www.palgrave-journals.com/thr

Using decision trees to identify tourism stakeholders: The case of two Eastern North Carolina counties
Erick T. Byrd* and Larry Gustke Received (in revised form): 1st May, 2006
*Department of Recreation, Tourism, and Hospitality Management, The University of North Carolina at Greensboro, PO Box 26170, Greensboro, NC 27402, USA Tel: + 1 336-334-3041; Fax: + 1 336-334-3238; E-mail: etbyrd@uncg.edu

Erick T. Byrd is an Assistant Professor in the Department of Recreation, Tourism, and Hospitality Management at The University of North Carolina at Greensboro. His current research interests focus on community participation in tourism development. Larry Gustke is an Associate Professor in the Department of Parks, Recreation and Tourism Management at North Carolina State University. His current research interests focus on community tourism planning. ABSTRACT KEYWORDS: decision tree analysis, stakeholder inclusion, sustainable tourism, tourism planning

their support for sustainable tourism development in their community.
Tourism and Hospitality Research (2007) 7, 176–193. doi:10.1057/palgrave.thr.6050049

This paper explores stakeholder involvement in tourism planning, development, and management. For tourism planners to include stakeholders in the tourism planning process those stakeholders and their interests need to be identified. The research reported in this paper describes and applies an analytical technique that is not traditionally used to identify stakeholders. A questionnaire was developed and mailed to stakeholders in two rural communities in North Carolina. The data were analysed with an Exhaustive Chi-square Automatic Interaction Detection decision tree. From the results of the decision tree, stakeholder groups were identified in relation to

INTRODUCTION Tourism has become one of the main industries identified as having the potential to assist local communities through diversifying the economy, enhancing community pride and awareness, and supporting the increased development of public services (Hassan, 2000; Long et al., 1990). For tourism development to be successful however, it must be planned and managed responsibly (De Oliveira, 2003; Inskeep, 1991; Martin, 1995; Southgate and Sharpley, 2002; Yuksel et al., 1999). Inskeep (1991) furthered the concept of responsible management stating, ‘ill-conceived and poorly planned tourism development can erode the very qualities of the natural and [social] environments that attract visitors in the first place’ (p. 460). Communities that use or plan to use tourism as a tool to diversify their economy must develop policies for the sustainable development of the community (De Oliveira, 2003; Pucako and Ratz, 2000; Southgate and Sharpley, 2002; Yuksel et al., 1999). Gunn (1994) suggests that the success and implementation of a tourism development plan is often based on the support of stakeholders such as citizens,

176

Tourism and Hospitality Research

Vol. 7, 3/4, 176–193

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Byrd and Gustke

entrepreneurs, and community leaders. Therefore, tourism as a development strategy requires deliberate planning with the inclusion of stakeholders. A stakeholder is defined as ‘any group or individual who can affect or is affected by the achievement of the organisations objectives’ (Freeman, 1984: 46), and that group or individual has a legitimate interest in the organisation (Donaldson and Preston, 1995). To include stakeholders in tourism development planning, they must be identified and their interests and needs understood. Traditionally, stakeholders have been identified in a community based on geographic (Cottrell, 2001; Davis and Morais, 2004; Murphy, 1983), demographic (De Lopez, 2001; Hassan, 2000; Ryan, 2002), and sociographic characteristics (Mathieson and Wall, 1982; Murphy, 1985). Planners identify groups based on characteristics and then develop and implement strategies to include the different groups in the planning process. There are inherent biases to these approaches that limit their usefulness. The major bias is that planners make subjective judgments on who and what groups are included to represent stakeholders. The purpose of this research paper is to use decision tree analysis in the identification of stakeholders in a community. Decision tree analysis is a tool for partitioning a data set based on the relationships between a set of independent variables and a dependent variable.The research reported here tests the application of, decision tree analysis, an analytical technique that is not traditionally used to identify stakeholders, but has been used in healthcare, market analysis, credit scoring, and policy studies.
STAKEHOLDERS AND TOURISM Stakeholder involvement in policy development is associated with the early ideas of community participation and public involvement that are central to basic democratic beliefs (Fiorino, 1990), as well as modern business management concepts. There is a substantial literature on stakeholders and stakeholder involvement in both business management (Clarkson, 1995;

