THE INFLUENCE OF TRAVEL PREFERENCES ON DOMESTIC TOURISM PARTICIPATION BEHAVIOUR IN KENYA: AN ANALYSIS OF TOURISTS AND NON-TOURISTS

Purpose of the study: The purpose of this study was to examine the influence of travel preferences on domestic tourism participation behaviour of domestic tourists by comparing the preferences of both domestic tourists and non-tourists in Nairobi county, Kenya. Methodology: The study adopted descriptive and explanatory research designs. The target population was residents of Nairobi County aged above 18 years. Close-ended questionnaires were used to collect data from 337 domestic tourists in five tourist sites and 339 non-tourists in eight shopping malls within Nairobi. Descriptive analysis, the independent t-test and multiple regression tests were used to analyze the resultant data. Main Findings: The results from the independent t-test showed significant differences between domestic tourists and domestic non-tourists with the tourists displayed stronger preferences than the non-tourists (t =-3.04, df =674, p=0.002). The regression analysis revealed that preferences significantly predicted participation behaviour for both domestic tourists (β=0.287, p<0.001) and domestic non-tourists (β= -0.316, p<0.001). Applications of this study: The findings of this study contribute to the understanding of the influence of tourism preferences on domestic tourism participation from the viewpoint of both tourists and non-tourists. This is key to supporting both product developers and destination marketers to avail of the right products and formulate the right marketing messages. It also serves to extend the debate on domestic tourism non-participation. Novelty/Originality of this study: The study contributes to existing knowledge by clearly bringing out the gap in the current domestic product offering and marketing messages through a comparison of the preferences of domestic tourists and nontourists. This knowledge is required to maintain the existing domestic market (current tourists) and also to tap into the lucrative potential market that is made up of the current non-tourists.


