The future of youth tourism in Ghana : motives , satisfaction and behavioural intentions

Purpose – The study presents the push–pull motives and behavioural intentions of youth tourists and how these provide the foundation for the planning and development of Ghana’s tourism future. Since youth tourism (YT) is regarded as a niche market globally, the purpose of this paper is to identify the prospects of this form of tourism in a developing nation. Design/methodology/approach – The study has utilised a convenience sampling approach in gathering data from 557 youth tourists, adapting existing instruments for measuring push–pull motives, satisfaction and behavioural intentions. The reliability and validity of the instruments were established through confirmatory factor analysis, exploratory factor analysis and Cronbach’s α analyses. Structural equation modelling is used to establish relationships. Findings – The results revealed the positive effects of push and pull factors on tourists’ satisfaction as well as the significant influence of tourists’ satisfaction on youth behavioural intentions. However, push factors positively influenced pull factors of youth tourists. Research limitations/implications – The conclusion and recommendations of this study might not be congruent with the factors that motivate adults or student tourism, satisfaction and behavioural intentions. Practical implications – The findings of the research validate the viability of YT activities and the behavioural intentions for future tourism market in Ghana. Ghana’s tourism sector should design interesting and competitive offers that attract youth tourists and address tourism growth. Originality/value – To date, investigation into motives, satisfaction and behavioural intentions of youth tourists as the basis for future tourism development remains a virgin field in Ghana. This study has timely attempted to address this gap.


