Examining Attributes Associated with Tourist Arrivals to Forest Parks through Linear and Curve Estimations

Advances in Hospitality and Leisure

ISBN: 978-1-78769-304-3, eISBN: 978-1-78769-303-6

ISSN: 1745-3542

Publication date: 5 December 2018

Abstract

This research constructs the critical predictors of visitation that shall allow the practitioners to foresee the visitation in the years to come through secondary data. For this study, tourist arrival data associated with the most popular forest park (i.e., Xiton Forest Park) in Taiwan along with relevant socio-economic data are utilized. This research adopts a group of analytical procedures involving correlation analysis, regression, and curve estimation analyses. The results show that the number of holiday per month and the average monthly rainfall have positive and negative correlations, respectively, with the visitation. Meanwhile, average monthly temperature and monthly gross domestic product per capita show a positive correlation in all three analytical methods and therefore are regarded as the primary predictors of tourist arrival. Consequently, this study provides managerial implications to increase the tourist arrivals to the forest park.

Keywords

Citation

Liu, W. (2018), "Examining Attributes Associated with Tourist Arrivals to Forest Parks through Linear and Curve Estimations", Advances in Hospitality and Leisure (Advances in Hospitality and Leisure, Vol. 14), Emerald Publishing Limited, pp. 93-109. https://doi.org/10.1108/S1745-354220180000014006

Download as .RIS

Publisher

:

Emerald Publishing Limited

Copyright © 2019 Emerald Publishing Limited


Introduction

Like any other developing countries, while transforming to a newly industrialized economy, Taiwan has experienced an escalating demand for goods and services associated with the travel business, along with the inexorable threat to the health of ecology and natural environment. Consequently, a variety of initiatives and movements toward ecological conservation have been initiated. In the context of leisure pursuits, ecotourism as a form of nature-based travel with consideration for conservation, for example, has flourished (Newsome, Moore, & Dowling, 2013). In ecotourism settings, the tourists are able to reinforce their cognition and behavior toward the conservation of ecology as well as become viable partners in contribution to sustainable development (Chiu & Chih, 2012; Lee & Huang, 2006).

What distinguishes ecotourism from other forms of nature-based tourism is that it bears critical agendas in support of conservation efforts. Hence, ecotourism attractions aim to furnish recreational opportunities to tourists to engage with nature while promoting the concept of environmental protection. There are various types of attractions that stage ecotourism experiences; for example, in Taiwan where forest land accounts for 60% of total landscape, a handful of forests are designated for conservation which are referred as forest parks and have been utilized as venues for outdoor recreation and environmental education and in the promotion of ecotourism.

Although the ecotourism has been evoked for decades, investigative efforts have largely centered on the attitudinal and behavioral issues associated with ecotourists through both inductive and deductive methods of inquiry by using primary survey data. Thanks to the availability of open-source data collected by government agencies, nonprofit agencies, and other parties, nowadays researchers are able to quickly access this data to analyze critical issues of interest. Focusing on the benefits of open-source data, this study is in an attempt to take an innovative path to glean issues of significance impacting management strategy and service delivery at ecotourism destinations. That is to utilize analytical schemes of linear and curve estimations to explore secondary data sources and evaluating the attributes/indicators explaining the total tourist arrivals to ecotourism destinations via a case study of the most visited forest park in Taiwan. The contribution of the present study is twofold. First, it presents a methodological approach to highlight possible factors predicting total tourist arrivals to a nature-based destination. Second, it supplies managerial implications to the forest park under study that may be used as lessons for other destinations with similar characteristics.

Concerning the predicting attributes of the tourist arrival to forest recreation areas, extant literature of forest park management in Taiwan has rendered scholarly discourses. This study summarizes possible predictors that could link to total tourist arrival: (1) temperature, (2) average rainfall (Leong et al., 2008; Li, 2013; Liu, 2010; Wu, 2008; Xu, 2006), (3) number of holiday days (Chiu, Tai, Tsai, & Chan, 2011), (4) domestic tourist arrival (Liu, Li, & Wang, 2014), (5) arrival of tourists from mainland China (Chiang, 2009; Wang & Cheng, 2011), (6) the severity of the occurrence of natural disasters (Dai, 2009), and (7) seasonality (He & Deng, 2007).

