Consumer behaviour: the influence of age and family structure on the choice of activities in a tourist destination

Changes in the structure of consumers’ (tourists’) needs that shape tourism in accordance with the modern way of life attribute greater importance to different types of consumers/tourists. Understanding consumer behaviour (tourists as consumers) is significant in terms of tourism destination management. This paper presents the results of a research conducted with the aim of identifying activities (visiting cultural and historical attractions, sport and recreation, health, fun, gastronomy, shopping, cultural entertainment events) that consumers/tourists prefer in a tourist destination depending on their age and family structure. Two hypotheses were set in this paper: H1: There is a statistically significant difference between the age of tourists and the activity preferences in a tourist destination; and H2: There is a statistically significant difference between the family structure of tourists and the activity preferences in a tourist destination. The sample included 1117 respondents from the area of the Autonomous Province of Vojvodina. A survey was conducted from January to March 2017. In accordance with the defined hypotheses, descriptive statistics and a statistical test of multivariate analysis of variance (MANOVA) were used. ARTICLE HISTORY Received 10 October 2017 Accepted 10 December 2018


Introduction
Numerous changes characterize modern tourism. Quantitatively, the changes are reflected in the increasing number of participants (tourists) in international tourism who, according to the official data of the World Tourism Organization, reached a total of 1322 million worldwide in 2017 (World Tourism Organization UNWTO). Representing an increase of close to 7%, this marks the eigth consecutive year of above-average growth following the 2009 global economic crisis. Besides, over the past few decades the tourist industry has been regarded as the key driver of growth in national economies, in both developed and developing countries (Mari c, studies are tourism activities (Mumuni & Mansour, 2014). Like in other studies (Lehto, Fu, Li, & Zhou, 2017), the purpose of this paper was to improve the understanding of tourists from Vojvodina by linking 'what do they want?' to 'what do they do?' at tourist destinations.