Introduction
It is obvious that youth tourists of today are the global tourists of the future.According to United Nations World Trade Organisation (UNWTO) & Worldwide Youth in Science and Engineering (UNWTO-WYSE, 2011), youth tourism (YT) is expected to increase to 300m travels by 2020 and dominate one-sixth of the global tourists' market (Moisa, 2010).Since the significance of YT was highlighted in the first World Tourism Organisation (WTO) conference in New Delhi in 1991 (Richards and Wilson, 2003), research interest in the field has steadily increased in other parts of the world (Moisa, 2010).Beyond attracting academic interest, YT has also been recognised as a niche market that provides equal opportunities to government and private sector practitioners (UNWTO, 2008).Youth travel is one of the fastest growing and most dynamic markets of the global tourism sector.In 2004, the total value of international youth market was almost $286bn and estimated to increase to 370 m youth travellers, accounting for a total spending over $460m in 2020 (UNWTO, 2016).Moreover, tourism literature provides evidence that YT plays a huge commercial role in the society (e.g.Carr, 1998;Horak and Weber, 2000;Johnson et al., 2006).It is obvious therefore that the sector is destined to be one of the most important target markets that will feed global tourism.However, in Africa, YT is yet to receive commensurate attention.
In the African context, prior studies in tourism by Dieke (2003), Du Rand andHeath (2006), Blackson et al. (2009), Akyeampong (2011), Fourie and Santana-Gallego (2013), Ezeuduji (2013), Amuquando and Asafo-Adjei (2013) and Otoo et al. (2016) have focussed mainly on event tourism, tourism positioning, food tourism, eco-tourism, rural tourism and tourism sustainability.Research in domestic youth tourism (DYT) is, however, still a virgin field in Ghana in particular and Africa as a whole.Domestic tourism aids in increasing tourism investments, reducing the rates of unemployment and developing natural areas suitable for recreation for families and friends (Mustafa, 2012).Ghana is characterised by a wide range of cultural, natural and historical attractions, which include a variety of festivals, mountains, pottery, paintings among others.Currently, the domestic tourism sub-sector accounts for 65 per cent of all tourism activities in Ghana (GLSS 6, 2014).This field is, therefore, a research field worth exploring.However, this study is focussed on DYT, the reason being that the youth account for 50 per cent of the total population.Given such a population, the youth are the key to the future development and sustainability of Ghana's tourism as they are the expected target market for tourism growth.Following the scanty empirical studies into YT on the continent, a study by Dayour (2013) in Ghana suggested future studies into youth travel motives.This study area has also been considered a virgin field which needs the attention of academics and practitioners (UNWTO, 2008).
The definition and categorisation of youth vary across geographical areas with a few overlaps seen in some countries.The UN defines youth as the category of persons between the ages of 15 and 24 years (UNESCO, 2013).According to Moisă (2010), in Romania, Youth Law defines the youth bracket as people between the ages of 14 and 35 years whiles in Malaysia the range is between age of 15 and 40 years (Ministry of Youth and Sports Malaysia, 2006).In Ghana, the youth are classified between the ages of 15 and 35 years (GLSS 6, 2014;National Youth Policy of Ghana, 2010).The Ghanaian definition, with a focus on education and familiarising heritage within the country, therefore serves as the working definition of youth in this study.One of the most appropriate definitions of YT is given by WTO stating that: Youth travel includes all independent trips for periods of less than one year by people aged 16-29 which are motivated partly or fully, by a desire to experience other cultures, build life experiences and/or benefit from formal and informal learning opportunities outside one's usual environment (UNWTO-WYSE, 2011).Youth tourists are typically financially independent and travel unaccompanied or in groups of their peers within and beyond their countries of residence (Dzikiti and Llewellyn, 2016;Swart, 2014;Theuns, 1992).In Richards and Wilson (2003), young tourists were classified into three: backpackers, travellers and tourists with ages ranging between 25 and 30.Independent trip of youth tourists is central to the UNWTO and WYSE (2008).However, in the African context, the concept of independent youth travel is unfamiliar, if not, unwelcomed, because of African cultural values which emphasise communality instead of individuality.Again, most parents are unwilling to let their young travel alone out of concerns about safety at destinations, unfamiliarity and poor access to quality healthcare in various destinations.In addition, young people's inability to earn income by themselves because of lack of jobs coupled with low-income levels of most parents undermine the concept of the independent trip of YT in developing countries as compared to the advanced nations, where youth can embark on independent tour because there is sufficient income.In order to encourage YT therefore, parents, schools and religious bodies should organise heavily subsidised trips.Arguably, the absence of organised and responsible travel and tour firms interested in domestic tourism might also be a contributing factor to the low levels of "independent YT" in the country as the majority of these firms pay most of their attention to international tourism to earn high profit.Finally, the future of YT might still be at stake if service quality (trained tour guides, health facilities and availability of tour information) at destinations, accessible roads, public education and government policies are not improved.Ghana's present and future tourism market rely on the socio-economic values of YT, as young people are becoming adventurous due to increase participation in formal education, global rise of the internet culture and opening up of new destinations.This explains the interest in investigating YT motives, satisfaction and behavioural intentions in Ghana.
A study on DYT in Ghana is critical because of a number of factors.First, the youth number up to 12.5m which constitutes 50 per cent of the total population 25m (Gyampo, 2015;Ghana Statistical Service, 2010).This segment represents an unsatisfied and understudied niche market (UNWTO, 2008).Second, YT is estimated to generate 65bn US dollars towards global tourism receipts (UNWTO, 2010) yet, Ghana has not fully tapped into the multibillion-dollar potential embedded in YT.Third, DYT is instrumental in planning tourism development of Ghana.
According to O' Leary and Deegan (2005, p. 247), tourist motivation is the "combination of the needs and the desires that affect the propensity to travel in a general sense".Various motives have been identified for youth travel across the globe.These include motives such as meeting people, nature, and wellness, adventure, recognition, education, social contact, relaxation and quest for unique experiences (Kihara, 2015;Tibon, 2012).Others are relaxation, seeking friendships, opportunities to put their skills and knowledge to the test, altruistic motives and exploring other cultures (Richards and Wilson, 2003).Identifying these motives will enable destination marketers to better match youth tourists' needs and wants.However, to date, no information has been documented on motives, satisfaction and behavioural intentions of youth tourists in Ghana as well as its viability in the future.Again, there is no known study that converges youth motivation to participate in domestic tourism, the economic and social-cultural importance of YT (UNWTO, 2008) and the benefits gained by local businesses (D'Anjou, 2004).This study will fill these gaps by examining the effect of the push and pull (PP) motivational factors on DYT, the influence of PP factors on satisfaction, and finally, the effect of satisfaction on behavioural intentions regarding the future of DYT.