Method

Study Site

The Xitou Nature Education Area (XNEA), located in the mountain region in Central Taiwan, was selected as the study site as it is the mostly visited forest park in Taiwan (Cheng, Chen, & Wu, 2003; Wang, Lin, Chang, Huang, & Wang, 2011). However, in the last two decades, the study site has suffered from nature disasters including a record-breaking earthquake in Taiwan and devastating typhoons. Thanks to the Taiwanese government’s reconstruction efforts, modern infrastructure and new points of interest have been built. Meanwhile, several new activities associated with ecotourism have been developed surrounding the park.

Data Analyses

To identify the contributions of variables, a linear regression analysis model was established:

(1) Y = α + β i X i + β j D j + ε, ε N ( 0 , σ 2 )
where Y represents the number of tourists per month and denotes the dependent variable; α is the constant; X i is a critical variable with a magnitude among the variables affecting visitor count (i = 1, 2, …, 7) and entails (1) climate factors (average monthly temperature and rainfall), (2) number of holiday days per month, (3) total domestic visitor, (4) Monthly Gross Domestic Product (MGDP) per capita, and (5) severity of a natural disaster; D j is a dummy variable (j = 1, 2, 3). In this study, three dummy variables were established: “whether the ‘Monster Village’ is open,” “whether an ‘A Friend of Xitou’ card is issued,” and “seasonality.” Finally, a curve estimation procedure was adopted to analyze the correlations among the explanatory variables and the number of tourist arrivals. The following estimation equation was used, where, in addition to the aforementioned Y and α variables, X k represents the instance variable with a magnitude among the variables affecting visitation (k = 1, 2, …, 8).
(2) Y = α + β X k + ε, ε N ( 0 , σ 2 )

Selection of Research Variables

This study also analyzed the variables in association with the tourist arrivals to the study site. Those variables consist of the average monthly temperature, average monthly rainfall, number of holiday days per month, domestic visitor count, number of visitors from mainland China, severity of a natural disaster, seasonality, and MGDP per capita. Moreover, two specific attributes likely to impact the tourist arrival to the forest park under study were also included. They are “whether the ‘Monster Village’ is open,” and “whether an ‘A Friend of Xitou Card’ is issued.” The following section further explicates the characteristics of research variables under investigation.

Tourist Arrivals Data

The monthly tourist arrivals data between 1996 and 2014, obtained from the Taiwan Tourism Bureau (2014), show no tourist arrival in certain periods of time because the operation of park was disrupted by natural disasters. Those closures relate to the 921 Earthquake (i.e., October 1999–January 2000), Typhoons Toraji and Nari (i.e., August–October 2001), and Typhoon Mindulle (i.e., May–June 2004). From 1996 to 2011, the arrival data show an upward trend. However, after 2011, the arrival data exhibit a declining tendency. Fig. 1 presents the monthly changes in tourist arrivals from 1996 to 2014.

Fig. 1. 
XNEA Tourist Arrival in Months from 1996 to 2014.

Fig. 1.

XNEA Tourist Arrival in Months from 1996 to 2014.

Explanatory Variables

Average Monthly Temperature

Wu (2008) showed that the temperature of destination tends to affect tourists’ comfort of visit, which subsequently impacts the total number of visitors; thus, the higher the degree of comfort is, the greater the number of the arrival is. This study analyzed the average monthly temperatures from 1998 to 2014 in Nantou County, where the study site is located, as reported by the Central Weather Bureau, Executive Yuan (2014a). The temperatures (measured in °C) showed no significant difference, with the highest temperatures occurring between June and August – 28.6°C in July 2003 – and the lowest temperatures being observed between January and February – 12.5°C in January 2011. Fig. 2 presents the average monthly temperatures in Nantou County.

Fig. 2. 
Average Monthly Temperature of Nantou County from 1996 to 2014.

Fig. 2.

Average Monthly Temperature of Nantou County from 1996 to 2014.

Average Monthly Rainfall

A study (Li, 2013) revealed that tourist arrival is affected by average monthly rainfall. The lower the average monthly rainfall is, the greater the tourist arrival is. The present study looked into the average monthly rainfall (measured in millimeters) in Nantou County from 1996 to 2014, as reported by the Central Weather Bureau, Executive Yuan (2014a). The rainy season lasts from May to August in Nantou County; the dry season is from October to March the following year. Typically, typhoons occur from July through September in Taiwan. During 1996–2014, the highest average monthly rainfall appeared in September 2008 (1391.5 mm). Fig. 3 shows the changes in average monthly rainfall in Nantou County.