INTRODUCTION
The United Nations World Tourism Organization (2010) defines domestic tourism as activities of persons travelling to and staying in places outside their usual environment but within their country of residence for not more than one year for leisure, business and other personal purposes not related to work or employment. Studies have shown that domestic tourism comes with a host of benefits to the destination. Domestic tourism presents a more predictable and stable demand as it is not prone to seasonality (Ndivo et al., 2012). It, therefore, acts as a cushion against the volatilities of international tourism and is indeed considered to be the backbone of the strong tourism industry for most destinations (World Travel Tourism Council, 2019a). Additionally, domestic tourism accounts for a significant proportion of the tourism industry globally. The UNWTO has approximated that domestic tourism flows account for up to three times the flow of international tourism globally (United Nations World Tourism Organization, 2019). Furthermore, the sector yielded 73% of total travel and tourism spending in 2017 (World Travel Tourism Council, 2019b). It is therefore considered to be the key driver of the tourism sector globally.
Domestic tourism plays a key role in both national and regional economic development (Cheloti, 2011). It promotes balanced regional development and economic growth (Goh et al., 2014) Furthermore, it has been known to ensure the spreading of economic development through visits by metropolitan tourists to both popular and less known areas of the country (World Travel Tourism Council, 2019a). Domestic tourism can, therefore, be said to facilitate the adjusting of regional economic development gaps (Wang & Chen, 2013). Furthermore, domestic tourism contributes to a country's GDP and employment through direct purchases of goods and services by the various sectors that deal directly and indirectly with the tourists (World Travel Tourism Council, 2018). It is, therefore, a major source of government revenue (GOK, 2007). Moreover, domestic tourists tend to avoid prepaid packages and purchase local products and services (Schmallegger et al., 2011;Choo, 2015) hence boosting local entrepreneurship, reducing leakages and maximizing benefits to the locals.
Domestic tourists also have more realistic expectations of local attractions making it easier to meet and satisfy their needs (Fennell, 2015; Kihima, 2015). The level of investment required for domestic tourism is in most cases less than that required for international tourism demand (Goh et al., 2014). It, therefore, acts as a simpler substitute for investors compared to international tourism since it has fewer barriers to overcome. It also results in import substitution, as the country saves on foreign exchange that would have been used if its citizens engaged in outbound tourism (Magableh & Kharabsheh, 2013). Domestic tourism promotes cultural understanding, cohesion, goodwill, national pride, and identity. This serves to reduce intertribal conflicts hence promoting peace and cross-cultural understanding (Mazimhaka, 2007); (Brouder, 2012); (Sheykhi, 2008). Additionally, it also contributes greatly to the conservation of natural resources (Okello et al., 2005) and awareness of cultural heritage (Magableh & Kharabsheh, 2013). Moreover, domestic tourism tends to exhibit fewer negative social impacts on the destination since domestic tourists tend to indulge less in hedonistic and potentially problematic activities /behaviour when at home than when abroad (Canavan, 2012). This is reiterated by (Kihima, 2015) who opined that the Kenyan domestic tourist who travels from Nairobi to Mombasa for hedonism ends up doing the same activities that they would have done in Nairobi.
Despite its obvious significance and lucrative nature, the uptake of domestic tourism in developing countries is still lower than the international threshold (United Nations World Tourism Organization, 2017). Most countries cannot create, increase and sustain local demand for domestic tourism (Magableh & Kharabesh, 2013). Kenya which is the focus of this study has experienced notable improvement in its domestic tourism flows over the past decade. However, it has not been able to attain its target of 6.5 million bed nights despite the various promotional efforts by the Kenya Tourism Board (Government of Kenya, 2018). Additionally, its current performance of 2,025,206 international tourists against 4,955,800 domestic tourists according to statistics from the (Tourism Research Institute, 2019) indicate that the country has not attained the international thresholds of domestic tourism being three times international tourism. This trend raises concern as it depicts a weak domestic segment that cannot adequately cushion the tourism industry against the volatilities of the tumultuous international market.
There have been several visible efforts by Kenya Tourism Board and other stakeholders to promote domestic tourism amongst them being the Tembea Kenya campaign; Twende Tujionee campaign, Twende Ushago campaign, Twende Tujivinjari campaign, Holidays expos, using celebrity bloggers, magical Kenya website, Using mass and social media (A. Munguti, personal communication, August 1, 2016). However, tourism marketing strategies can only be effective if the needs and wants of the tourists are correctly identified (Kamau et al., 2015). These needs and wants are presented as preferences in this paper. Preferences are critical in creating typologies that are used for market segmentation. However, most of these typologies are old and most have gaps in domestic tourism (Horner & Swarbrooke, 2007). Previously, the industry and marketers have wrongly assumed that the Kenyan domestic tourist would enjoy the same products as an international inbound tourist (Okello et al., 2012). Thus, very little research has been done on the preferences of domestic tourists. Kruger & Douglas (2015) reiterated this by stating that little remains known on South Africa's preferences, needs and motivation for domestic tourism. Indeed, as posited by (Moyle et al., 2017) researchers have paid very little attention to measuring visitor preferences in tourism. Previous studies have also ignored the views of the non-tourists who are significant because they form a pool of potential tourists. These gaps provide the impetus for this study as there is a need for research on preferences to inform both product development and marketing.
The study sought to establish the following; i.
The preferences of the domestic tourists ii. The preferences of the domestic non-tourists iii.
Comparison of the preferences of domestic tourists and non-tourists iv.
The relationship between preferences and domestic tourism participation behaviour The study also sought to test the following hypotheses: H 0 1-There is no significant difference in tourism preferences between participants and non-participants of domestic tourism H 0 2-There is no relationship between tourism preferences and participation in domestic tourism