Youth tourism
Research into YT has been profuse and yet restricted to European, American and Asian contexts.Many works in this field can be cited such as Kim et al. (2006), Moisa (2010), Govender andRogerson (2010), McDonald et al. (2011) and Demeter and Bratucu (2014).Generally, YT has been touted as the future of tourism by many researchers (Cavagnaro and Staffieri, 2015;Richards and Wilson, 2003) and rightly so because tourism is a generational phenomenon.As the next generation, the youth are the pivot of continuity and sustainability.Youth travel behaviour and attitude towards tourism are, therefore, predictive of the future performance of the industry.Many findings have been made relative to YT and these findings have focussed on varying aspects including "meaning" and "motivation" among others.Example, Cavagnaro and Staffieri (2015) found that meaning attributed to travel by students was classified into five components: growth and development, sustainability, socialisation, entertainment and escapism and relaxation, which are comparable with earlier findings of Giri (2004), Richards and Wilson (2003).Govender and Rogerson (2010) also found volunteerism as a key motivator of YT.Cavagnaro and Staffieri (2015) also identified correlations between individual categories of meanings and socio-demographic characteristics of youth, suggesting that understanding the meanings young tourists attach to travel and tourism is complex and multivariable and it necessitates a pluralism of studies across nations.Moreover, as Moisa (2010) found, youth travel is significant for the development of the entire tourism sector.

Push and pull motivational theory
The theory of PP motivations was used as the foundation of understanding behaviour in this study to explain why individuals travel (Uysal and Jurowski, 1994).Research in leisure travel considered push factors as socio-psychological needs that influence a tourist's decision to tour ( Jang et al., 2009), while the pull factors are regarded as those features that attract a tourist to a specific destination once the decision to tour has been concluded (Baniya and Paudel, 2016;Khuong and Ha, 2014;Pearce and Lee, 2005).Undeniably, many types of research have applied the PP motivation theory in explaining reasons for selecting and travelling to a destination choice.Crompton (1979), Dann (1981), Iso-Ahola (1982), Chul et al. (1995), Baloglu and Uysal (1996), Yoon and Usyal (2005), Battour et al. (2014), Dayour (2013) have all argued along the same line that PP motivation theory is applicable in studying the motives of tourists' destination choice.
The theory advocates the behaviour of tourists in two ways: push factors, explained by internal desires to travel and pull factors, explained by a person's decision on destination choice attributes (Crompton, 1979;Yoon and Usyal, 2005) may also be regarded as highly personal (Baloglu and Uysal, 1996;Kim et al., 2003).Furthermore, it is assumed that these two sets of drivers might be both independent and interdependent (Battour et al., 2012).The push-pull motivational theory was, therefore, selected as the theoretical guide for this study.However, note should be taken that this study used historical or cultural attraction, services delivery and ecological heritage as the pull motives whereas knowledge seeking, fun, escape, relaxation, ego-enhancement, novelty and adventure served as the push motives.Marketers in the tourism sector need to understand what motivates youth to select and travel to a destination.The needs and desires of youth tourists could be linked to their satisfaction as well as their behavioural intentions to purchase the same or similar tours in the future.Yoon and Usyal (2005) buttressed this point and concluded that the outcome of research works into tourists' motivation and behaviour must go beyond understanding their needs and wants.

Relationship between push and pull motivators
Prior studies have examined the relationship between PP motives of tourists due to the possibility that one might influence the other.The push factors preceded pull factors (Dann, 1977(Dann, , 1981)).Dann assumed that push factors influenced pull factors, hence causation is established.Furthering this line of support, findings of Chhabra ( 2010), Kim et al. (2003), Klenosky (2002) and Qiao et al. (2008) also confirmed the significant relationship between the factors with one influencing the other.Klenosky (2002) argued that the PP factors relate to two distinct decision points made at two separate points in time, one emphasised on whether to go and the other examined where to go.The following motives of YT were identified in literature: individual background, ethnic origin, family/social context, education and marketing (Boukas, 2013), meeting people and relaxation, nature and wellness, education and quest for unique experience as push motives (Tibon, 2012).In addition to the above, research findings also suggested that the scope of destination product offering and delivery factors were more significant pull factors influencing YT than the natural appeal destination of push factor (Njagi et al., 2017).It is concluded that all the aforementioned studies on PP motives have centred on the general population of tourists visiting a destination yet youth tourists in developing countries has received a limited exposure.Establishing the relationships between these two factors are very important in order to inform marketers about the directions of the motivators be positive or negative.Based on the empirical review, the researchers hypothesise that: H1.Push motivational factors influence pull factors among Ghanaian youth tourists.