Fig. 3. 
Average Monthly Rainfall of Nantou County from 1996 to 2014.

Fig. 3.

Average Monthly Rainfall of Nantou County from 1996 to 2014.

Number of Holiday Days Per Month

In this study, the number of holiday days per month was derived from the data of Central Personnel Administration, Executive Yuan (Fig. 4), which includes holidays for anniversaries, such as the Founding of the Republic of China (Taiwan) and Peace Memorial Day, and holidays for folk festivals, such as the Lunar New Year Festival, Lunar New Year’s Eve, Tomb Sweeping Day, Dragon Boat Festival, and Mid-autumn Festival. Since January 1, 2001, the “two-day weekend” policy was implemented, which increased the number of holiday days per year from an average of more than 60 days in 1997 and 80 days in 1998 to more than 110 days in 2001. The highest number of holidays was reported in 2011 (115 holidays). In 2012, the number decreased by three days compared with 2011, totaling to 112 holidays, which comprised two-day weekends and make-up holidays.

Fig. 4. 
Number of Holiday Days per Month in Taiwan from 1996 to 2014.

Fig. 4.

Number of Holiday Days per Month in Taiwan from 1996 to 2014.

Number of Domestic Trips

In 2001, the average number of domestic trips per person was 5.26, annually. By 2011, it reached to 7.42, indicating a gradual increase in domestic trips (Liu et al., 2014). The relevant research data acquired from the Taiwan Tourism Bureau (2014) manifested that from 1996 to 1999, the annual domestic trip was considered as stable (Fig. 5). In 2000, the number began to gradually increase; growth in domestic trip began in 2008 and, compared with that in 2009, increased substantially (55%) in 2011. However, since 2012, the number of domestic trips slightly decreased, compared with that of 2011.

Fig. 5. 
Domestic Tourist Arrival from 1996 to 2014.

Fig. 5.

Domestic Tourist Arrival from 1996 to 2014.

Number of Tourists from Mainland China

Since Taiwan launched the “visiting relatives in China” program in 1987, social, cultural, and economic exchanges have gradually increased between mainland China and Taiwan. Through negotiations with the Straits Exchange Foundation and Association for Relations Across the Taiwan Straits, after the implementation of the mini-three-links expansion plan, beginning June 19, 2008, the number of visitors from mainland China has increased significantly, which has led to an economic impact on tourism in a colossus scale. Wang and Cheng (2011) reported a positive correlation between the number of tourists from mainland China and the total tourist arrivals to attractions in Taiwan. In 2008, the annual tourist arrivals from mainland China to Taiwan were 320,000, of which 100,000 came for purely sightseeing (Fig. 6). In 2010, the number of tourists from mainland China increased by 49%, compared with that in 2009. In 2009, 2010, 2012, and 2013, the highest number of visitors from mainland China was reported in April of those years. As shown in Fig. 6, the number of visitors from mainland China in 2013 increased by 374.59%, compared with that in 2008.

Fig. 6. 
Mainland Chinese Arrival from 1996 to 2014.

Fig. 6.

Mainland Chinese Arrival from 1996 to 2014.

Monthly Gross Domestic Product (MGDP) Per Capita

With an increase in Taiwan’s MGDP, the central government enacted new policies for holidays involving a two-day weekend for civil servants since 2001 and a work condition of no more than 48 hours per week. The increased leisure time has subsequently given ample opportunity for individuals to amply partake in different types of short leisure trips. As per the domestic trip survey conducted by the Taiwan Tourism Bureau, the total number of domestic travels in 2010 increased by 26.5%, compared with that in 2009, and the total travel spending increased by 30.3%. In 2006, the Directorate-General of Budget (DGOB) announced that the consumption of goods and services increased by 2.2%, of which recreation and culture-related services accounted for 18.7% (an increase of 1.9% compared with that in 1996) (Fig. 7). It seems to reveal the propensity of Taiwanese people that they are willing to spend more money on travel goods and services. For the analysis, this study used the average MGDP per capita data issued by the DGOB (2014).

Fig. 7. 
Monthly Gross Domestic Product Per Capita from 1996 to 2014.

Fig. 7.

Monthly Gross Domestic Product Per Capita from 1996 to 2014.