LITERATURE REVIEW
This study is anchored upon the travel decision making process in the tourism destination choice model as proposed by This model posits that the travel decision making process is made up of two main stages namely, "whether to go" and "where to go (Crompton, 1979). Both these stages provide the two main variables for this study, namely tourism participation behaviour, and travel preferences. The first stage deals with the critical concern of whether to participate or not participate in tourism i.e tourism participation behaviour. Those who participate are referred to as tourists while those who do not are referred to as nontourists in this study. The second stage focuses on destination preferences and what to do once at these destinations. Preferences for accommodation, mode of transportation, food and beverage, activities (Kim & Jogaratnam, 2002), as well as souvenirs (Swanson & Horridge, 2006) by tourists, have been subjects of study in the past. They are used to segment the market by grouping the tourists into typologies based on these preferences (Horner & Swarbrooke, 2007). Travel preferences in this paper, are represented by preferences for attractions, activities, and accommodation. The final decision is moderated by other factors such as demographic characteristics, constraints and the motivation for travel which are out of the scope of this paper.

Tourism Participation Behaviour
This refers to the intention to either participate or not participate in domestic tourism. In this paper, those who had participated in domestic tourism were referred to as tourists while those who had not were referred to as non-tourists. Hung & Petrick (2012) defined tourism participants (tourists) to be those people who take a trip for leisure. In contrast, he defined tourism nonparticipants (also known as non-tourists) as those who either do not take part in tourism or drop out from an existing activity or do not take part in a new activity due to existence of a barrier. Literature has also defined non-tourists as potential tourists who have not participated in tourism activities during the last few years This study seeks to establish the influence of preferences on domestic tourism participation behaviour by examining both the relationship between the two variables and also comparing the preferences of domestic tourists and non-tourists.

Preference for Activities and Attractions
Just like international tourists, domestic tourists have been known to exhibit a preference for certain attractions and activities. These preferences have been known to range from low expense activities such as traditional sightseeing to high-end special interest tourism (Canavan, 2012). Okello et al., (2012) noted that Kenyans enjoyed visiting parks for wildlife just like foreigners. This is supported by (Arlt, 2006) who also noted that Chinese domestic tourists were interested in national parks. However, the majority of the tourists did not go back to the parks for revisits and expressed interest in other recreational products that were lacking within the parks. In line with those sentiments, various destinations have come up with specialized, hedonistic tourist experiences such as wellness spas, green tourist resorts and volunteer tourist experiences that cater to specific needs of niche tourist markets (March & Wilkinson, 2009). Domestic tourists have also been known to engage in hiking, visiting museums, backpacking, camping, kayaking and visiting historical sites (Hudson & Ritchie, 2002). Others have been known to prefer historical sites, cruising, spas, skiing and scuba diving (Cai et al., 2002). South African domestic tourists displayed a preference for shopping, social activities including visiting friends and relatives, participating in nightlife, visiting natural attractions and beaches (The National Department of Tourism, 2012).
Ndivo et al., (2012) posited that the most preferred destination for domestic tourism in Kenya was the coast while the least preferred was Northern Kenya. This was attributed to the availability of a variety of activities in the former destination and lack of awareness /security concerns in the latter. Cheloti (2011) noted that domestic tourists visiting the Kenyan coast engaged in reading, sleeping, shopping, relaxing, carrying out business transactions and visiting beauty parlors as recreational activities. Kihima (2015) noted that the Kenyan domestic tourists enjoyed shopping, clubbing and visiting other entertainment spots. They, however, displayed minimal visitation to museums and other cultural sites. Some activities such as bird watching are not considered as being tourist activities by domestic tourists (Canavan, 2012).

Preferences for Accommodation
Domestic tourists have been known to exhibit a preference for low-cost accommodation establishments such as guest houses, campsites, and hostels (Hudson & Ritchie, 2002). Geerts (2017) noted that Chinese domestic tourists preferred budget hotels and short-term rentals such as those found on online homestay booking platforms. Tiwari & Jain, (2009) noted that Indian domestic tourists tended to stick to budget and midrange hotels even when there was an increase in their income. Similarly, (Okello et al., 2012) noted that domestic tourists craved for affordable accommodation facilities. Also, they sought facilities that were culturally acceptable, that served familiar food and drinks and whose staff accorded uniform treatment to all its guests. This means that there is a need to invest in affordable accommodation facilities for domestic tourism.
Several destinations have successfully brought onboard communities to develop affordable accommodation facilities that are still profitable to the community. The development of low-cost housing beach huts by the community in Samoa, for instance, has been a major boost to domestic tourism in the area (Scheyvens, 2007). This allows communities to invest in tourism. Contrary to this, a study in Kenya showed that domestic tourists preferred hotels to camps (Kihima, 2015). A study from the British Isles shows that domestic tourists preferred tents, motorized homes, guest houses, and their second homes as accommodation (Canavan, 2012). The second home concept is an additional residence away from the primary residence usually used for vacations, weekends and get-away. It is rapidly gaining popularity, especially amongst the more affluent populations.

