PUSH TRAVEL MOTIVATIONS AS A BASIS FOR SEGMENTATION OF TOURISTS IN EMERGING RURAL TOURISM DESTINATIONS: THE CASE OF CROATIA

Rural tourism has become one of the most popular strategies for rural development, and motivation is one of the most important criteria for identifying market segments. The aim of this paper is to explore the travel motivations of domestic rural tourists in Croatia and to identify a profile of rural tourists based on push travel motivation and tourist behaviour. An online survey was conducted on a convenient sample of 307 respondents. The collected data were analysed using factor analysis followed by cluster analysis. For rural tourists in Croatia, the most important travel motives are seeking relaxation and learning new things. Using a factor–cluster approach to market segmentation, the following four segments were


INTRODUCTION
In many countries, rural tourism has become one of the most popular development strategies not only for rural regions but also for tourism destinations as a whole (Kastenholz et al., 1999;Farmaki, 2012;Rid et al., 2014).Emerging rural tourism destinations usually have good natural and socio-cultural resources that are underutilised.Community leaders and businesses involved in rural tourism in these destinations tend to have less experience in the tourism sector and therefore lack the knowledge and resources for tourism development (Popp & McCole, 2016, Shields & Schibik, 1995).
One of the emerging rural tourism destinations in Europe is Croatia.Since the upsurge of large-scale international tourism, Croatia has been an important tourist destination in the Mediterranean (Orsini & Ostojić, 2018).However, interest in rural tourism has been growing since the 2000s, when national and local authorities became aware of its potential for rural development.Croatia has significant and numerous natural and socio-cultural resources for rural tourism development in all its areas, even 91.6% of the national territory is classified as rural area (Demonja & Ružić, 2010).However, rural tourism in Croatia is still underdeveloped.The contribution of the tourism sector to the national GDP is one of the highest in the EU, amounted to 19.6% in 2018 (before COVID 19).At the same time, rural tourism revenues account for only 1 percent of total tourism revenues (Bartoluci et al., 2018).Croatia's rural areas face serious problems such as depopulation and ageing of the population in rural settlements, which leads to a decline in the labour force.These problems have led to a stagnation of economic development as well as a deterioration of the quality of life in rural areas (Grgić et al., 2010).Rural tourism can help to reduce these problems (Očić et al. 2019;Marzo-Navarro et al., 2015) by diversifying the income of the rural population.In Croatia, there are almost 500 family-run farms that participate in rural tourism.Of these about 80% are excursion farms, i.e. they have tasting rooms and offer only a tour of the property, while 20% have accommodation facilities (Ministry of Tourism, 2018).
In order to develop a high quality tourism product, it is essential to carry out market research to better understand the characteristics of demand, segment the market and select target segments to which tourism products should be adapted.Identifying a well-defined market segment is very important to gain a competitive advantage, as it provides valuable information about customers and allows a destination planner to customize its offer to better meet their needs (Matzler et al., 2004).
Many published studies examined the attitudes of local tourism service providers and residents towards tourism development topics (Akis et al., 1996;Diedrich & Garcia-Buades, 2008;Harun et al., 2018;Peters et al., 2018;Ryu et al., 2020) and fewer studies focused on motive-based segmentation in rural tourism.
Despite the growing demand in rural tourism, to the best of our knowledge there are no published studies on motiv-based segmentation in rural tourism in Croatia.Most of the previous studies on rural tourism in Croatia have mainly focused on analysing the situation and current supply (Lukić, 2002;Ružić, 2012;Demonja, 2014;Galijan et al., 2017;Svržnjak et al., 2017;Čehić et al., 2020), as well as the impact of rural tourism on sustainable development and the local population (Bosnić, 2011;Krajnović et al., 2011;Demonja, 2014;Očić et al., 2019).
Therefore, in order to better understand rural tourism users, the aim of this work is to explore the travel motivations of domestic rural tourists in Croatia and to identify the profile of rural tourists based on their travel push motivations, tourism behaviour and socio-demographic characteristics.The results will provide an information base for targeted marketing in rural tourism.