Severity of a Natural Disaster

Taiwan is situated in a region with earthquakes. Moreover, Taiwan is in the tropical zone where the occurrence of typhoons is rather frequent. Since forest recreation areas are mostly situated in mountains regions, these natural disasters frequently damage the infrastructures and facilities of forest parks. Thus, it has impacted the tourists’ willingness to visit (Dai, 2009) due to the perceptions of unsafe infrastructure. In this study, the severity of a natural disaster (such as typhoons, earthquakes, and others) was divided into five levels (1–5), as shown in Fig. 8. From 1996 to 2012, the number of tourist arrivals to the study site also changed, this could be attributed to major natural disasters such as Typhoon Herb, the 921 Earthquake, Typhoon Nari, Typhoon Mindulle, and Typhoon Morakot, as well as the epidemic of severe acute respiratory syndrome.

Fig. 8. 
Severity of a Natural Disaster from 1996 to 2014.

Fig. 8.

Severity of a Natural Disaster from 1996 to 2014.

Seasonality

Tu, Wu, Lin, and Jen (1999) and He and Deng (2007) reported that the tourist arrivals during the peak season (fall and summer) are significantly higher than that during the off-season (spring and winter). This study examined the tourist arrivals in the following four periods: spring (March–May), summer (June–August), autumn (September–November), and winter (December–February). In data coding, spring was D1 = 2, summer was D1 = 3, autumn was D1 = 4, and winter was D1 = 1. The seasons referred to time sequence data that repeat a similar pattern of change during a specific time of year.

Open of “Monster Village”

The distance between the study site (XNEA) and the second entrance of the Ming Shan Resort is about 100 m. In 2009, the resort established the “Monster Village,” which was officially opened in 2011. The “Monster Village” has attracted a large number of tourists who subsequently visited the XNEA. This study determined whether the establishment of the “Monster Village” had an influence on the tourist arrival. Hence, for this dummy variable, D2 = 0 represents the closed-to-public “Monster Village” before 2010; D2 = 1 represents the open-to-public “Monster Village” after 2011.

Number of A-Friend-of-Xitou Cardholders

Following the 921 Earthquake that struck on September 21, 1999, the XNEA introduced the A-Friend-of-Xitou card as a promotion mechanism to enhance revisit intention. The cardholders are entitled to an unlimited number of admissions to the XNEA within the validity period. In recent years, the number of XNEA tourist arrivals has continued to increase. To maintain leisure quality for other types of tourists during the peak season, the dates for free admissions for the cardholders were adjusted. Since December 1, 2010, a zero-admission fee has been offered to the cardholders for the period of one year, excluding Saturdays and Sundays from July to August. In this study, the dummy variable D3 was established to distinguish between the cardholders and non-cardholders. D3 = 0 represents the period without the card (before 1999) and D3 = 1 represents the period after the card was introduced (after 2000).

Results

Correlation Analysis

This study first employed the Pearson correlation analyses to examine the correlation between tourist arrival and 10 attributes discussed in the method section. The results (see Table 1) illustrate that monthly tourist arrival demonstrated a reversal tendency with two variables encompassing (1) number of holiday days per month and (2) severity of a natural disaster; however, the correlations were not statistically significant. Nevertheless, the remaining eight variables under study show a significant correlation (P < 0.001) with the number of tourist arrivals.

Table 1.

Variable Correlation Analysis.

Xitou Visitor Count Average Monthly Temperature Monthly Rainfall Number of Holiday Days per Month Domestic Visitor Count Whether “Monster Village” is Open Whether “A Friend of Xitou Card” is Issued Mainland Chinese Visitor Count Seasonality Severity of a Natural Disaster MGDP Per Capita
Xitou visitor count 1
Average monthly temperature 0.207** 1
Monthly rainfall 0.094 0.486*** 1
Number of holiday days per month −0.012 −0.247*** −0.188** 1
Domestic visitor count 0.574*** −0.188** −0.008 −0.003 1
Whether ‘Monster Village’ is open 0.647*** −0.085 −0.021 −0.017 0.791*** 1
Whether ‘A Friend of Xitou Card’ is issued 0.167** −0.272*** −0.023 −0.016 0.660*** 0.244*** 1
Mainland Chinese visitor count 0.645*** −0.079 0.016 −0.062 0.730*** 0.876*** 0.280*** 1
Seasonality 0.228** 0.665*** 0.207** −0.269*** −0.003 0.003 0.029 0.035 1
Severity of a natural disaster -0.123* 0.447*** 0.249*** −0.072 −0.346*** −0.269*** −0.218** −0.275*** 0.371*** 1
MGDP per capita 0.640*** −0.204** 0.018 −0.071 0.825*** 0.627*** 0.699**** 0.658*** 0.105 −0.299*** 1