Motivation and satisfaction
In the marketing literature, satisfaction is considered as a key concept that guides decision of marketers, customers and consumers.According to Zabkar et al. (2010), Devesa et al. (2010) and Yoon and Usyal (2005), tourist satisfaction is significant to marketing destinations successfully.By deduction, therefore, the level of satisfaction an individual tourist gains are related to how the consumer would be motivated to share or market a destination to other potential customers who might be willing to travel.Early studies have established the link between tourists' satisfaction and destination attributes as well as the improvement of a destination through the quality of products and services and tourists' interests (Yoon andUsyal, 2005, Fang et al., 2008).Qu and Ping (1999) pointed that it is important to have a clear picture of motivation and levels of satisfaction as these elements could behave differently.Subsequent studies (Yoon and Usyal, 2005;Fang et al., 2008;Chi and Qu, 2008;Zabkar et al., 2010;Battour et al., 2012;Battour et al., 2014) have confirmed that PP factors influence travel satisfaction.This research, however, concentrated on the PP motives of youth tourists and its connections to satisfaction as one of the aims of the study.Based on the empirical review, these propositions were formulated and tested: H2.Push motivation factors positively influence youth tourist satisfaction.

H3. Pull motives positively influence youth tourist satisfaction.
PAGE 8 j JOURNAL OF TOURISM FUTURES j VOL. 5 NO. 1 2019

Satisfaction and behavioural intentions
Establishing relationships between tourists' satisfaction and behavioural intentions is essential in selecting a destination.Behavioural intentions have been defined as the outcome of short-and long-terms consequences (Oliver, 1997).The short-term consequences include word-of-mouth, recommendation and complaints while the long-term consequence includes customer loyalty.Another study by Zeithaml et al. (1996) found five types of behavioural intentions which are loyalty to the organisation, willingness to pay more, propensity to switch and both external and internal responses.Satisfaction has been found to be one of the predictors of behavioural intentions (Zeithaml et al., 1996;Baker and Cronmpton, 2000;Prayag, 2009;Chan et al., 2015) and satisfied customers are more likely to purchase the same product or service again (Rust et al., 1995).For instance, Chan et al. (2015) found that behavioural intentions were influenced by tour guide service and tourist satisfaction while Baker and Cronmpton (2000) indicated that a better perception of festival performance increased tourists' satisfaction and influenced loyalty and willingness to pay more.In the light of this, youth tourist satisfaction may influence behavioural intentions in patronising the same destination choice or even recommend destinations to other potential tourists.Therefore, we hypothesise that: H4.Youth tourists satisfaction positively influences behavioural intentions.
The proposed research framework for the study is shown in Figure 1.

Data collection and sampling
A convenience sampling method was used to select the young Ghanaian tourists aged between 15 and 35 years and self-completed questionnaires were administered to them.In total, 25 final-year undergraduate students grouped into five parties with a leader each were trained for the fieldwork and assigned to collect data from youth who were voluntarily ready to participate in the study after they were educated on the purpose of the research.Generally, data were gathered during a three-day conference organised by National Youth Authority of the Ministry of Youth and Employment, at the National Theatre, where large numbers of YT were available.Screening questions were used to select only young Ghanaian tourists that have visited at least one tourist site in Ghana.
The screening questions were also used to avoid double sampling.The sample size for the study was determined by the previous literature reviewed and based on the proposed data analysis technique (Malhotra, 2007).One of the proposed data analysis techniques for this research was confirmatory factor analysis (CFA).Tabachnick and Fidell (2007) indicated that CFA is very sensitive to sample size and less steady when estimated from small samples.Studies of Hair et al. (2010) pointed out that there are no generally accepted criteria for determining a precise sample size using CFA.They suggested that the minimum sample size is 100 when considering models consisting of five or fewer constructs, each with more than three items with communalities above (0.6).Hair et al. (2010) also suggested that 100 is the practical minimum size for using SEM.Finally, Cliff's (1987) sample size recommendation for a 40 variable (item) scale was 150.These suggestions provided the basis for a decision on the sample size for this study.Thus, a usable sample size of 557 representing 92.4 per cent response rate of 603 administered questionnaires was used in this study due for the CFA and SEM applications and the study also utilised 45 items with 10 constructs.In total, 46 questionnaires were discarded due to incomplete answers.