Literature review
Changes in the structure of consumers'/tourists' needs shape tourism in accordance with the modern way of life and the values that it creates. Socio-demographic changes, such as higher life expectancy, healthier ageing, smaller family size and higher level of education, impact the development of tourism. These changes represent external factors that shape tourism demand, affecting the decisions to participate in tourist activities, the type of destination chosen, the length of a vacation, etc. (Glover & Prideaux, 2009). Furthermore, the number of activities undertaken during the vacation has not been reduced at the same rate as the duration of the vacation which leads to a conclusion that approximately the same number of activities undertaken during the previously longer vacation is nowadays fitted into one week or even a weekend (Poon, 2003). At the same time, activities are evaluated by tourists according to their ability to satisfy needs. Tourism literature points out the importance of 'push' and 'pull' factors in choosing vacation destinations and activities undertaken during the vacation. In this way 'push' factors refer to intrinsic desires of the tourists while 'pull' factors are related to the attractiveness of a destination (including activities offered in destination) (Kozak, 2002).
Modern trends in the global tourism market attribute greater importance to different types of consumers/tourists who expect different values. In order to create a value, especially a superior one, a good knowledge of the behaviour of the various tourist segments is a necessity ( -Dordevi c & Ze cevi c, 2015). Understanding the behaviour of consumers/tourists is significant in terms of tourism destination management. Studies that have examined consumer behaviour in tourism suggest that it is vital for destination managers to develop a better understanding of specific segments of consumers in order to accommodate their distinct needs and establish efficient and effective marketing and management strategies (Kim, Wei, & Ruys, 2003;Mykletun, Crotts, & Mykletun, 2001). To this end activity-based segmentation is a solution for developing tourist segments across different nations and cultures (Mumuni & Mansour, 2014;Choi, Murray, & Kwan, 2011;Beritelli & Boksberger, 2005). According to Pesonen (2015) activity-based segmentation is most useful for guiding destination management and marketing and according to Eusebio, Carneiro, Kastenholz, Figueiredo, and Soares da Silva (2017) it is an action-relevant destination marketing tool. Asan and Emeksiz (2018) used it as a method based on the activity in which the tourist prefers to participate. In this sense tourist destination managers must understand preferences for activities and actual participation of consumers in them. When tourists visit a destination, they often eat somewhere outside the hotel enjoying local gastronomy, visit cultural or historical places, shop, visit fun or cultural entertainment events and participate in some sports/recreation/health activities. These activities make a contribution to the overall tourist experiences with their vacations at the destination (Kozak, 2002).
Market researchers often group consumers into generations or other socio-economic groups in order to create a general picture of the market segment. A cohort analysis is useful in the tourism industry because different cohorts are attracted to different tourist activities (Bojani c, 2011). This is important since understanding these values is necessary if one wants to meet the needs, aspirations and desires of consumers/tourists (Nordin, 2005). The segmentation of consumers into generations is based on the assumption that they share common lifestyle characteristics and a common value system. According to Swarbrooke and Horner (2006), segments in the tourism industry are often based on demographic criteria. These authors seem to assume that young people look for parties, elderly people prefer sedate activities, while parents are preoccupied with the need to keep their children happy during their vacation. Thus, for the members of the 'Baby Boom' generation (born between 1943 and 1959) vacations are seen as a way to maintain health and interrupt daily routines. According to Nimrod (2008), leisure travel appears to be taking a central role in the retirement life of seniors. Members of Generation X (born between 1960 and 1979) enjoy adventure trips. Members of Generation Y (born between 1980 and 2000) extensively use modern technology when choosing a tourist destination, and that is why this generation is characterized as the 'next generation', having a relatively high discretionary income and travelling frequently (Leask, Fyall, & Barron, 2013;Bilgihan, Okumus, & Cobanoglu, 2013). Members of this generation are gradually replacing the Baby Boomer and Generation X consumers in the workforce and are becoming the main source of visitors of some tourist destinations (Benckendorff, Moscardo, & Pendergast, 2009). The youngest segment of consumers/tourists members of Generation M (born after 2001) are directly influencing the decision-making process of the members of Generation X while choosing a tourist destination (Bartoluci, 2013;Nordin, 2005).
According to Kozak (2010), different members of one household are typically jointly involved in travel decisions with the specific dynamics due to power relations among family members. The size and composition of the family affect household preferences regarding the destination choice and participation in activities during a vacation. Smaller families with no children are more inclined to travel than large families which have more economic and physical constraints (Bernini & Cracolici, 2015). Based on the degree of interactivity, activities shared with family members are either parallel or joint (Lehto et al., 2017). Parallel activities require minimal interaction (hunting, listening to music, visiting a museum) while joint activities require significant interaction among family members (playing games, visiting friends, camping). Active, creative and goal-oriented activities enhance family interactivity more effectively (Lehto, Lin, Chen, & Choi, 2012).
According to a survey conducted (between 24 and 30 January 2013.) by the TNS Political & Social network (with 30,628 respondents from different social and demographic groups) in 27 Member States of the European Union and Croatia, Turkey, the Former Yugoslav Republic of Macedonia, Iceland, Norway, Serbia and Israel, respondents aged 55þ are most likely to mention wellness/spa/health treatments, nature and culture as main reasons for visiting a tourist destination. On the other hand, 25-54-year-olds prefer the sun/the beach, while 15-24 year-olds are most likely to say that a specific event is the main reason for visiting a tourist destination. (Flash Eurobarometer,370). Authors Bel, Lacroix, Lyser, Rambonilaza, and Turpin (2015) identified different groups of tourists according to their age and family structure. Using activity-based segmentation 'water-based activities' were found to suit families with children, 'outdoor pursuits' suit adults while 'natural and cultural heritage discovery' and 'gastronomy' were found to suit adults over 50. In their study authors Mumuni and Mansour (2014) have identified three clusters: (1) conservative cluster (knowledge-seeking activitiesvisiting historical and cultural attractions); (2) cluster of fun seekers (entertainmentnightclubs, music concerts, movies, shopping, amusement parks, beaches); and (3) cluster of variety seekers (knowledge-seeking activities and entertainment). Depending on their age and family structure, the conservative cluster is dominated by relatively older and married respondents while the cluster of fun seekers is young and single respondents.