Note: *p < 0.05, **p < 0.01, ***p < 0.001

Linear Regression Analysis

Results of linear regression analysis are presented in Table 2. After adjustment, the regression correlation coefficient is 0.703. The explanatory power of this model reached 70.3%, with a Durbin–Watson value of 1.063. The numerical value is not close to 2, but considerably close to 0, indicating a relatively strong correlation between the residuals. Furthermore, the regression analysis results were further used in residuals analysis to determine the difference between the observed dependent variable and the estimated values (see Table 3). The results demonstrate that when the residual scores are standardized, the overall mean is 0, the standard deviation is 1, and the standardized residual scores are distributed between −2 and 2. The standardized residual values are between −2.772 and −2.681, indicating that the regression model possesses a considerable amount of explanatory power.

Table 2.

Regression Analysis.

Measured Dimension Coefficient (t Value)
Dependent variable
Total number of visitors/month
Independent variable
Constant −324278.512
Average monthly temperature (degrees Celsius) 6094.915***
(6.339)
Monthly rainfall (mm/month) −21.317*
(−2.060)
Number of holiday days per month (number of days/month) 4536.722**
(2.745)
Domestic visitor count (domestic visitor count/month) 0.001
(0.375)
Mainland Chinese visitor count (number of people/month) 0.102
(1.264)
MGDP per capita (USD/month) 6.635***
(9.853)
Severity of a natural disaster (5 scales: 1–5) −429.684
(−0.214)
Seasonality (four seasons: 1–4) −3792.938
(−1.445)
Whether “Monster Village” is open (closed: 0; open: 1) 16316.385
(1.207)
Whether “A Friend of Xitou Card” is issued (not issued: 0; issued: 1) −44195.139***
(−5.138)
R 2 0.703
Adjusted R 2 0.687
Significance 0.000***
DW value 1.063

Note: *p < 0.05, **p < 0.01, ***p < 0.001.

Table 3.

Linear Regression Residuals Statistics.

Minimum Value Maximum Value Mean Standard Deviation Number
Predictive value 3378.05 195394.38 79145.36 39250.775 204
Residuals −83064.484 80335.750 0.000 29370.172 204
Standard predictive value −1.930 2.962 0.000 1.000 204
Standard residuals −2.772 2.681 0.000 0.980 204

Results of the regression analysis (Table 2) suggest that the higher the average monthly temperature is, the higher the tourist arrival is. For the number of holiday days per month variable, β = 4536.722 (P < 0.01), indicating that the higher the number of holiday days per month is, the higher the arrival is. For the MGDP per capita variable, β = 6.635 (P < 0.001), indicating that the higher the MGDP per capita is, the higher the tourist arrival is. For the “whether an A-Friend-of-Xitou card is issued” variable, β = –44195.139 (P < 0.001), indicating that once the card is issued, the total tourist arrival tends to decline. As shown (see Table 2), the higher the number of tourists from mainland China is, the higher the tourist arrivals to the study site is. Evidently, the fast growth of tourism economy in mainland China, to a certain degree, brings economic benefits to forest parks in Taiwan.

Curve Estimation Analysis

In this study, curve estimation analysis was deployed to examine the relationship among the explanatory variables and the monthly tourist arrival. The results are presented in Table 4.

Table 4.

Curve Estimation.