Questionnaire design
To investigate the most important attributes of PP motives, a list of measurement items relating to tourist motivations, satisfaction and behavioural intentions were developed after a thorough review of the prior literature.In total, 15 and 17 items were adapted to measure PP motivations for visiting tourist sites, respectively from prior studies of Al-Haj and Mat Som (2010), Baloglu and Uysal (1996), Battour et al. (2014), Dayour (2013), Tibon (2012) and Yousefi and Marzuki (2015).
Tourist satisfaction was measured using nine items adapted from studies of Prayag (2009) and Qu and Ping (1999).Tourists behavioural intentions were measured with four items adapted from Zeithaml et al. (1996).The instruments adapted from the existing literature showed good indexes of reliability and validity in their earlier studies.Thus, these instruments were considered appropriate for the current study.The final questionnaire developed to measure the four constructs had a total of 45 items and the format used was a five-point Likert scale ranging from strongly disagree (1) to strongly agree (5) and four questions related to the personal information about the tourists.
However, the data collected were subjected to preliminary statistical analyses such as missing data analysis utilising expectation-maximisation algorithm method and test for normality.Next, the data were pooled and Cronbach's α was calculated to check the internal consistency of the instruments as well as computed scores for the composite reliability of the constructs identified.
CFA was employed to validate the instruments and assess whether the specified model fit the data and was appropriate for SEM postulated by Yuan et al. (2005).In addition, the convergent and discriminant validity of the model was tested using average variance extracted (AVE).To measure convergent validity, the AVE should exceed 0.50 as posited by Hair, Anderson, Tatham and Black (1998a) revealing evidence of convergent validity.Discriminant validity was performed to ascertain whether the latent constructs are distinct from one other.Fornell and Larcker (1981) indicated the square of a construct's AVE must be greater than the correlations between the constructs and other constructs in the model.

Demographics
Descriptive statistics were used to describe the participants' personal data.The results from the respondents' demographics showed that the majority of participants were female (51.5 per cent) and the highest age bracket were 15-24 years old (93.5 per cent).Majority (45.5 per cent) of the participants indicated the tours were organised by their individual schools (see Table I).Furthermore, the study also confirmed that majority of the respondents visited waterfalls (11.6 per cent), sea, river, dam, lagoon accounted for 11.4 per cent, zoo (10.5 per cent), national parks (9.9 per cent), forts and castle (9.4 per cent), botanical garden (9.2 per cent), museum (8.6 per cent), historical places (7.0 per cent), palace (4.1 per cent) and slavery market recorded only (1.3 per cent).
Tables II and III show the mean, standard deviations and the composite mean scores of the studied constructs: natural/ecological heritage (M ¼ 4.172), historical/cultural attractions (M ¼ 3.952), Service Delivery (M ¼ 3.833), and accessibility and good value (M ¼ 3.532).The composite mean values revealed that out of the four underlying pull factors, natural/ecological heritage was identified as the most important motive of youth tourists, followed by historical/cultural attractions, service delivery, accessibility and good value.However, the composite mean values of push factors were: rest-relaxation (M ¼ 3.569), knowledge seeking (M ¼ 3.432), novelty (M ¼ 4.049), ego-enhancement (M ¼ 4.159).This showed that ego-enhancement was regarded as the most important push factor of DYT.However, the mean values of composite indicators of satisfaction showed: tourist satisfaction (M ¼ 4.209), satisfaction-connectedness (M ¼ 3.708), satisfaction-restoration (M ¼ 3.415) and behavioural intention (M ¼ 4.157).It implied that tourist satisfaction was considered as the most relevant factor and the respondents have also agreed that they have intentions of visiting other tourist sites as well encouraging friends and relatives do so.