Tourist motivation
Travel motivation has been the focus of tourism science since the 1960s (Allan and Shavanddasht, 2019;Kim et al., 2007) in order to better understand and predict the factors that influence travel decisions.According to Backman et al. (1995, p. 15), motivation is conceptualized as "a state of need, a condition that serves as a driving force to display different kinds of behaviour toward certain types of activities, developing preferences, arriving at some expected satisfactory outcome ".Previous research has shown that travel motivations can significantly predict behavioral intentions (Oyunchimeg & Gantuya, 2021) and overall satisfaction of domestic tourists as well as their post-visit behavior (Bayih & Singh, 2020).Crompton (1979) noted that travel motivation is influenced by psychological needs that cause a psychological imbalance that can only be corrected through the experience of travel.There are several typologies in the literature of the key motives that encourage people to travel.One of the earliest studies was by author Lundberg (1972), who developed a motivation measurement scale of 18 motivational elements that lead tourists to travel.Crompton (1979) identified nine motives based on in-depth interviews, seven of which he classified as socio-psychological motives or push motives: escape, exploration of self, relaxation, prestige, regression, enhancement of kinship relationships, and facilitation of social interaction.The other two, novelty and education, were characterized as cultural motives or pull motives.
According to Goeldner and Ritchie (2009) the most important motives in tourism are classified into the following categories: • physical, for example relaxation; • cultural, for example discovering new geographical areas; • interpersonal, for example socializing and meeting new people; and • prestige, for example self-esteem and self-actualization.
There are several theories in the literature dealing with motivation in tourism.Some of the most commonly used are the push-pull theory (Dann, 1977), Iso-Ahola's motivation theory (Iso Ahola, 1982;Dunn Ross & Iso-Ahola, 1991) and the Travel Careers Ladder (TCL), which is based on Maslow's pyramid of needs (Pearce, 1988).Iso-Ahola's motivation theory divides motives into personal (vacation, prestige, learning about new cultures, etc.) and interpersonal, (social interactions, meeting locals, etc.) and assumes that consumers in the tourism market balance between these two groups of motives.According to TCL theory, tourism motivation is divided into 5 different motive categories: "physiological needs, safety needs or security needs, love and belonging needs or social needs, self-esteem and self-actualization needs or self -realization needs " (Pearce & Lee, 2005, p. 227).The TCL states that people's travel needs change throughout their lives and that tourists increasingly seek satisfaction of higher level needs as they gain experience.One of the best known and most widely accepted theories in the field of tourism is the push-pull theory (Yuan & McDonald, 1990;Uysal & Hagan, 1993).Push-pull theory explains the reasons why tourists choose one destination over another, the type of experience they want to have and the type of activity they want to engage in (Prayag & Hosany, 2014cited in Albayrak & Caber, 2017).Push motives are those that promote thinking about travel and the desire to travel, while pull motives influence people's decision about where to travel (Allan & Shavanddasht, 2019).

Tourist segmentation and typologies of rural tourist
According to Sarigöllü and Huang (2005), market segmentation is a process of classifying customers into groups based on different needs, characteristics or behaviours.Market segmentation is recognized as a key tool for market positioning and market success (Kompula, 2005;Park & Yoon, 2009;Farmaki, 2012;Rid et al., 2014;Pesonen, 2014Pesonen, , 2015;;Fernández-Hernández et al., 2016;Allan & Shavanddasht, 2019).More specifically, market segmentation is used in tourism research to help tourism service providers evaluate new tourism product opportunities, including rural tourism offerings (Beane & Ennis, 1987).The main characteristics of tourists studied by researchers are geographical characteristics (larger or smaller area in which respondents live), demographic characteristics (gender, age, household size, number of children in the household, stage in the person's life cycle), socio-economic characteristics (occupation, education, property owned by the person, person's income and any factors that define a respondent's social status), psychographic characteristics (lifestyle, interests, activities, needs), and behaviours, including motivations, activities chosen, benefits sought, direct expenditures, use of information sources and travel behaviour (Bigne et al., 2008, cited in Albayrak & Caber, 2018).According to Albayrak and Caber (2018, p. 65), the selected characteristics should be measurable and allow researchers to obtain market segments that include customers with similar characteristics.These segments are often presented as tourism typologies (Garrod, 2008).In order to understand the behaviour of rural tourist and help tourism businesses market their products more effectively, a limited number of studies have been conducted on rural tourism segments based on travel motivation (see Table 1). Table