Measured Dimension Coefficient (t Value) Measured Dimension Coefficient (t Value) Measured dimension Coefficient (t Value)
Dependent Variable: Xitou Tourist Arrivals (Number of Tourists/Month) Dependent Variable: Xitou Tourist Arrivals (Number of Tourists/Month) Dependent Variable: Xitou Tourist Arrivals (Number of Tourists/Month)
Independent Variable Independent Variable Independent Variable
Constant 22123.010 Constant 74679.678 Constant 83569.190
(1.156) (16.573) (3.185)
Average monthly temperature (degrees Celsius) 2870.509 Monthly rainfall (mm/month) 20.432 Number of holiday days per month (number of days/month) −518.422
(3.002) (1.344) (−0.186)
R 2 0.043 R 2 0.009 R 2 0.000
Adjusted R 2 0.038 Adjusted R 2 0.004 Adjusted R 2 −0.005
Significance 0.003* Significance 0.180 Significance 0.852
Measured Dimension Coefficient (t Value) Measured Dimension Coefficient (t Value) Measured Dimension Coefficient (t Value)
Dependent Variable: Xitou Tourist Arrivals (Number of Tourists/Month) Dependent Variable: Xitou Tourist Arrivals (Number of Tourists/Month) Dependent Variable: Xitou Tourist Arrivals (Number of Tourists/Month)
Independent Variable Independent Variable Independent Variable
Constant −43547.737 Constant 62329.553 Constant 54270.784
(−3.464) (21.310) (6.738)
Domestic visitor count (domestic visitor count/month) 0.015 Mainland Chinese visitor count (number of people/month) 0.602 Seasonality (Four seasons: 1–4) 9781.888
(9.971) (11.961) (3.326)
R 2 0.330 R 2 0.415 R 2 0.052
Adjusted R 2 0.327 Adjusted R 2 0.412 Adjusted R 2 0.047
Significance 0.000*** Significance 0.000*** Significance 0.001**
Measured Dimension Coefficient (t Value) Measured Dimension Coefficient (t Value)
Dependent Variable: Xitou Tourist Arrivals (Number of Tourists/Month) Dependent Variable: Xitou Tourist Arrivals (Number of Tourists/Month)
Independent Variable Independent Variable
Constant 90581.773 Constant −135193.881
(11.986) (−7.402)
Severity of a natural disaster (5 scales: 1–5) −5146.125 MGDP per capita (NTD/month) 5.161
(−1.750) (11.832)
R 2 0.015 R 2 0.409
Adjusted R 2 0.010 Adjusted R 2 0.406
Significance 0.082 Significance 0.000***

Note: *p < 0.05; **p < 0.01; ***p < 0.001.

The results indicate that the average monthly temperature has a positive relationship with tourist arrivals. If the weather condition is favorable, tourists’ willingness to visit tends to increase, implying that the degree of comfort toward climate has a certain effect on the number of tourist arrival. Moreover, the total domestic trips affect the tourist arrival. The total tourist arrival from mainland China has grown rapidly since the implementation of the “opening-up policy.” The results further show that the number of tourists from mainland China has a significant relationship with the tourist arrival to the forest park under study.

The frequency of travel may differ according to the season; the tourists’ selection of travel destinations may also vary according to the season. In this study, the tourist arrival was divided into two categories: summer months (i.e., peak season) and winter months (i.e., off-season). During the off-season, the forest park may organize special activities to attract those (e.g., retirees) not affected by the seasonality issues and to improve the low-season visitation.

Conclusions

Given the prevalence of open-source data, this study presents an example of exploring secondary data sources to unveil the issues of significance in the context of forest parks. The study is able to extract the critical predictors of the tourist arrival to a forest park. In addition to the tourist arrival, the study data consist of the attributes of the effect of average monthly temperature, monthly rainfall, number of holiday days per month, number of domestic trips, number of tourists from mainland China, average MGDP per capita, severity of a natural disaster, seasonality, “whether the ‘Monster Village’ is open,” and “whether an A-Friend-of-Xitou card is issued.”

According to the combined results of correlation, regression, and curve estimation analyses, the effects on the total tourist arrival have been explored.

Table 5.

Research Analysis Compilation.

Correlation Analysis Regression Analysis Curve Estimation Analysis
Average monthly temperature Positive correlation/significant Positive correlation/significant Positive correlation/significant
Monthly rainfall Positive correlation/not significant Negative correlation/significant Positive correlation/not significant
Number of holiday days per month Negative correlation/not significant Positive correlation/significant Negative correlation/not significant
Domestic visitor count Positive correlation/highly significant Positive correlation/not significant Positive correlation/significant
Mainland Chinese visitor count Positive correlation/highly significant Positive correlation/not significant Positive correlation/significant
MGDP per capita Positive correlation/highly significant Positive correlation/highly significant Positive correlation/significant
Severity of a natural disaster Negative correlation/not significant Negative correlation/not significant Negative correlation/not significant
Seasonality Positive correlation/significant Negative correlation/not significant Positive correlation/significant
Whether “Monster Village” is open Positive correlation/significant Positive correlation/not significant
Whether “A Friend of Xitou Card” is issued Positive correlation/significant Negative correlation/highly significant
The study discovers various attributes significantly affecting the tourist arrival in a positive fashion; these include the average monthly temperature, number of holiday days per month, domestic visitor count, number of visitors from mainland China, average MGDP per capita, seasonality, and “whether the ‘Monster Village’ is open.”