Exploratory factor analysis
Exploratory factor analysis was done using the principal component analysis with the varimax rotation to identify the orthogonal structure of each of the dimensions of the proposed model of PP factors, satisfaction and behavioural intentions.All the three constructs: PP factors and satisfaction were multidimensional (see Tables II-IV ).Factor analyses of 17 items of pull factors resulted in three factors extracted namely: natural/ecological heritage, historical/cultural attractions, service delivery, accessibility and good value.The Kaiser-Meyer-Olkin and Bartlett's test (KMO ¼ 0.766, df ¼ 136, po 0.05) showed the adequacy of sampling and applicability of factor analysis (Hair, Anderson, Tatham and Black, 1998a, b;Hair, Anderson, Tatham and Black, 1998).A total of 60.79 per cent variance was explained where natural/ ecological heritage recorded (21.69 per cent), historical/cultural attractions (19.80 per cent), service delivery (13.37 per cent) and accessibility and good value (5.92 per cent) with communalities above 0.6 indicating that the items were included in the factor extracted (Hair et al., 2010).The factor scores were computed for each of the constructs utilising the regression factor score method due to its maximisation of the validity of the items used (DiStefano et al., 2009).The factor scores were used in the subsequent structural equation model (SEM).However, factor analyses of 15 push factors items resulted in four factors extracted that accounted for a total variance of 60.03 per cent, where rest-relaxation recorded (22.25 per cent), knowledge seeking (18.57per cent), novelty (10.68 per cent) and ego-enhancement (8.53 per cent).The Kaiser-Meyer-Olkin and Bartlett's test (KMO ¼ 0.829, df ¼ 406, po 0.05) showed the adequacy of sampling and applicability of factor analysis (Hair, Anderson, Tatham and Black, 1998a, b;Hair, Anderson, Tatham and Black, 1998).The results showed the Cronbach's α analyses of the items (see Tables II and III) of pull factors ranged between (0.763-0.777), and push factors (0.811-0.831).Again, the Cronbach's α reliability of tourist satisfaction revealed (0.706-0.751) and behavioural intention stood at (0.688-0.879) (see Tables IV and V ).Composite reliability of pull factors recorded between 0.637 to 0.673 push factors presented (0.645-0.658)where tourist satisfaction recorded (0.789-0.666).Indeed, the various Cronbach's α values presented exceeded the recommended value of 0.7 (Hair et al., 2010).Therefore, it is concluded that there is internal consistency among the items employed in the study.In addition, factor analysis was conducted for tourist satisfaction and behavioural intentions.Tourist satisfaction was identified as multidimensional, with three factors extracted namely: tourist satisfaction, satisfaction-connectedness and satisfaction-restoration.The Kaiser-Meyer-Olkin and Bartlett's test (KMO ¼ 0.762, df ¼ 45, po0.05) showed the adequacy of sampling and applicability of factor analysis (Hair, Anderson, Tatham and Black, 1998a, b;Hair, Anderson, Tatham and Black, 1998).The total variance explained by the three factors accounted for 62.19 per cent, where tourist satisfaction represented 25.91 per cent, satisfaction-connectedness 21.18 per cent and satisfaction-restoration 15.11 per cent.The behavioural intention was identified as a unidimensional with only one factor extracted.The Kaiser-Meyer-Olkin and Bartlett's test (KMO ¼ 0.739, df ¼ 6, po0.05) showed the adequacy of sampling and applicability of factor analysis (Hair, Anderson, Tatham and Black, 1998a, b;Hair, Anderson, Tatham and Black, 1998) and a total variance of 62.71 recorded.