Donaldson and Preston, 1995; Freeman, 1984; Johnson and Scholes, 1999; Jones, 1995; Stoney and Winstanley, 2001) which focuses on the management and power of the stakeholder, and public administration (Ansari and Phillips, 2001; Arnstein, 1969; Beierle, 1998; Carmin et al., 2003; Carter and Darlow, 1997; Crosby et al., 1986; Curry, 2001; Fiorino, 1990; Simrell et al., 1998; Steelman, 2001) which focuses on the right a stakeholder has to be involved no mater their level of power. Researching stakeholder groups and the significance of their interests has also been a reoccurring theme in the tourism literature (Allen et al., 1993; Andereck and Vogt, 2000; Davis and Morais, 2004; De Lopez, 2001; Gunn, 1994; Markwick, 2000; Murphy, 1983; Ryan, 2002; Robson and Robson, 1996; Vincent and Thompson, 2002; Yuksel et al., 1999). Current tourism concepts about stakeholders and their role in tourism development are built on the business management and public administration literature. In general, there are four major stakeholder perspectives identified in tourism. These perspectives are the tourist, the residents, the business owners, and the local governmental officials (Goeldner and Ritchie, 2002). In previous research (Allen et al., 1993; Murphy, 1983; Pizam, 1978; Lankford, 1994), these perspectives were used to categorise different groups involved in the tourism development process. Much of the research investigating the complex relationship between stakeholders and tourism has focused on individual stakeholder groups; residents, visitors, business owners, or government officials; and their perceptions and attitudes (Andereck and Vogt, 2000; Brunt and Courtney, 1999; Long et al., 1990; Martin, 1995; Mathieson and Wall, 1982; Murphy, 1985; Pizam et al., 2000). Hardy and Beeton (2001) stress the need for studies that look at multiple stakeholder groups and compare them based on their interests. Also, investigations of multiple tourism stakeholder groups revealed differences in attitudes and perceptions (Byrd, 1997; Kavallinis and Pizam, 1994), suggesting the need for a mechanism to explain and manage these different views.

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Vol. 7, 3/4, 176–193

Tourism and Hospitality Research

177

Using decision trees

Researchers have argued that tourism planners should consider the interests of all stakeholders before proceeding with development efforts (Hardy and Beeton, 2001; Sautter and Leisen, 1999; Vincent and Thompson, 2002). Yuksel et al. (1999) stated that the incorporation of stakeholder views and interests can reduce conflicts in the long term by ‘drawing on the knowledge and insights of stakeholders (359)’. Sautter and Leisen (1999) found that as agreement across stakeholder interests increased, so did the likelihood of collaboration and compromise. Including Sautter and Leisen (1999) there are a number of researchers and professionals who are advocating the inclusion of stakeholders in the planning process (Hardy and Beeton, 2001; Gunn, 1994; Ioannides, 1995; Markwick, 2000).Two distinct areas of thinking about stakeholders and tourism have emerged. The first is closely related to the classical idea of stakeholder management.The central agency considers the interest of the stakeholders and develops policies and practices based on the stakeholders’ power and influence. Those with more power would be given more consideration than those with less (De Lopez, 2001; Hunt and Haider, 2001; Markwick, 2000; Reed, 1997). The second area of stakeholder theory that has emerged over the past few years includes the concept of collaborative thinking (Bramwell and Sharman, 1999; Jamal and Getz, 1995; Sautter and Leisen, 1999; Yuksel et al., 1999). This idea is based on the normative approach to stakeholder theory. It implies that consideration should be given to each stakeholder group without one being given priority over others (Sautter and Leisen, 1999). In particular, stakeholder identification and involvement is the main step towards achieving community partnerships and collaboration within tourism (Hardy and Beeton, 2001). Yuksel et al. (1999) stated while collaborative planning may be time-consuming and difficult, it can be justified because potentially it can ‘avoid the costs of resolving conflicts in the long term, it is more politically legitimate, and