METHODOLOGY
This study targeted both domestic tourists and non-tourists drawn from residents aged eighteen and above residing in Nairobi County, Kenya. The study adopted both descriptive and explanatory research designs. For domestic tourists, systematic random sampling was used to select a total of 337 domestic tourists from five tourist destination areas within the County as illustrated in table 1 below. The domestic tourists were singled out from the international ones through the differentiated entrance rates charged at the destinations.  This was a three-stage sampling that involved selecting constituencies, then wards and finally the shopping centers. According to the Nairobi County Integrated Plan 2013-2018, Nairobi County is made up of 17 constituencies and 85 wards (County government of Nairobi, 2014). The first stage of sampling involved forming the primary unit from the constituencies that make up Nairobi County. The study purposively picked the constituencies that house the tourism destination sites picked for the active domestic tourists. This was done to create a level playing ground for comparing the participants and non-participants. This step yielded two constituencies namely Westlands and Lang'ata. Since the area that is made up of 10 wards in total was still large, the second stage of sampling was applied to select four wards from the two constituencies through simple random sampling. The selection of four out of ten wards was done based on the 30% sample size selection threshold by (Mugenda & Mugenda, 2003). The 10 wards were allocated numbers and random numbers generated using a computer. This yielded a total of four wards namely, Karen, Mugumoini, Parklands/Highridge, and Kangemi, which formed the secondary sampling units.
Since the study population was still large at this point, the third stage of sampling was applied to the shopping centers within the four wards to yield the final sampling units. In each ward, all shopping centers were listed and then two shopping centers were randomly picked using a computer, making a total of 8. This approach is similar to that used by Li et al., (2015) and is consistent with the suggestion by (Veal, 2017). In the listing of the shopping centers, deliberate efforts were made to prepare two lists, one for upmarket and another for modest shopping centers in each ward to include respondents from diverse economic backgrounds. These shopping centers formed the sites for the actual survey. The list of the 8 shopping centers selected is indicated in table 2. The shopping center was deemed an ideal site for this study because it is a meeting point for people from diverse backgrounds. The technique was also used in a study by (Mutinda & Mayaka, 2012) on domestic tourism in Nairobi. The actual survey at the shopping centers was conducted in form of a street intercept as suggested by (Veal, 2017) who proposed the method as an appropriate technique for conducting tourism surveys at shopping centers, malls or on busy streets. This means that the respondents were intercepted as they either got in or left the shopping center (with permission from the management and security of the establishments). Since eligibility for inclusion in the study was dependent on the elimination questions posed at the beginning, all those who had not participated in domestic tourism within the stipulated period at a shopping center were included in the study till the quota for that site was achieved. Additionally, to obtain views of both genders, deliberate efforts were made to issue the questionnaires to respondents of both genders within each site.
A structured questionnaire was used to collect the data from both sets of respondents. The respondents were asked to indicate their preferences for the items on a 5-point Likert scale ranging from 1 to 5 where 1 was not preferred, 2 was least preferred, 3 was fairly preferred, 4 was preferred while 5 was most preferred. Descriptive analyses in the form of means and standard deviation were used to summarize data and provide general trends. Inferential analyses in the form of independent t-test and multiple regression were used to test hypotheses at the level of α = 0.05. Specifically, the independent t-test was used to compare travel preferences between tourists and non-tourists, hence testing the hypothesis on the significant difference between the two groups. On the other hand, multiple regression was used to predict the influence of travel preferences on domestic tourism participation behaviour hence test the relationship between the two variables.