Source: Authors' own compilation
Previous studies have used different approaches to segment rural tourists based on their travel motivations.Push motivations were used in the study by Park and Yoon (2009), who identified six dimensions of rural tourism motivation: relaxation, socialization, learning, family togetherness, novelty, and excitement.
Using factor and cluster analysis, they identified four market segments, namely: family togetherness seeker, passive tourist, want-it-all seeker, and learning and excitement seeker.Pull motivations, or benefits that tourists expect from a destination, were implemented in the studies by Kastenholz et al. (1999), Frochot (2005) and Rid et al (2014).The studies by Kastenholz et al. (1999) and Frochot (2005) identified four fairly similar segments that differed only in the proportions in the samples studied.Rid et al (2014) used pull motivations to analyse the market potential for rural tourism activities in Gambia.A combination of push and pull approach was used in the studies by Allan and Shavanddasht, (2019), Pesonen (2012), Molera and Albaladejo (2007).

METHODOLOGY
The survey was conducted online on a convenient sample of 307 respondents (domestic rural tourists), aged 18 years and older, who have visited some rural tourism destinations in Croatia at least once in the last three years.The link to the survey was sent to respondents via email and social media.Prior to conducting the survey, pre-test of the questionnaire was performed with 10 respondents to correct any ambiguities in the order and wording of the questions.The survey took between 7 and 10 minutes to complete and was conducted between June and July 2019.
The questionnaire included questions on the frequency of visits to rural tourism destinations, preferences on the type of travel, preferred leisure activities and companions, and expenditure per person per day.The questionnaire was designed based on the literature on travel motivation (Park & Yoon, 2009;Frochot, 2005;Kastenholz et al., 1999).The importance of the motives was measured on a 5-point Likert scale (ranging from 1: not at all important to 5: very important).The questionnaire asked respondents to rate the importance of 24 general travel motives related to trips they had taken in the last three years.The measurement scale for push motivations was taken and adapted from Park and Yoon's (2009, p. 103) study.The last group of questions included socio-demographic characteristics such as gender, age, education, occupation, where they grew up, personal average monthly income and children in the family.
Using a one-way analysis, the frequencies of consumer responses were calculated (via IBM SPSS software, v. 21.0).The importance of tourist's motives was calculated as average values of the Likert-scale scores.Factor and cluster analyses were used to divide the respondents into segments, using the push motives as input variables for the factor analysis.Bartlett's test and the Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) (Hair et al., 1998) were used to determine the adequacy of each intercorrelation matrix for factor analysis.Following this step, a principal component analysis (PCA) was conducted for the importance ratings of the 24 push motivation factors.As suggested by Kaiser (1974), factors with eigenvalues greater than one and variables with factor loadings greater than 0.6 were used for further analysis.Varimax rotation was used to facilitate the interpretation of each factor.The identified factors were used as input variables for Hierarchy cluster analysis.According to Bigne et al. (2008, p. 152), most studies in tourism use hierarchical approaches for segmentation.The factor scores of the push motives were used to groups tourists into segments using Euclidean distance and Ward's aggregation method in the analysis.
The validity of the identified clusters was examined using discriminant analysis.To determine whether there are differences between the clusters in terms of tourists' motives, the Kruskal-Wallis test and Dunn's test were performed.The differences between the segments in terms of socio-demographic characteristics and tourists' travel behaviour were examined using the Chi-square test.