it can build on the store of knowledge and capacities of the stakeholders’ (p. 315). Collaboration and partnerships are essential to the development of bonds and networks among diverse stakeholders for their benefit (Briassoulis, 2002). The idea of collaboration is not new to stakeholder theory. Freeman (1984) stated there were three main concepts that need to be understood to effectively manage stakeholders. The first two concepts focus on the identification of stakeholder interests and the process necessary to understand the relationship between organisations and stakeholder. The third focuses on the management of exchanges between all stakeholders (Freeman, 1984). Failure to identify the interest of even a single primary stakeholder group could result in the failure of the process (Clarkson, 1995). Therefore, there is a need for techniques and tools to assist in the identification of stakeholders.
Stakeholder identification A way to accomplish stakeholder group identification is to apply the concept of segmentation. Simply, segmentation is the dividing of groups (markets, stakeholders) into smaller groups based on specific characteristics. The process begins with the identification of a structure to segment the market. Traditionally, this has been done thorough geographic, demographic, psychographic, and behavioral variables (Kotler et al., 2003). Segmentation assumes people are different and that the differences are related to a specific behavior or attitude. Based on the differences people can be grouped into segments (MacKay et al., 2002). To be useful stakeholder group identification through segmentation should have four characteristics. The first is measurability. The main aspects that need to be measured are the size and the interests of the segment. The second is accessibility. For a segment to be used it must be accessible. The third is substantiality, the size of the segment is large enough to be of use. The fourth is actionability or the effective designing of programs (Kotler et al., 2003).

178

Tourism and Hospitality Research

Vol. 7, 3/4, 176–193

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Byrd and Gustke

Owen (1998) in his review of the market segmentation research and literature concluded that there is a need for new approaches to segmentation. Arimond and Elfessi (2001) echoed this need. This need also holds true for stakeholder identification in a community. Owen states that the new techniques should be conducted computer-based software to lessen the ‘bias of human judgment’. Arimond and Elfessi (2001) indicated new segmentation approaches should allow for simpler data collection and survey methods. Also, these new approaches should provide graphic displays that reveal the relationships inherent in the data.
Decision trees One approach that addresses Owen’s (1998) and Arimond and Elfessi (2001) ideas for new segmentation is the use of decision tree analysis. A decision tree is a model that can be used to classify or predict variables (Kass, 1980; Koskela, 1999). It is a method used to study the relationship between a dependent variable and independent variables (Huba, 2003). Decision trees are designed to handle a large number of independent variables at differing levels of measurement. Decision trees identify groups by dividing the dependent variable by the independent variables. The results of the decision tree analysis indicate which independent variables are most strongly related to the dependent variable. Decision tree analysis has been used in financial settings for credit scoring, manufacturing settings for quality control, and healthcare settings for determining treatments (SPSS, 2002). This study used the Exhaustive Chi-square Automatic Interaction Detection (CHAID) software available from SPSS. The Exhaustive CHAID is a second-generation CHAID algorithm (Huba, 2003). It was selected for the study because it is a more comprehensive method and yields more accurate results than CHAID and the other decision tree methods within SPSS Answer Tree (Huba, 2003; SPSS, 2003). CHAID was first designed for use with nominal variables (Kass, 1980). SPSS extended

the capabilities of CHAID so that it can include nominal, ordinal, interval, and ratio variables (Huba, 2003). CHIAD splits the data into subsets that best describe the dependent variable (Kass, 1980). For nominal and ordinal variables, chi-square analyses are used, and for interval and ratio variables an analysis of variance is used (Huba, 2003). The split is based on which variable has the lowest p-value. If a tie occurs between two or more variables the variable that has the highest F-value is selected as the predictor. The strength of the CHAID method is found when attempting to detect patterns in large data sets that are comprised of differing levels of measurement. CHIAD can handle independent and dependent variables at all levels of measurement. Also, not all independent variables have to be at the same level of measurement. In other words if a data set was comprised of independent variables that were nominal, categorical, interval and ratio the CHIAD decision tree analysis can include all the variables in the analysis (Huba, 2003). Through decision trees a visual representation of the variables and their interaction is developed. Due to its parsimonious nature decision trees facilitates understanding by both researchers and other tourism professionals.
STUDY METHODOLOGY During the winter and early spring of 2003, a study of tourism stakeholders was conducted in two rural counties in eastern North Carolina. The counties selected for the study were Johnston County and Martin County. The sample for the study, which included the county residents, business owners, local government officials, and tourists, were mailed a questionnaire inquiring about their attitudes and perceptions of tourism development, their participation in local political and tourism activities, and their view of the environmental impacts of tourism. A questionnaire was developed based on previous research and variables identified in the literature (Mason and Cheyne, 2000;