Preferences for attractions
The findings from the study revealed that coastal beaches were a favorite amongst majority of both groups of respondents with the domestic tourists displaying a higher mean (mean= 4.31; SE= 0.059) than the non-tourists (mean= 4.22; SE= 0.059). However, the preferences for the two groups differed for the remaining attractions. Majority of the domestic tourists exhibited preference for game parks (mean = 4.05; SE= 0.059), followed by entertainment facilities (mean= 3.55; SE= 0.066) and then sporting/recreational facilities (mean =3.52; SE=0.083). The least preferred attractions were cultural/historical sites (mean= 3.21; SE=0.078), followed by health and wellness spas (mean = 3.26; SE= 0.078) and spectacular landscapes (mean=3.46; 0.076). In contrast, majority of the non-tourists preferred sporting/ recreational facilities (mean= 3.99; SE= 0.056), followed by entertainment facilities (mean=3.80, SE= 0.067), and then the health and wellness spas (mean= 3.54; SE= 0.067). The least preferred attraction was the game parks (mean= 2.41; SE= 0.053) followed by cultural/historical sites (mean= 2.91; SE= 0.086), and spectacular landscape (mean= 3.27, SE=0.80). These findings are shown in table 3.

Mean difference between preferences of domestic tourists and non-tourists
The results of the independent t-test as shown in table 6 revealed that there was a significant difference in preferences between tourists and non-tourists (t =-3.043, df =674, p=0.002). The tourists (mean=3.56, SE= 0.037) displayed stronger preferences than the non-tourists (mean=3.43, SE= 0.031). The study, therefore, rejected (H 0 1) and concluded that there was a significant difference in travel preferences between domestic tourists and non-tourists. Moving on to the analysis of the specific variables on preferences, the results as displayed in table 8 showed the significant differences between tourist and non-tourist of domestic tourism. All the preference variables displayed significant differences apart from three namely visiting coastal beaches, visiting friends and relatives and exploring cultural/ heritage sites. These displayed equalities of means between the tourist and the non-tourist. Thus, they were preferred equally by the two groups.

Relationship between travel preferences and domestic tourism participation behaviour
The findings from the study showed that preferences significantly predicted domestic tourism participation behaviour for both the tourists (β=0.287, p<0.001) and the non-tourists (β= -0.316, p<0.001). Thus, for preferences, the study rejected the H 0 2 to conclude that there was a significant relationship between preferences and domestic tourism participation behaviour for both tourists and non-tourists. The regression analysis results further indicated that there is a positive and significant relationship between preferences and participation (B=.405; p<0.001; t= 4.323) for tourists. This implies that an increase by 1 unit in preferences leads to a 0.405 increase in participation. The results show that there is a negative and significant relationship between preferences and non-participation (B= -.184; p<0.001; t= -4.143). This implies that an increase by 1 unit in preferences leads to a 0.184 decrease in non-participation.