4.
RESULTS AND DISCUSSION

Sample profile
The sample analysis presented in Table 2 showed that 67.9% women and 31% men participated in the study.Most participants were in the age groups of 18 to 29 years (45.6%) and 30 to 45 years (38.8%).Almost half of the sample has a university degree and one fifth has a PhD.More than two-thirds of the respondents are employed and 26% are students.Most respondents (34.4%) have a personal monthly income between 681 and 1 180 EUR.Of the total number of respondents, more than half (60.6%) are childless.Almost three quarters of the respondents (72%) grew up in a city, while 28% grew up in the countryside.

Travel behaviour of rural tourists
The characteristics of the travel behaviour of rural tourists are presented in Table 3.The results show that most of the respondents have visited some rural destinations once (36.8%) and 35.5% two to three times (35.5%) in the last three years.In line with recent trends in tourism, slightly more respondents (57.7%) preferred to visit multiple destinations during a trip, while others (42.3%) preferred single destinations.
Visiting rural tourism is very often a family activity, so the largest share of respondents most often visit rural tourism destinations with family (43.6%) and friends (26.4%).The largest number of respondents (36.2%) spend an average of EUR 15-20 per person per day when visiting a rural tourism destination, while 23.5% of respondents spend between EUR 21-30.In terms of leisure activities, most respondents (40.7%) prefer to spend their free time travelling.The next most preferred leisure activities are sports and games (21.5%), followed by watching TV and sleeping (16.3%), housework (11.7%) and artistic hobbies (9.8%).

Principal components analysis
A PCA was used to determine the basic dimension of the push motivation items.Following the guidelines proposed by Hair et al. (1998), three variables (feel at home away from home; being physically active; personal safety, even when travelling alone) were excluded from further analysis due to low factor loadings and their negative effect on reliability.
After eliminating these variables from the analysis, a new factor solution emerged based on 21 items.Using the Kaiser's criterion of an eigenvalue greater than 1 and varimax rotation, a six-factor solution was found that explained 66.87% of the total variance.The factor loadings of all relevant variables in the rotated factor matrix (except "Not have to rush") were above 0.6 and were clearly associated with only one factor each.The Cronbach's alphas for the six factors ranged from .50 to .83.The Kaiser-Meyer-Olkin (KMO) statistic of 0.829 was The factors identified in the factor analysis are shown in Table 4 and are named as follows: FACTOR 1 -Learning new things explains most of the variance (29.02%) with a reliability coefficient of 0.833.This factor includes five motivation statements (Explore new places, Experience new and different lifestyles, Learn new things, Increase knowledge, Travel to historical heritage sites).
FACTOR 2 -Relaxation explains 10.14% of the total variance and includes five motivation statements related to relaxation, such as get refreshment, escape from a busy job, relax away from the ordinary etc.).
FACTOR 3 -Excitement explains 8.61% of the variance.It includes push motives such as doing exciting things, finding thrills and excitement, having fun away from the city.
FACTOR 4 -Family togetherness includes motives related to family togetherness and explains 7.49% of the variance.
FACTOR 5 -Socialisation identifies the socialisation components of a society.This factor explains 6.84% of the variance and includes four motivation items (Share a familiar place with others, inspire community consciousness, meet people with similar interests, go to places friends haven't been).
FACTOR 6 -Novelty represents seeking new and unusual experiences and includes two motivation items (experience solitude, indulge in luxury).This factor explains the least share of variance (4.76%).
The six factors identified represent specific dimensions of motivation to participate in rural tourism.We can conclude that the motives relaxation (mean=4.37)and learning new things (mean=4.19)are the central motivation for Croatian rural tourists.The results are in line with the findings of previous studies that cite 'relaxation' as the most important motive for rural tourists (Sharpley & Sharpley, 1997;Correia et al. 2007;Chang, 2007;Saayman et al., 2009;Park & Yoon, 2009).
However, in our study, the motive "learning new things" explained the largest proportion of the total variance (29.02%), which is in contrast to the Korean study where it was found that "relaxation" was the factor that explained the largest proportion of the total variance (Park & Yoon, 2009).The results obtained are consistent with the findings of Jang and Wu (2006) who, in a comprehensive review of the literature on push motives, highlighted knowledge seeking, relaxation and family togetherness as the main motives.