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Vol. 7, 3/4, 176–193

Tourism and Hospitality Research

179

Using decision trees

McFarlane and Boxall, 2000; Perdue et al., 1990; Stein et al., 1999; Vincent and Thompson, 2002). Questions included perceptions about tourism development and the environment in relation to the local community, perceptions of tourism impact in local community, types and amount of community participation, individual’s participation in recreational activities, and basic demographics. The questionnaire consisted of 56 items that included nominal, categorical, and interval level of measurement. The questionnaire included both open ended and closed ended questions. The majority of the items were five-point Likert scale questions. The values of 1 to 5 represented a respondents’ agreement or disagreement with the statement. A 1 represented strong disagreement and a 5 represented strong agreement. The format and scale were selected based on the format of previous studies (Andereck and Vogt, 2000; Lankford and Howard, 1994; Maddox 1985; Mason and Cheyne, 2000; Vincent and Thompson, 2002; Williams and Lawson, 2001). The independent variables for the study were: perspective, perceived impact (an index that was developed from 13 Likert scale questions) participation in 13 recreational activities, length of residency, age, location of residence, education, gender, household status, and whether the respondent belonged to an environmental organisation. For this study, an index of sustainable tourism for the community was developed and served as the independent variable for the study. The index represented the respondent’s level of support for sustainable tourism development due to the complex nature of sustainability. There was no definition of sustainability provided to the respondent. Answers to the questions were considered as indicators of their levels of support to different aspects of sustainable tourism development.The index was developed based on recommendations by Babbie (1995). An additive index was used because of the investigative nature of the study and the lack of literature that would justify another type of index. Also, employing

additive indices keeps the study conservative (Babbie, 1995). The index consisted of variables that indicated the stakeholders’ perceptions towards tourism development, the natural environment, the local economy, and the community. Face validity was obtained by developing questions about variables that previous research (Mason and Cheyne, 2000; McFarlane and Boxall, 2000; Perdue et al., 1990; Stein et al., 1999; Vincent and Thompson, 2002) indicated were relevant. Each question was evaluated for its relevance to the counties that were being studied. After reviewing the face validity and the reliability for each measure, the variables that were found to be relevant were retained and others deleted. This resulted in a total of 17 questions or measures in the Support for Sustainable Tourism Index (SSTI) (see Table 1). Sampling was conducted by drawing independent samples from each of the four perspective groups. Random selection of residents and business owners was achieved using a, USADATA, commercial source. A list of visitors that had contacted the local convention and visitors bureaus over the past 12 months was used to generate the population of tourists. A random number table was employed to conduct sampling of the tourist. The total population for the government officials was utilised, obtained form the county office. Each group was sampled independently to ensure that each perspective was represented in the study (Goeldner and Ritchie, 2002). A total of 2,527 surveys were mailed to tourism stakeholders in Johnston and Martin Counties using a modified Total Design Technique (Dillman, 2000). Two weeks after the initial mailing a reminder postcard was sent to nonrespondents.Two weeks following the postcard, a second questionnaire was sent to county resident nonrespondents. Out of the 2,527, a total of 542 were returned usable. Two hundred and sixteen were undeliverable because of incorrect address or requests not to be included in the study. This resulted in a total response

180

Tourism and Hospitality Research

Vol. 7, 3/4, 176–193

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Byrd and Gustke

Table 1: Index items for support for sustainable tourism index Item 1 2 3 4 Tourism protects the environment Tourism creates new jobs Tourism activities should be integrated with the county’s conservation programmes Tourism development should be discouraged when it harms the natural environment Economic gains are less important than the natural environment Tourism must be managed so that the environment is sustainable over time Tourism must not be allowed to damage natural resources Environmental education programmes lead to improvements in natural resources Children need to learn about the environment A community’s economic stability is more important than its environmental concerns The restoration of historical sites promotes tourism Tourism should be managed to meet the needs of the present Tourism should be managed to meet the needs of the future Tourism development should respect the scale, nature, and character of my county Involvement of residents in environmental activities enhances tourism Economic development funds should be used to promote tourism Local government should be involved in tourism development Mean 3.01 4.16 3.81 3.97 Std. deviation 0.89 0.76 0.86 0.96 Range 4 4 4 4 Min 1 1 1 1 Max 5 5 5 5

