables were used to analyze the first phase -need awareness, which we grouped into seven categories (Table 2).Firstly, MANOVA analysis was conducted to check whether there are significant differences in the phase 1 (need awareness) in relation to the education level.
The results of the MANOVA analysis (F=0.002; p<1.000), indicate that there is no difference between the responses of potential tourists to the questions of the first phase of the decision making process on the tourist destination choice.
Wilks' Lambda and Roy's Largest Root tests were used to test the existence of statistical differences between and in the tested groups.Tests proved statistical significance in the several categories1: The analysis of discrimination coefficient (Table 3) has shown that it is the highest with the presence of ways to spend free time (A1) (0.111), preferred form of tourist travel (A2) (0.103) and sociological factors presence at decision making process about travelling (A5) (0.076).Way of spending free time (A1) is determined with five variables: no answer, active, passive, with entertainment, by computer, other, while the preferred tourist travel forms (A2) are determined by nine variables: wellness, sea, cruises, mountains, travels, village, safari, sport related travels, cultural property, religious tourism.

Analysis of the second phase decision making process -information search
In this part of the research the aim was to confirm or reject the claim of the existence of significant differences between opinions and attitudes of potential tourists having different education level (secondary school education, higher and university degree, or Master of Science degree or PhD) in relation to the information search (11 indicators that define the first phase (n)) as a second phase in the decision making process about the choice of tourist a destination.
Total of 53 variables were used to analyze the second phase -searching for information, which we grouped into 11 categories (Table 4).

Factor Code
Influence of already possessed information about the tourist destination on the intensity of collecting additional information inf1 Influence of satisfaction with services rendered on the intensity of information search about a tourist destination inf2 The most trustworthy sources of information on tourist destination inf3 Marketing source types of highest influence in the process of information search inf4 The types of personal sources that have the greatest influence in the process of gathering information inf5 Types of neutral sources that have the greatest influence in the process of gathering information inf6 Loyalty influence on the intensity gathering information about the tourist destination inf7 Discount influence offers on the intensity of information by searching about tourist destination inf8 The influence of urgent selection and reaching the decision of travelling on the intensity of collecting information about travelling inf9 Ways of collecting information about travel arrangements inf10 Degree of satisfaction with the amount and quality of information about tourist offer inf11 Based on the results of the MANOVA analysis (F=2.217; p<0.00), it is noticeable that there is a significant difference between the responses of potential tourists to the questions of the second phase of the decision making process on the choice of tourist destinations in relation to the level of education.
Values of discrimination coefficient (Table 5) are the highest for: The most trustworthy sources of information on tourist destination (inf3) (which is determined with five variables: marketing sources, personal sources, opinions of family and friends, neutral sources and other), and The types of personal sources that have the greatest influence in the process of search for information (inf5) which is defined by four variables: experience, attitudes; opinions and experiences of the closest people opinion of a respected person.Wilks' Lambda and Roy's Largest Root tests were used to test the existence of statistical differences between and in the tested groups.Tests proved statistical significance in the all factors2.

Analysis of the third phase decision making processestimation of alternatives
In this part of the research the aim was to confirm or reject the claim of the existence of significant differences between opinions and attitudes of potential tourists with different education level (secondary school education, higher and university degree, or Master of Science degree or PhD) in relation to the estimation of alternatives as a third phase in the decision making process about the choice of a tourist destination.
A total of 13 variables were used to analyze the third phase-estimation of alternatives, which are grouped into 2 categories (Table 6).The most significant factors in estimation of tourist destination alternatives E21 The results (F=7.780,p<0.000) show that there is a significant difference between the responses of potential tourists to the questions of the third phase of the decision making process on the tourist destination choices in relation to the education level.
Wilks' Lambda and Roy's Largest Root tests were used to test the existence of statistical differences between and in the tested groups.Tests proved statistical significance3.Discrimination coefficient proved to be the highest with the estimation of alternatives (E21) (0.061) and then with the forms of decision making in the process of tourist destination selection (El1) (0.010).

Analysis of the fourth phase decision making process -purchase of tourist product
In this part of the research the goal was to confirm or reject the claim of the existence of significant differences between opinions and attitudes of potential tourists with different education level (secondary school education, higher and university degree, or Master of Science degree or PhD) in relation to the purchase of a tourist product as a fourth phase in the decision making process about the tourist destination choice.
A total of 32 variables were used to analyze the fourth phase -decision making by purchase of a tourist product, which are grouped into 6 categories.

