Choice behavior of tourism destination and travel mode: A case study of local residents in Hangzhou, China

Abstract

The social and economic growth as result of promoting the rapid development of tourism in China has brought tremendous pressure on the urban transportation systems. Research of travel behavior concerning the characteristics of tourists has provided effective information for transportation planning. Due to different city plans, public transportation system design, car parking design and management, etc., the local situation in developed countries differs from the counterpart in China. However, little research has studied the factors influencing the choice of travel destinations in tourism. The research aims to study the tourism destination and mode choice behavior of tourists and provides suggestions to improve tourism transportation service system. An online questionnaire survey is used to collect data including the travel characteristics and personal attributes of local tourists in different holidays in Hangzhou, China. A multinomial logit model is constructed with the trip destination set as the dependent variable. Results show that age, residential type, car ownership, companion type and holiday length have a significant impact on destination choice. To determine what influences modal choice for such trips, a second logit model is established with travel mode set as the dependent variable with the explanatory variables of age, gender, companion type, car ownership, holiday length and travel destination found to be significant. The results demonstrated that people aged 26 to 44 prefer suburban areas, and they are the main group driving to their travel destination. Public transport use frequency decreases when the destination is located outside of the main tourist area. Finally, suggestions have been proposed to mitigate the congestion and parking problem based on model analysis from the perspective of the bus line setting, transfer improvements, and the policy to limit cars, respectively.

Introduction

In transition countries such as China, traveling is no longer a luxury activity. As a basic activity to improve and enrich people's daily leisure and entertainment, it has largely come into the public life as it did in countries who had greater wealth in the 20th century. Along with this expanding leisure activity, the problems of tourism transportation have been drawn more and more attention in China. Western countries have witnessed this problem previously and various responses can be seen. For example, the construction of the Bronx River Parkway in New York in 1923 was a manifestation of the integration of tourism transportation into road traffic planning (Lu, 2009). By the 1990s, tourism transportation planning occupied an important position in the transportation planning of the United States. The German federal government also set up the summer holiday traffic forecast information system with traffic signs around tourist site, starting in 1978 (Huang et al., 2007). With the continuous development of the social economy in China and other such countries, the number of private cars and tourists increase year by year, which indicates the growth in demand for tourism transportation and a higher development requirement of infrastructure. It is difficult to ensure a highly efficient transport service when the basic transport infrastructure supply fails to meet the need demand of tourist traffic. In order to provide advice for the establishment of tourism planning, tourism traffic information platforms, and tourism traffic service systems, it is necessary to investigate, analyze and summarize the behavioral characteristics of tourists from the perspective of travel demand.

In the Case of Hangzhou, it received a total of 184.03 million tourists in 2018, with a year-on-year increase of 13%. The total tourism revenue reached 358.9 billion Yuan ($51.11 billion, with a year-on-year increase of 18%), while foreign exchange revenue from tourism reached $3.83 billion (a year-on-year increase of 8.1%) (2018 Hangzhou Statistical Bulletin on National Economic and Social Development, 2020). As the capital of Zhejiang province and the central city of the Yangtze River delta city cluster, Hangzhou is a major tourist city in China. The increasing number of tourists not only promotes the continuous prosperity of Hangzhou's tourism and culture industry, but also brings great pressure to Hangzhou's transportation system. Studying the travel behaviors of tourists may be an efficient way to help put forward several advices for improving the system.