Confirmatory factor analysis
A CFA was conducted to validate the constructs where all items loaded above the threshold of 0.5 provided strong evidence of the validity of the constructs proposed for the research model.Anderson and Gerbing (1988) and Bagozzi and Yi (2012) postulated that high factor loadings and high composite reliability scores provide evidence for convergent validity.However, if the AVE of a construct is greater than 0.5, then it shows that there is convergent validity of the constructs (Fornell and Larcker, 1981).The AVEs of the four constructs recorded 0.741, 0.766, 0.796 and 0.792 respectively, and these were all greater than 0.5 (see Tables II-V ).These results established adequate evidence of convergent validity of the constructs used for the study.However, in order to satisfy the requirement of the discriminative validity, the square root of the construct's AVE is greater than their correlations between the constructs shown in Table VI.For example, the square  root of the of the AVEs for two constructs, pull and push factors, are 0.741 and 0.766 in Tables II and III which are more than the correlation between them, 0.321 in Table VI.This shows that the constructs are distinct from one another; hence there was adequate discriminative validity.
The study calculated the inter-factor correlation analysis among each construct of PP motives, tourist satisfaction and behavioural intentions.
Table VI shows the inter-correlation coefficient values which were all below 0.700, ranging from 0.311 (pull motives and tourist behavioural intentions factors) to 0.617 (between push motives and satisfaction factors) and the correlations were significant at.

Structural model results
The results of the overall fit on applying the results of CFI, TLI, RMSEA and SRMR were within the acceptable level.Figure 2 shows the outcome of the proposed model.
The statistics for the model was, root mean squared error of approximation (RMSEA) was 0.001, comparative fit index (CFI) and Tucker-Lewis index (TLI) were 0.945 and 0.924 respectively, and standardized root mean squared residual (SRMR) was 0.001 revealing that the measurement model fit the data and met the cut-off points (Barrett, 2007;Hu and Bentler, 1999;Steiger, 2007).However, the significant χ 2 test is a common outcome in research and was not surprising with regard to sample size (Bagozzi and Yi, 2012).The remaining fit indices (RMSEA, CFI, TLI and SRMR) of the model surpassed the recommended thresholds.The model was, therefore, accepted and regarded as being valid.All of the paths estimated in the model were significant and hypotheses were supported in accordance with the previous literature.

Discussion
The result showed four factors each of both PP motives of DYT in a developing country: natural/ ecological heritage, historical/cultural attractions, service delivery, accessibility and good value, rest/relaxation, knowledge seeking, novelty and ego-enhancement.Natural/ecological heritage and ego-enhancement were identified as the most important pull-push factors for DYT consistent with earlier findings of Kihara (2015) and Tibon (2012).It is shown that youth tourists were pulled by the desire to experience historical and cultural attractions and pushed by the ego-enhancement derived from visiting novel places that they can talk about.
However, the findings of this study showed a compatibility of the effects of push factors on pull factors as demonstrated in the studied framework which confirmed the previous findings of Baloglu and Uysal (1996), Battour et al. (2012), Dann (1977), Dann (1981) and Crompton (1979) and also confirmed the significant relationship between the factors (Chhabra, 2010;Kim et al., 2003;Klenosky, 2002).Thus, it confirmed the push-pull motivational theory as applicable in a developing country.It is observed that the push factors explained the internal desires of the youth to travel while the pull factors relate to their personal decision in choosing a destination site (Baloglu and Uysal, 1996;Kim et al., 2003).
Furthermore, the findings established the positive effects of PP factors on tourist satisfaction as well as the significant influence of tourist satisfaction on youth behavioural intentions presented in the conceptual model and supported by prior studies (Battour et al., 2012;Battour et al., 2014;Chi and Qu, 2008;Fang et al., 2008;Yoon and Usyal, 2005).Zabkar et al. (2010) confirmed that PP factors influence travel satisfaction.Studies of Baker and Cronmpton (2000), Chan et al. (2015), Prayag (2009), andZeithaml et al. (1996) also provided evidence of travel satisfaction.The study also proved that tourist satisfaction will influence the behavioural intentions of young Ghanaian tourists in the future.Given that increased positive satisfaction could affect their future behaviour, the travelling experience of the respondents has practical implications for the future development of DYT since for some of them, this is only the beginning.It is confirmed that first travel experience of youth tourists determines their future travel behaviour (Eusebio and Carnerio, 2015).In effect, it is deduced that satisfaction of the youth tourists, which might also stimulate patronage of the same tour in the future or evoke recommendations of destinations to potentials tourists is one of the explanatory factors of tourists' behavioural intentions.