Factor Code
The influence factor on deciding on the purchase of a particular tourist product P1 The selling point of tourist arrangements P2 The way to purchase tourist arrangements P3 The price and payment option influences on the purchase of a tourist product P4 Purchase preferred form of tourist product P5 Unpredicted situation influences at the selling point on reaching the final decision P6 The results showed that there is a significant difference (F=7.446;p<0.000) between the responses of potential tourists to the questions of the fourth phase of the decision making process on the choice of tourist destinations in relation to the education level.
Wilks' Lambda and Roy's Largest Root tests were used to test the existence of statistical differences between and in the tested groups.Tests proved statistical significance4 Discrimination coefficient analyses (Table 8) has shown that it is the highest with the price and payment options' influence on the purchase of a tourist product (P4) (0.065), influence factor of deciding on the purchase/purchase decision of a particular tourist product (P1) (0.045) and by selling point of tourist arrangements (P2) (0.043).The price and payment options influence on the purchase of a tourist product (P4) and the influence factor on deciding on the purchase of a particular tourist product (P1) in relation to the education level of potential tourists are determined with thirteen variables.Variables used for measuring the price and payment options influence on the purchase of a tourist product are: Not at all, Small, Somewhat, Significant, Large influence.Variables used for measuring the influence factor on deciding on the purchase of a particular tourist product (P1) are: No answer, Personal assessment, Preferred product, Attitudes of others, Unexpected events, Inflation and Supplements, Unstable situation, Other.

Analysis of the fifth phase -purchase evaluation
In this step, the research indented to explore if there are significant differences between opinions and attitudes of potential tourists with different education level (secondary school education, higher and university degree, or Master of Science degree or PhD) in relation to the purchase evaluation as a fifth phase in the decision making process about tourist destination chioce.
A total of 51 variables were used to analyze the fifth phase -post purchase evaluation, which are grouped into 10 categories.

Factor Code
Satisfaction influence on repeated purchase with the same travel agent PE1 Tourist travel frequencies PE2 The amount of money spent on tourist travel from the annual income PE3 Decision making results of tourist travels PE4 Tourism priority forms PE5

Traveling in season PE6
Traveling in the preseason PE7 Traveling in the low season PE8 Satisfaction infulence with the expected service quality PE9 The way of expressing dissatisfaction with the expected level of tourism services PE10 Based on the results of MANOVA analysis (F=10.047;p<0.00), it is noticeable that there is a significant difference between the responses of potential tourists to the questions of the fifth phase of the decision making process on the choice of tourist destinations in relation to education level.
Wilks' Lambda and Roy's Largest Root tests were used to test the existence of statistical differences be-tween and in the tested groups.Tests proved statistical significance5.
The highest discrimination coefficient analyses (Table 10) has the frequency of tourist travels (PE2) (0.074), the priority forms of tourism (PE5) (0.048).Tourist travel frequency (PE2) is determined by the following variables: No answer, Continuous, Occasional, Other.For the analysis of the priority forms of tourism (PE2): Continuous, Occasionally, Other.For the analysis of priority forms of tourism (PE5): Summer holidays at sea, mountain winter holidays, Vacations in ecological environment, Extreme sports; Active holiday in the countryside, Exotic travel, Cruises, Short travels, Other.