Considerable research exists on tourists' travel behavior. As for the destination choice, much research focus on international tourism. Factors such as cultural distance (Bi and Lehto, 2018), familiarity (Lee and Tussyadiah, 2012), demographic characteristics (Xiang et al., 2017) and trip expenditure (Guillet et al., 2011) are found to have influence on outbound tourists' destination choice. Keshavarzian et al. (Keshavarzian and Wu, 2020) found that tourists destination choice behavior can be affected when the airline ticket information is presented before the tourism features, by establishing a multinomial logit model based on a Stated Preference (SP) choice experiment. Topics on travel mode decisions have also been chosen by great amount of relevant research. Some scholars found that factors related to the physical environment, such as land use, population density (Ding et al., 2017), travel activity selection (Krygsman et al., 2007) and random regret minimization psychological mechanism (Chorus et al., 2008) exert varying degrees of influence on the decision of travel mode. Mars et al. (2018) stated that limitations, changes in original locations, and new activity locations are determinants for rescheduling travel mode choices (Mars et al., 2018). In 2019, based on Support Vector Machine classification technique, Pirra proposed a new approach to recognize travel mode choice (Pirra and Diana, 2019). Zhang et al. (Zhang et al., 2017) selected three scenic spots in Beijing for questionnaire survey, and used the multinomial logistic model to show the factors including age of the tourists, cars ownerships, monthly income, the number of peers and the type of companions have significant impact on the choice of the tourists' travel mode. For the methodology, Logit model is widely used in the tourism traffic (Guan et al., 2005), especially in the choice behavior of destination and travel mode, which also can be seen in the literature reviewed above.

Although public transport and private cars are often taken as a focus in such research, mode choice of shared mobility has also been scrutinized. Chen et al. (Chen et al., 2017; Chen et al., 2018) ran two intercept surveys including scenic spots to shed light on the characteristics of bike-sharing users and found that leisure is one of the main reasons of that mode's selection. Campbell et al. (Campbell et al., 2016), on the other hand, analyzed the factors influencing the choice of shared bicycles and shared electric bikes, pointing out that e-bikeshare choice is more sensitive to user heterogeneities.

In tourism research, the choice of trip chaining is also an important area of research as people often have a series of trips within a day. Zhao and Guan (2012) used travel attributes and personal attributes as influencing factors to conduct multiple logit modeling of trip chaining choice (Zhao and Guan, 2012). The results show that career, age, travel time, familiarity with scenic spots, local tourists or not and how to arrange travel activities all have significant influence on trip chaining choice. In terms of the decisional order between mode choice and trip chain decisions, Islam et al. (Islam and Habib, 2012) used a 6-week travel diary collected in Thurgau, Switzerland, to unravel the relationship between trip chaining and mode choice, reporting that mode choice decision precedes the trip chaining decision for non-work trips on weekdays. Finally, Yang et al. (2016) compared the relationship of travel mode and trip chaining choices between holidays and weekdays (Yang et al., 2016) and found that in holiday travel people tend to choose the travel mode before the trip chaining which means a whole travel schedule including not only scenic spots but also shop stores and restaurants.

The former research provides solid foundation both theoretically and methodologically. However, most of the research on destination selection based on tourists' motivation preferences, psychological factors (Yoo et al., 2018), marketing, business-related added value and tourism management. Little research has taken tourists' travel behavior as a starting point, such as who to travel with, when to travel and how to travel, which may provide a lot of useful information for tourism transportation system improvement. In previous studies, domestic research on travel behavior generally focused on the choice of travel mode and trip chain. The topic of destination choice still needs to be explored greatly. In addition, the local situation in developed countries differs from those in transition countries, such as China. So, it is necessary to scrutinize through the specific travel behavior and mode choice in the field of tourism in Chinese cities.

This study aims at investigating choice behavior of tourism destination and travel mode, summarizing and analyzing the travel characteristics of tourists, identifying the factors influencing the choices and establishing the relationship among personal attributes, travel attributes and choice behavior. Suggestions could be made to improve tourism transport services according to useful information provided by the research results and findings. The site of the research is additionally meaningful because most of the previous research in transition countries has focused on mega-cities such as Beijing, the Chinese capital. Research on smaller centers is needed as mega-cities are an exception. In this research, the types of holidays are taken into consideration to provide basic information for different transportation policies in different periods of time.

The remainder of this paper is organized as follows: the following section explains the online survey design and sample description; the third and fourth sections introduce the multinomial logistic models for destination choice and for travel mode respectively. The final part of the article contains the discussion and conclusion.