Research methodology
The aim of this study was to identify activities that tourists prefer during their vacation depending on their age and family structure (as socio-demographic determinants of respondents). For this purpose, a convenience sample of 1117 respondents from the area of the Autonomous Province of Vojvodina was used. According to Mumuni and Mansour (2014), convenience sampling is widely used in tourism studies. Likewise, people who live in a certain region often share the same values and have similar needs which differ from the needs of people who live in other regions ( -Deri, Armenski, Te sanovi c, Bradi c, & Vukosav, 2014). The characteristics of respondents that participated in the research are presented in Table 1.
What is evident from Table 1 is that most of respondents were female respondents (55.7%). Most of respondents were within the age group of 45þ (35.1%). The average age of respondents was 38.02 years (SD ¼ 14,487, range 16-84), of which the average age of male respondents was 38.22 years (SD ¼ 14,462, range 16-84), and the average age of female respondents 37,86 years (SD ¼ 14,515, range 17-84). Besides, most respondents (54.9%) have completed secondary school. Within the family structure, the majority of respondents was living in a family with two or more children (36.5%). Face-to-face and web survey techniques were used in order to collect data. The survey was conducted from January to March 2017, including the pilot test. The questionnaire was composed of three parts: (1) questions related to basic sociodemographic characteristics of respondents; (2) questions related to respondents' general vacation preferences; (3) questions related to preferred activities during a vacation. The list of activities was compiled from a review of the activity-based segmentation literature (Mumuni & Mansour, 2014;Lehto, O'Leary, & Morrison, 2004). In order to indicate how necessary each activity is a 7-point response scale was used (1-most; 7-least).
Descriptive statistics of activity preferences in a tourist destination by age of tourists is presented in Table 2. The analysis was conducted on a reduced sample of 1067 respondents, since 50 respondents did not provide answers to the questions relevant to the analysis.
Descriptive statistics of activity preferences in a tourist destination by family structure of tourists is presented in Table 3. The analysis was conducted on a reduced sample of 1045 respondents, since 72 respondents did not provide answers to the questions relevant to the analysis.
Starting from the basic subject and issues as well as the research goal of this study, and taking into account recent scientific research on this topic, two hypotheses were tested: H 1 : There is a statistically significant difference between the age of tourists and the activity preferences in a tourist destination. H 2 : There is a statistically significant difference between the family structure of tourists and the activity preferences in a tourist destination.
In accordance with the defined hypotheses, statistical test of multivariate analysis of variance (MANOVA) was used. In the first part of the research, MANOVA was carried out in order to explore the differences between the age of tourists and activity preferences in the tourist destination. In the second part of the research, MANOVA was carried out in order to explore the differences between the family structure of tourists and activity preferences in the tourist destination. Statistical package SPSS IBM Statistics Version 20 was used for data processing.

Testing the hypothesis 1
As defined in the research plan, the first hypothesis testing was enabled by applying a statistical test of multivariate analysis of variance (MANOVA). The test of the assumption of homogeneity of covariance matrices was carried out before the application of MANOVA. This test was performed by applying the Box test in accordance with the SPSS procedure. It was observed that the data passed the test of homogeneity of covariance matrices (Box test ¼ 0.044 > 0.001). The results of MANOVA test are shown in Table 4: One-way multivariate analysis of variance was used in order to explore the differences between the age of tourists (16-29, 30-44, 45þ) and activity preferences in the tourist destination. Seven dependent variables were used as activities in the tourist destination: visiting cultural and historical attractions, sport and recreation, health, fun, gastronomy, shopping, cultural entertainment events. The independent variable was the age of tourists. The preliminary examination verified the assumptions of normality, linearity, univariate and multivariate atypical points, homogeneity of variance-covariance matrix, and multicollinearity (Green & Salking, 2014). Violation of assumptions were not detected. There was a statistically significant difference between the age of tourists and the combination of the dependent variables, F (14, 2116) ¼ 18.00, p ¼ .000 < 0.05; Wilks' lambda ¼ 0.798; partial eta squared (the proportion of variance in the dependent variable) ¼ 0.106 (large impact) (Cohen, 1988, 284-287) (Table 5).
A review of the average value of the results showed that tourists aged 45þ have a stronger preference for activities related to health (M ¼ 5.18, SD ¼ 1.901) compared to tourists aged 29 or under (M ¼ 3.45, SD ¼ 1.832). Tourists aged 29 and under showed a stronger preference for fun (M ¼ 5.01, SD ¼ 2.110) compared to tourists aged 45þ (M ¼ 3.25, SD ¼ 2.124). Tourists aged 45þ expressed a stronger preference for gastronomy (M ¼ 4.00, SD ¼ 1.780) compared to tourists aged 29 or under (M ¼ 3.49, SD ¼ 1.825). Tourists aged 29 or under showed a stronger preference for shopping (M ¼ 3.93, SD ¼ 1.880) compared to tourists aged 45þ (M ¼ 3.23, SD ¼ 1.849).
Therefore, it can be concluded that the hypothesis H 1 is confirmed: there is a statistically significant difference between the age of tourists and activity preferences in the tourist destination.