In particular, the variables of (1) average monthly temperature and (2) MGDP per capita display a positive relationship in all three analytical methods. Thus, these two variables may be regarded as the primary predictors for the tourist arrival. The variables that have a positively significant relationship in two analytical methods are (1) monthly rainfall, (2) number of domestic trips, (3) number of tourists from mainland China, and (4) seasonality. Therefore, these variables are considered as secondary predictors. The variables with a positively significant relationship in one analytical method are number of holiday days per month and whether the “Monster Village is open.” The variable “whether A-Friend-of-Xitou card is issued” has a positive relationship in correlation and curve estimate approach. However, it shows a significant negative relationship in the regression analysis. Therefore, the effect of this variable on the tourist arrival remains unknown. It is suggested that further study deploying questionnaire survey to assess the underlying impact of the A-Friend-of-Xitou card in relation to the willingness to visit be undertaken. Furthermore, monthly rainfall is the only variable that exhibits a negative relationship in the regression analysis.

Managerial Implications

First, the results show that the higher the monthly rainfall is, the lower the tourist arrival is. It is noted that the weather condition is rather unstable in the mountain area under study during the raining season. Park managers may consider developing additional indoor facilities and relevant services to address inconveniences to the tourists that transpire during the rainy season. For example, it may construct additional pavilions to provide shelter from the wind and rain and render umbrella rental service at different locations.

Second, given that the tourists from mainland China have a significant effect on the number of the tourist arrivals, park managers may train interpreters to be familiar with the socioeconomic and culture background of the tourists. It would enable the park staffs to motivate the tourists from mainland China to enjoy ecotourism experiences involving the different types of natural plants, topography, and landscape surrounding the forest. It is also important to consider using simplified Chinese characters in printed materials (e.g., tour pamphlet) and website contents along with traditional Chinese characters, which are used officially in Taiwan. Moreover, a mobile application relating to the forest park, which supports multiple languages and characters (e.g., simplified Chinese), may be developed and download freely by all tourists.

Third, since the increase in MGDP per capita has a significant effect on the tourist arrival, it is advantageous to develop a pricing strategy to fulfill the diverse needs of tourists. For example, park managers may consider providing favorable pricing strategy to those living in adjacent cities to encourage them to utilize the park facilities more frequently.

Fourth, although the correlation between the tourist arrival and the perception of severity of a natural disaster is not significant, it is important to keep tourists’ safety. Disaster prevention policies should be adapted to local conditions and appropriate planning is required to safeguard the interests of public as well as the life of the tourist. Additionally, typical natural disasters such as earthquakes and typhoons are likely to disrupt the operations of the park. Hence, it is imperative that park managers implement preventive measures in advance and practice disaster prevention schemes on a regular basis.

Finally, the more holidays per month, the more tourist arrivals. Since the number of tourist arrivals is much lower during the off-season, park managers may create campaigns to attract different groups of tourists whose schedules would be less affected by holidays. For example, photography contests could be conducted on non-holidays. The park managers may also consider to offer discounted admission tickets during weekdays.

References

Central Personnel Administration, Executive Yuan (2014) Central Personnel Administration, Executive Yuan . (2014). Personnel Statistics. Retrieved from http://www.dgpa.gov.tw/np.asp?ctNode=386&mp=7

Central Weather Bureau, Executive Yuan (2014a) Central Weather Bureau, Executive Yuan . (2014a). Climate Statistics. Retrieved from http://www.cwb.gov.tw/V7/climate/dailyPrecipitation/dP.htm

Central Weather Bureau, Executive Yuan (2014b) Central Weather Bureau, Executive Yuan . (2014b). Disaster Prevention Weather Forecast. Retrieved from http://www.cwb.gov.tw/V7/prevent/

Cheng, Chen, & Wu (2003) Cheng, F. S. , Chen, F. S. , & Wu, W. L. (2003). Xitou Forest Recreation Area No. 3 Landslide Hazard Remediation. In Proceedings of 10th Conference on Current Researches in Geotechnical Engineering in Taiwan. Taipei: National Taiwan University, July 22, 2003.