Cluster analysis
In the next phase, cluster analysis was carried out and the results are presented in Table 5.The clusters, i.e. the tourist segments, were named as follows: rural explorers and relaxation seekers, excitement seekers, passive tourists and family rural tourists.The discriminant analysis shows that all defined factors had a significant influence (p < 0.05) on the differentiation of the clusters and that 90.6% of the originally grouped respondents were correctly classified, which proves that the classification of the respondents into the segments was not just random.Rural explorers and relaxation seekers (cluster 1, 16.8% of the sample) highly rated the motives of relaxation (mean 4.68) and learning (mean 4.41).This segment of tourists has less need for the socialisation component of the vacation and shows little interest in family togetherness.They are not inclined to experience solitude (mean 2.67) and indulge in luxury (mean 2.16).
Excitement seekers (cluster 2, 29.0% of the sample) place more value on excitement, fun and entertainment compared to the other segments.This segment includes tourists who highly value the motive to learn new things, i.e. the opportunity to explore new places and sites during a tourist trip (mean 4.64).Compared to the other segments, being with the family on holiday and visiting places where the family comes from as well as socialising are the least important for them.Significantly more than the other segments, the excitement seekers want to meet tourists with common interests and visit new places where their friends have not been before.Passive tourists (cluster 3, 22.8% of the sample) include tourists who can be described as disinterested.They represent rural tourists who rate most motives below 4 (important).According to the expressed interest, the most important thing for this segment is relaxation, refreshment (mean 4.36) and relieving daily tension (mean 4.17).
Family rural tourists (cluster 4, 31.4% of the sample) is the largest segment.This segment includes tourists with a wide variety of needs and expectations for a tourist trip.Most motives were rated higher than 4, with the exception of the motives socialisation and novelty.Compared to the other segments, being with the family is most important for family rural tourists (mean 4.62).The opportunity to learn new things, expand knowledge about the destination and relax during the trip proved to be very motivating for them.For this segment, it is also important to socialise, meet people with similar interests and promote community awareness.They tend to experience solitude and indulge in luxuries more than other segments.
The profile of rural tourists in Croatia in terms of their travel behaviour is shown in Table 6.The segments did not differ in terms of frequency of visits to rural tourism destinations and preferred type of travel (p > 0.05).There was a statistically significant difference between the segments in terms of preferred travel companionship.Rural explorers and relaxation seekers were most likely to visit the rural destination with their family (44.0%), as were passive tourists (59.7%) and family rural tourists (50.0%), while excitement seekers preferred to travel with friends (40.0%).In all segments, tourists spend on average between EUR 13 and 20 per person per day.Regarding leisure activities, rural explorers and relaxation seekers (54.9%), excitement seekers and family rural tourists (36.8%) prefer to travel, while passive tourists prefer leisure activities such as sports games, watching TV and sleeping (26.1%).Note: There is no statistically significant difference between the segments in relation to all variables, except for preferred type of travel companions and preferred leisure activities (p <0.05) A clear profile of tourists in terms of socio-demographic and behavioural characteristics was not found.This suggests that these factors do not explain the profiles of the different rural tourists as well as expected.Therefore, research designs on rural tourists in emerging destinations should be further developed, e.g.including theories of consumer culture.Another reason for the insignificant influence of socio-demographic and behavioural characteristics might be related to the underdeveloped and poorly differentiated supply of rural tourism in emerging rural tourism destinations such as Croatia, resulting in insufficiently developed demand.Therefore, a stronger segmentation of consumers by socio-demographic characteristics can be expected in the coming years.