DISCUSSION / ANALYSIS
From the above-mentioned findings, it is evident that travel preferences significantly influence domestic tourism participation behaviour for both tourists and non-tourists. This is in tandem with findings from a study by Hsu et al., (2009) that concluded that preferences are an important factor that influences the travel decision-making process for potential tourists. Thus, availing of products that cater to these preferences will contribute significantly to increasing participation in domestic tourism. For the tourists, it would result in satisfaction with existing products leading to return visits and also the desire to try out new products/destinations. For the non-tourists, the availability of what they desire may be the "pull" that they require to finally participate in tourism. Furthermore, the results show that travel preferences for domestic tourists and non-tourists are not the same. They differ significantly (t =-3.043, df =674, p=0.002), hence the need to add variety to the existing domestic product offering. This difference may be the reason behind non-participation as the non-tourists may either be unaware or not able to access their preferred choices amongst the readily available product offerings by the industry (Li et al., 2015). Out of the country's two signature products namely the beach and wildlife The World Bank, ( 2010), the coastal beaches remained a favorite for both sets of respondents. However, game parks and game drives registered as the least preferred amongst the nontourists though it was the second favorite amongst the active domestic tourists. The preference amongst the active domestic tourists supports study findings by (Okello et al., 2012). It could be attributed to the current marketing messages which lean strongly towards promoting wildlife tourism.
The reverse trend amongst the non-tourists could be due to the restrictive nature of activities that can be carried out in these protected areas compared to other destinations such as coastal beaches which have a variety of activities (Mutinda & Mayaka, 2012). There is, therefore, a need to integrate the needs of potential tourists into the design of nature-based tourism experiences (Moyle et al., 2017). The non-popularity of game parks as a tourism destination choice for domestic tourism is consistent with findings by Stone & Nyaupane, (2016) who posited that domestic tourists in Botswana did not perceive visiting protected areas as a tourism activity. This is further reinforced by Canavan, (2012) who postulated that some activities such as bird watching were not considered to be tourism activities by domestic tourists. The other top preferences for the tourists included entertainment facilities and sporting/recreational facilities. In terms of activities, the favorites included dancing/clubbing and shopping. Conversely, the non-tourists preferred sporting/ recreational facilities, followed by entertainment facilities and then the health and wellness spas. For activities they preferred sight-seeing, participating in sporting/recreational activities and dancing/clubbing. This reinforces the need for additional product offerings and marketing of the same to lure the potential market that is made up of the non-tourists.
In regards to accommodation, tourists highly favored lodges /resorts followed by self-service establishments such as cottages/apartments and budget hotels. This is in contrast to non-tourists whose most preferred choice was the budget hotels, followed by self-service establishments and then luxury hotels. Thus, both tourists and non-tourists tended to favor low-cost accommodation in the form of budget hotels and self-service establishments. This resonates with the sentiments by (Okello et al., 2012) who opined that domestic tourists yearned for affordable accommodation facilities. Bel et al., (2014) also supported these findings as their study found that domestic tourists prefer campsites and rural cottages. The same sentiments were shared by (Kihima, 2015). However, it was evident from the findings that a portion of both tourists and non-tourist craved high-end accommodation in the form of lodges/resorts and luxury hotels. This could be an opportunity for the industry to provide accommodation as a product offering for fine dining and not only as an accessory for other products. Additionally, since accommodation constitutes a large portion of tourism expenditure Mapelu et al., (2013), these facilities could adopt dual pricing based on season and nationality to make it more affordable to the domestic tourist.

CONCLUSION
From the deliberations above, it is prudent to conclude that travel preferences contribute greatly to the decision to participate in domestic tourism. It is also evident that the current product offering does not appeal wholesomely to the potential market made of non-tourists. The Kenyan domestic tourist seems to be fun-oriented and biased towards activities that require social interaction with others be it, family or friends. Non-participation could be caused by either lack of products that satisfy a given preference or lack of knowledge on where to access such products. There is, therefore, need to: i.
Diversify the product offering to include sports, recreation, entertainment, health, and wellness ii. Diversify the marketing message away from the signature beach and wildlife products targeting the non-tourists iii. Segment the market and introduce niche products based upon identified preferences iv. Incorporate niche activities into the traditional wildlife product to supplement game viewing such as entertainment, socially interactive activities, family-friendly activities

LIMITATION AND FUTURE SCOPE OF WORK
The study has various limitations that could form a basis for future research. Firstly, the study was limited to Nairobi County hence leaving out tourists and non-tourists from other parts of Kenya. Future studies could target a wider scope to incorporate other regions and subsequently other types of products apart from urban tourism products that were captured here. Secondly, the variables selected to investigate the influence of preferences on domestic participation behaviour were limited to three. Future studies can expand this to look at other preferences beyond activities, attractions, and accommodation. The study was also limited to quantitative methods, future studies could take a qualitative approach to unearth deeper meaning and understanding of the variables.