Discussion and conclusion
From the practitioner point of view, distinguishing and meeting of potential tourist needs is extremely important.The ability to determine the influence of education level of potential tourists at the final purchasing decisions is an important point for marketing planning and shaping marketing strategies.A possible research suggestion may include the development of marketing programs to attract the attention of both groups of tourists: those with secondary education and those with higher education in order to make their decision-making process easier.
The main hypothesis of this research was that there is a significant difference considering education level of the tourists in all five phases of decision makingprocess when choosing a tourist destination.However, MANOVA results have shown that there are no differences in the first phase of the decision-making process (need awareness, F=0.002, sig.=1.000) regarding education level of the tourists.This means that nevertheless of their educational level, all respondents have similar factors affecting their decisions to travel, as well as similar needs and motives.The difference between them occurs when they go further in the process of thinking how to satisfy those needs and motives.
Additional analysis showed that respondents with secondary education in the first phase -need awareness, preferred rural tourism.The most common reason for tourist travel is seeing new scenery and getting to know new cultures.Respondents with higher education, in the same phase, preferred safari tourism.Getting to know new landscapes and cultures are the most common reasons for tourist travel.In the detailed analysis of psychological factors, pleasure that travel can provide stands out compared to sociological habits and expectations and economic propaganda messages in the media (TV commercials).
For respondents with secondary education, in the second phase -information search, the information they already have does not affect the new information search about the tourist destination.This may mean that they are more open to new information and that it is easier to shape their opinion by media or other means of promotion compared to respondents with higher education.The results show that even the loyalty to travel agency does not affect the intensity of information search.However, they rely on the wordof-mouth promotion, as the most reliable source of information is the family and opinion of friends.The urgent selection has a significant affect while discount offer has a decisive influence, meaning that different promotional offers such as last minute travel offers could be very attractive to this segment of tourists.For respondents with higher education, previously possessed information and satisfaction with services rendered only slightly affect the intensity of the new information search, meaning that they rely on their personal experience and previous knowledge more than people with secondary school.This could be because they are more self-confident due to their longer education and they trust information theiy already have.For respondents with higher education the preferred marketing source is consumer pamphlets, while of personal sources the most valuable are their attitudes and opinions about the tourist destination, which is in the line with previous findings.
In the third phase -the estimation of alternatives, for the respondents with secondary education the decision on the choice of a tourist destination is made in consultation with family and friends.This could mean that they do not rely so much on information they have and their own judgments, but they seek help from the close people.When evaluating alternative tourist destinations the greatest importance is placed on the all inclusive arrangements, courtesy of tourism workers and atmosphere at the selling point which indicates that people with secondary school expect high level of "pampering" at travel destination and high quality experience.On the other hand, in case of the respondents with higher education, the decision on the selection of a tourist destination is difficult to make while assessing all the alternatives.The most important factors of assessment are preferences, expectations and attitudes about tourist destinations meaning they rely on their own judgments.The results also indicate that they perfer individual arrangements, which are tailored to satisfy their specific desires and needs, which differes from people with secondary school who prefer all inclusive packages.Although there were no studies connecting education and decision making, it is interesting to mention that some similar studies found that people who are travelling with family members tend to place more emphasis on facilities and safety than those who are travelling alone (Lai, Graefe, 2000).It is also suggested that people are more willing to take a package tour for their first-time international pleasure travel, and normally place a higher value on word-of-mouth communication than more experienced repeat travellers (Wong, Kwong, 2004).Thus it is interesting for future research to include these variables and to explore their possible mediating effect on relationship between education and decision making.
In the fourth phase-purchase of a tourist product, for the respondents with secondary education, it can be concluded that travel packages are often bought at travel agencies where friends or relatives work.This is in the line with the finding that they rely mostly on their firends and family's opinion while choosing between travel alternatives (also, at purchase all alternatives are evaluated).Price and payment options have a large influence on the purchase of a tourist product, which is also connected with the finding that they opt for travel discounts.Unexpected events have the greatest impact on respondents with higher education.This could be connected with the nature of their job which requires a lot of unexpected travelling.Exclusive agencies are usually chosen and purchasing is done at once, meaning they trust in agencies which already became a brend on the market.Price and payment options do not affect the choice, which may be explained by the fact that people with higher education also earn more money.As they rely on their personal judgments, personal shopping is preferred.
In the fifth phase -post purchase evaluation, for the respondents with secondary education satisfaction has a small influence on the purchase at the same travel agency, meaning they like to experiment with travel agencies and they may opt for those that offer the cheapest itineraries at the moment.They mostly travel in preseason because of less overbooked capacities and in post season due to lower cost of arrangements.Satisfaction with the expected quality of service rendered results in recommendations to friends and acquaintances, as they also rely on their recommendation while deciding where to travel.Dissatisfaction with the quality of service is expressed through angry silence, which indicates that they are not prone to complaining and spreading the negative word of mouth.In case of the respondents with higher education it was found that satisfaction only slightly affects the repeated purchase.They travel regularly and continuously, and spend relatively a lot of money (more than one third of annual income).This may be explained by their higher income compared to people with secondary education, but also travelling may be a part of their job description.What is different compared to people with secondary education, is that their satisfaction with quality of service increases loyalty to the travel agency and result in the repeated purchase, meaning that they prefer quality over price and stay loyal customer even some agencies have cheaper offer.
On balance, the results of this research are of particular importance for improving business in terms of managing the development of new and adaptation of existing tourism products, according to the preferences of potential tourists.Moreover, undarstanding the way that two analyzed groups behave while chosing travel destinations can be of great importance for shaping the marketing activities and travel promotional tools in order to attract each of them.These findings could also help travel agents to attract the loyal customers and shape their communication strategies, which is important for business success of every company on the tourist market.One of the main limitations of the study was a lengthy questionnaire which prolonged the process of collecting the adaquate study sample.Moreover, the survey focused only on people choosing costal regions as tourist destination, so the future reseach could also include diverse destinations such as cities, mountains, spas etc. in order to check if the same results would be obtained.Future research can also analyze how other socio-demographic characteristics such as income, age, marital status affect the decision-making process when choosing a tourist destination.

Figure 1 .
Figure 1.Hypothetical model of research

Table 1 .
Sample characteristics (N= 252) Source: Data obtained by survey research

Table 2 .
Factors of the first phase of decision makingneed awareness

Table 5 .
Values of discrimination coefficient for information search indicator regarding level of education of a respondent

Table 6 .
Factors of the third phase of decision makingestimation of alternatives

Table 7 .
Factors of the forth phase of decision makingpurchase of tourist product

Table 8 .
Values of discrimination coefficient for purchase of a tourist product indicator regarding education level of a respondent

Table 10 .
Values of discrimination coefficient for purchase evaluation indicator regarding education level of a respondent