The limitations of the study
This study adds to the sparse literature that deals with tourist motivation, especially in emerging rural tourism destinations.However, there are some limitations in this study based on which recommendations can be made for future research.Some of the limitations of this study are the small sample size, the technique of contacting respondents (online) and the exclusive use of push motives.The advantages of online surveys are a higher response rate, lower cost, real-time access, convenience and the absence of an interviewer.On the other hand, the main disadvantages of online surveys are survey fraud, limited samples and respondent availability.
To improve the representativeness of the sample, future research needs to include a larger sample with a higher proportion of male and older respondents (+40), a face-to-face interviews and a combination of push and pull motives to better understand consumers' motives.In addition, the quality of the results could be improved if the data were collected directly in the destinations and collected over several years.One of the recommendations for future research is to investigate whether there are differences between domestic and foreign tourists in the motives for visiting rural tourism.

CONCLUSION AND RECOMMENDATIONS
Developing rural tourism in emerging rural destinations can be an effective strategy for revitalising rural areas.Considering that rural tourism participants are facing the increasing development, marketing, management and economic difficulties, this paper is a valuable contribution to the marketing response to this problem.A proper knowledge and understanding of tourists' behavior could help rural tourism stakeholders to successfully sell their products by meeting consumers' needs and expectations.Since previous research has shown that travel motivations can significantly predict behavioral intentions and tourist satisfaction, to attract more tourists, destination managers and marketers should consider the key motivational components in tourists' decision-making.In this study, six push motivational factors of rural tourists were found, namely ''relaxation'', ''socialisation'', ''learning'', ''family togetherness'', ''novelty'' and ''excitement'', thus the motives identified in this study support the generalizability of rural tourism motives.In addition, study confirms the appropriateness of push motives for segmenting tourists in emerging rural tourism destinations, such as Croatia.
Study results provided insights into the existence of different rural tourist segments based on travel motives, which enables to better understand rural tourists, as well as the factors that influence their decision making.
In order to improve the tourist offer for each of the four identified segments of rural tourists we suggest recommendations for practice.

The recommendations for practice
Our study found that the most important push motives of domestic rural tourists are relaxation motives, followed by the desire to learn something new.This means that rural tourism managers should consider how to shape the tourist experience by developing new or redesigning the existing tourism products, and how to position their destination as one that offers relaxation and travel-related education through marketing communications.
Using factor and cluster analysis, we identified four segments of rural tourists based on their travel motivations.The largest segment, "Family rural tourists", includes tourists with a wide range of travel needs and expectations.To increase the satisfaction of these tourists, it is important to offer them opportunities for family experiences, including relaxation, education and socialising with other people.However, they also want to experience solitude on their trips more than others.The offer on rural holiday farms needs to be enriched with facilities for children (e.g. an equipped children's playground, farm animals, sports facilities, education about plants, etc.).
The "Excitement seekers", value excitement and fun as the most important push travel motives.
In addition to fun, they also want to learn new things and explore new places and sights.The rural tourism offer for these tourists must be based on facilities such as cultural, fun, and sports events (e.g., cycling and mountain tourism).
The smallest segment, called "Rural Explorers and Recreation Seekers", can be recognised by a more pronounced motivation to relax and learn to new things, but is less inclined to excitement.Since this segment prefers to experience new and different lifestyles and learn about new things, this segment should be offered content such as visiting food and wine themed routes and trails (e.g.wine trails, olive oil trails, pumpkin oil trails, etc.) through which they can learn more about Croatia's gastronomic heritage.A good example of expanding the offer is the organisation of educational and creative workshops, as well as the possibility to participate in various agricultural activities on farms for rural tourism (e.g.grape and olive harvesting).
"Passive tourists" have lower expectations of their trips compared to the other segments.Their main motive for visiting rural areas is relaxation.This target group of tourists can be a good choice for rural areas where a local government plans to develop rural tourism with little investment.Their main advantage to attract tourists is their unspoiled nature and abundance of natural resources.This is ideal for passive tourists who do not have high expectations of their trip, but whose main motive for visiting rural areas is precisely to enjoy the natural environment.Passive tourists prefer leisure activities such as sports, TV and sleeping.Therefore, it is important to provide them with such entertainment facilities in the establishments where they stay.

Table 4
Factor analysis of the motivation of Croatian rural tourists

Table 5
Rural tourist segments based on motivation

Table 6
Profile of rural tourists in Croatia in terms of their travel behaviour (N=307)