Testing the hypothesis 2
As defined in the research plan, the second hypothesis testing was enabled by applying a statistical test of multivariate analysis of variance (MANOVA). The test of the assumption of homogeneity of covariance matrices was carried out before the  Table 6: One-way multivariate analysis of variance was used to explore the differences between the family structure of tourists (single, a family without children, a family with one child, a family with two or more children) and activity preferences in a tourist destination. Seven dependent variables were used as activities in the tourist destination: visiting cultural and historical attractions, sport and recreation, health, fun, gastronomy, shopping, cultural entertainment events. The independent variable was the family structure. The preliminary examination verified the assumptions of normality, linearity, univariate and multivariate atypical points, homogeneity of variance-covariance matrix, and multicollinearity (Green & Salking, 2014). Violation of assumptions were not detected. There was a statistically significant difference between the family structure of tourists and the combination of the dependent variables, F (21, 2972.5) ¼ 6.271, p ¼ .000 < 0.05; Wilks' lambda ¼ 0.883; partial eta squared (the proportion of variance in the dependent variable) ¼ 0.041 (medium impact) (Cohen, 1988) (Table 7).
A review of the average value of the results showed that families with two or more children have a stronger preference for health (M ¼ 4.83, SD ¼ 1.962) compared to single tourists (M ¼ 3.58, SD ¼ 1.903). Single tourists (M ¼ 4.77, SD ¼ 2.221) and families without children (M ¼ 4.38, SD ¼ 2.059) expressed a stronger preference for fun compared to families with one child (M ¼ 3.50, SD ¼ 2.164) and families with two or more children (M ¼ 3.65, SD ¼ 2.226).
Therefore, it can be concluded that the hypothesis H 2 is confirmed: there is a statistically significant difference between the family structure of tourists and activity preferences in the tourist destination.
Finally, in terms of seven vacation activities used in this study (visiting cultural and historical attractions, sport and recreation, health, fun, gastronomy, shopping, cultural entertainment events) authors conclude that older, married respondents with two or more children prefer health among all activities. On the other hand, younger, single respondents without children prefer fun among seven vacation activities. These  results correspond with earlier study of Swarbrooke and Horner (2006) who assumed that young people look for parties, elderly people prefer sedate activities. Research results also show that tourists aged 45þthe silver market (Branchik, 2010) have a stronger preference for activities related to health and gastronomy compared to younger tourists whose preferences are fun and shopping. Furthermore, it was observed that families with two or more children prefer health, while single tourists and families without children prefer fun during their vacation. These results correspond to the Gallup typology of tourists that distinguishes, among others, tourists traveling for hedonism, tourists traveling for medical reasons and tourists traveling for active leisure ( -Dordevi c & Ze cevi c, 2015). Furthermore, the obtained results are consistent with the results of the research conducted by the TNS Political & Social network, in which respondents aged 55þ are most likely to mention wellness/spa/ health treatments, nature and culture as main reasons for visiting a tourist destination. On the other hand, 25-54 year-olds prefer the sun/the beach.

Conclusion
Modern tourism market is characterized by the increasing number of tourist destinations and a competitive battle for consumers/tourists. The possibility of substituting one tourist destination with another is very high and therefore tourist destination managers are trying to diversify their offer by developing tourism products tailored to suit different segments of consumers/tourists (youth, elderly, singles, families). According to forecasts, more than half of the population in the developed countries will be 50þ by the year 2040, and it will result in changes of marketing and management strategies of tourist destinations (Smith, 2001). Since many tourist destinations rely on the activities taken by consumers/tourists during their vacation, it is of prime interest for destination managers to gain more knowledge about each segment (Karim & Geng-Qing Chi, 2010). It also seems clear that, in order to remain competitive, tourist destination managers have to design and implement marketing and management strategies in order to achieve the desired product positioning in target markets (Hawkes & Kwortnik, 2006). In future, the focus on the 'I', as a particular person, will increase. Consumers/tourists will seek 'custom' solutions tailored to correspond to their personalities (Popesku, 2011). Therefore, the main task for the tourist destination managers will be an attempt to link the needs of consumers/tourists and megatrends that appear on the market. In accordance with the results of the research, one of the megatrends that is necessary to pay attention to is a favorable development trend of wellness and spa tourism products. On the other hand, the gastronomic offeras an integral part of tourism experienceoften plays a vital role when choosing an adequate tourist destination (Ryu & Jang, 2006).
The main purpose of this study was to improve the understanding of tourists from Vojvodina by identifying activities that tourists prefer during their vacation. The data analysis identified different segments of consumers/tourists to whom it is necessary to adjust an offer of tourist destination in order to reduce the risk that is constantly present during marketing and management decision-making. In other words it can be concluded that the outbound tourist market of Vojvodina is not one homogeneous market that can be reached with only one offering and promotion.
Although there were some important findings in this study, there were some limitations as well. First, the sample included only respondents (tourists) from Vojvodina. According to Kozak (2002) undertaken tourist activities may vary according to countries of origin. The second limitation is related to the selection of destination. Namely, there is a lack of data about specific destinations visited by respondents. Finally, the time of travel is also uncertain because the sample did not show when (what season) respondents were travelling to the destination.
This study provided a first look at the outbound tourist market of Vojvodina. A future step could be to expand the sample with respondents from other countries in the region. Also, future research might be related to the specific destination or specific activity undertaken in it.
Adapting to a new system of market values and changes in consumer/tourist preferences causes an abandonment of the traditional offer. In this sense, segmentation is important because it provides the actors in the value chain with the opportunity to identify consumer/tourist segments. Furthermore, this allows them to create and deliver a value as a precondition for achieving a competitive advantage and making profit.