Chiang (2009) Chiang, H. M. (2009). The effect of “opening up of Taiwan to mainland tourists” from the perspective of globalization trend. Journal of Tourism and Travel Research, 4(2), 126.

Chiu, Tai, Tsai, & Chan (2011) Chiu, C. R. , Tai, C. H. , Tsai, W. L. , & Chan, W. K. (2011). Influences of natural disasters on tourist arrivals at national forest recreation areas in Taiwan. Quarterly Journal of Chinese Forestry, 44(2), 249264.

Chiu & Chih (2012) Chiu, T. Y. , & Chih, W. H. (2012). Study the determinants of ecotourism behavior. Marketing Review, 9(2), 203222.

Dai (2009) Dai, J. S. (2009). A Study on the Impact of Natural Disaster to the Number of Visitors of Forest Recreational Areas. Master’s thesis, School of Forestry and Resource Conservation, National Taiwan University, Taipei, Taiwan.

DGOB (2014) DGOB . (2014). National Income and Economic Growth. Directorate-General of Budget, Accounting and Statistics, Executive Yuan, R.O.C. (Taiwan). Retrieved from http://www.dgbas.gov.tw/ct.asp?xItem=37407&CtNode=3566&mp=1

He & Deng (2007) He, R. G. , & Deng, G. H. (2007). The model of time series prediction of the number of the tourists in Guilin. Journal of Guangxi Academy of Sciences, 23(3), 153156.

Lee & Huang (2006) Lee, M. T. , & Huang, Y. C. (2006). The research of tourist’s willingness to pay for interpretative service: A case of ecological reserve area in Kentin National Park. Journal of Leisure and Tourism Industry Research, 1(1), 1933.

Leong et al. (2008) Leong, R. , Chang, C. S. , Lin, T. Y. , Hsu, B. S. , Wey, T. H. , & Chou, K. H. (2008). Evaluation debris flow rebuilt by using the tourist number in Sitou after typhoon Toraji. Journal of the Experimental Forest of National Taiwan University, 22(1), 3547.

Li (2013) Li, C. L. (2013). The Impact of Weather on Visitor Flows: Evidence from Taiwan. Master’s thesis, Graduate Institute of Economics, National Chi Nan University, Nantou, Taiwan.

Liu (2010) Liu, C. (2010). Research about the tour climatic comfort degree. Journal of Anhui Agricultural Sciences, 38(23), 1288312884, 12887.

Liu, Li, & Wang (2014) Liu, K. Y. , Li, C. L. , & Wang, Y. H. (2014). Exploring the influence of values and socio-demographics on perception of crowding: The case of national park and forest recreation visitors in Taiwan. Quarterly Journal of Forest Research, 36(1), 2944.

Newsome, Moore, & Dowling (2013) Newsome, D. , Moore, S. , & Dowling, R. (2013). Natural area tourism: Ecology, impacts and management. Clevedon: Channel View Publications.

Taiwan Tourism Bureau (2014) Taiwan Tourism Bureau . (2014). Annual Tourism Statistics. Available: http://admin.taiwan.net.tw/statistics/year.aspx?no=134

Tu, Wu, Lin, & Jen (1999) Tu, A. , Wu, W. Y. , Lin, J. C. , & Jen, I. (1999). Fluctuation of tourist number and revenue of the National Forest Recreation Areas in Taiwan, 1990–1998. Journal of Outdoor Recreation Study, 12(4), 6172.

Wang & Cheng (2011) Wang, M. Y. , & Cheng, C.-F. (2011). Current status and future outlook of tourism industries in Taiwan. Journal of Commercial Modernization, 6(1), 209228.

Wang, Lin, Chang, Huang, & Wang (2011) Wang, Y. N. , Lin, L. C. , Chang, P. C. , Huang, Y. R. , & Wang, C. D. (2011). Preliminary investigation of visitor arrivals and management implication in Xitou Nature Education Area. Journal of the Experimental Forest of National Taiwan University, 25(3), 181191.

Wu (2008) Wu, I. L. (2008). A Study on the Climate Affecting Number of Visitors of the Chitou Forest Recreational Areas. Master’s thesis, School of Forestry and Resource Conservation, National Taiwan University, Taipei, Taiwan.

Xu (2006) Xu, C. T. (2006). The distinguishing feature, form, development and protection of the scene climates tourism resources by our country are researched. Journal of Zaozhuang University, 23(2), 103107.