Conclusion
PP motivations for DYT are among the deserted areas in tourism marketing on the continent.
Hence, this research contributes to the existing knowledge in the area by investigating the convergence of PP motivations, tourists' satisfaction and tourists' behavioural intentions as well as developing a causal model for future tourist behaviour in developing economies from a theoretical framework.Using factor analysis, the study identified four push factors: rest/relaxation, knowledge seeking, novelty and ego-enhancement.The pull factors included accessibility-good value, historical-cultural attractions, natural-ecological heritage and service delivery.The study results demonstrated the importance of natural/ecological heritage attractions and ego-enhancement as PP travel motives of young Ghanaian tourists, which differentiate the result from that of Njagi et al. (2017), where novelty of both places and lifestyle as well as scope of destination products offering and delivering were identified as PP motives of youth tourists in Kenya.Again, it was found out that push factors influenced pull factors and both factors were identified to have positive effects on tourists' satisfaction.Finally, the study's findings proved that tourist satisfaction indeed positively and significantly influenced behavioural intentions which distinguish this research from early study of Chan et al. (2015), where positive effects with insignificant results were examined.

Implications
The result is also consistent with PP motivational theory that advocated that tourists have reasons for selecting and travelling to a destination choice (Crompton, 1979;Dann, 1981;Iso-Ahola, 1982;Chul et al., 1995;Baloglu and Uysal, 1996;Yoon and Usyal, 2005;Battour et al., 2014;Dayour, 2013).One of the theoretical contributions of this paper is that the PP theory provided the foundation for investigating future DYT motives, utilising the proposed research framework within the context of Ghana.Youth tourists' motivations, satisfaction and behavioural intentions are important in managing tourism policies and designing tourism services.It is worth noting that the findings have shown that positive influences could be regarded as the genesis of the future of DYT.Indeed, it is important to align the identified motives of DYT with the current tourism activities and policies.The future development of DYT should be a collective discourse of the tourism practitioners, managers, governmental, youth agencies and other stakeholders together.This research's findings, thus, youth tourists being pulled by the desire to experience historical and cultural attractions and pushed by the ego-enhancement derived from visiting novel places that they can talk about have implications for tourism development.For future purposes, government and other stakeholders will need to find innovative ways of sustaining youth interests in domestic tourism through the maintenance of the existing tour sites and the preservation of the historical and cultural heritage of Ghana.Again, service delivery, accessibility, and good value, rest/relaxation as found by this study as PP factors for DYT require that the tourism industry should maximise efforts at providing excellent and amusing services and resources and the road network connecting tour sites should be more inviting and safe.It is imperative that tour operators and their partners in the industry strive to enhance satisfaction levels of the youth in respect of the PP factors in order to guarantee their (tour operators and other stakeholders') social acceptability and validation.The implications of the above are that travel and tour firms, as well as the tourism boards of developing countries, should find innovative cost-effective ways of enticing young people into independent travels.Again, much education and public assurance are needed to convince parents to allow their wards to make independent trips.
Further, to guarantee the future of tourism, the youth should be the target of tourism services and products.This is because they (youth) will replace the current working class, become future decision-makers and more likely to introduce the cultural values of tourism to their families, relatives, communities and the nation at large.Ghana's tourism sector should, therefore, design interesting and profitable offers that attract youth tourists and address tourism growth and development needs.

Figure 1
Figure 1 Proposed youth tourism model

Figure 2
Figure 2 Results of overall SEM model of youth tourism

Table II
Factor analysis of pull motives of youth tourism Demographic characteristics and destinations visited by the participants Note:The square root of AVE of pull motives ¼ 0.741TableI

Table IV
Factor analysis of youth tourist's satisfaction The square root of AVE of tourist satisfaction ¼ 0.796TableIIIFactor analysis of push motives of youth tourism The square root of AVE of push motives ¼ 0.759 PAGE 12 j JOURNAL OF TOURISM FUTURES j VOL. 5 NO. 1 2019

Table V
Factor analysis of youth behavioural intentions

Table VI
Inter-factor correlations