5 6 7 8 9 10 11 12 13 14 15 16 17

3.30 4.09 4.13 3.93 4.33 2.71 4.13 3.75 4.04 4.15 3.75 3.30 3.85

1.10 0.73 0.84 0.73 0.66 0.98 0.73 0.79 0.62 0.62 0.79 0.93 0.75

4 4 4 4 4 4 4 4 4 4 4 4 4

1 1 1 1 1 1 1 1 1 1 1 1 1

5 5 5 5 5 5 5 5 5 5 5 5 5

rate of 21.40 per cent. The response rate for the individual groups were 23.90 per cent residents, 20.00 per cent tourist, 20.90 business owners, and 35.60 business owners.
RESULTS AND DISCUSSION The purpose of this study was to investigate the use of decision trees as a tool to identify stakeholder groups based on their level of support for sustainable tourism development. A decision tree analysis was used to identify stake-

holder groups. The resulting tree, Support for Sustainable Tourism Decision Tree (SSTDT), has 24 nodes (see Figure 1). Of the 24 nodes, 13 are terminal nodes. A node represents a group or sub-group on the tree. When a node (group) cannot be divided further it is considered a terminal node. The SSTDT had 13 terminal nodes indicating that 13 distinct groups can be identified through the use of the decision tree. The 13 groups range in size from 10 respondents to 102 respondents. Their

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Vol. 7, 3/4, 176–193

Tourism and Hospitality Research

181

182 Using decision trees
Support for Sustainable Tourism Development Node 0 Mean 64.9823 Std. Dev. 6.8898 n 480 % 100.00 Predicted 64.9823 Perception of Tourism’s Impact on the Community Adj. P-value=0.0000, F=24.7761, df=2,477

−

Tourism and Hospitality Research 45 Node 3 Mean 69.1667 7.2041 Std. Dev. 84 n % 17.50 Predicted 69.1667 Participated in Bird watching Adj. P-value=0.0080, F=7.3903, df=1,82

− −

−

Visited Visitor Center Adj. P-value=0.0473, F=4.0485, df=1,86 Yes Yes No Node 7 Mean Std. Dev. n % Predicted 63.8810 5.4871 231 48.13 63.8810 Node 6 Mean 66.7662 6.9242 Std. Dev. n 77 % 16.04 Predicted 66.7662 Participated in Hiking Adj. P-value=0.0131, F=6.4594, df=1,75 Node 5 Mean 64.7778 6.1290 Std. Dev. n 27 % 5.63 Predicted 64.7778

No

Yes Node 8 Mean 72.8500 5.4219 Std. Dev. n 20 % 4.17 Predicted 72.8500

No Node 9 Mean Std. Dev. n % Predicted 68.0156 7.3387 64 13.33 68.0156

Vol. 7, 3/4, 176–193
− −
Gender Adj. P-value=0.0047, F=10.7891, df=1,229

Node 4 61.2295 Mean 8.1943 Std. Dev. n 61 % 12.71 Predicted 61.2295

Participated in Gardening Adj. P-value=0.0162, F=6.1334, df=1,59

−

Attended a festival Adj. P-value=0.0341, F=4.6958, df=1,62

−

Yes Yes Node 12 Node 13 Mean Std. Dev. n % Predicted 65.4231 6.9546 52 10.83 65.4231 Mean 69.5600 Std. Dev. 6.0833 25 n 5.21 % Predicted 69.5600 No Node 11 57.9091 Mean 8.3831 Std. Dev. n 22 4.58 % Predicted 57.9091

No

Male Node 14 Mean Std. Dev. n % Predicted 62.9275 5.4083 138 28.75 62.9275 Prespective of Tourism Adj. P-value=0.0501, F=9.4885, df=1,136

Female, Node 15 Mean Std. Dev. n % Predicted 65.2957 5.3223 93 19.38 65.2957

Yes Node 16 Mean 69.3182 6.8396 Std. Dev. n 44 % 9.17 Predicted 69.3182

No Node 17 Mean 65.1500 7.7546 Std. Dev. n 20 % 4.17 Predicted 65.1500

Node 10

Mean 63.1026 Std. Dev. 7.5631 39 n % 8.13 Predicted 63.1026

− −
Attended a Sports Event Adj. P-value=0.0077, F=8.5292, df=1,23 Yes Node 20 Mean 66.3333 5.8205 Std. Dev. n 12 % 2.50 Predicted 66.3333 No Node 21 72.5385 Mean Std. Dev. 4.7891 n 13 % 2.71 Predicted 72.5385

−

Area of Residency Adj. P-value=0.0066, F=8.2714, df=1,37

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00
Offical;Business, Node 22 60.6111 Mean Std. Dev. 5.9822 n 36 % 7.50 Predicted 60.6111

Urban

Rural

Resident;Visitor Node 23 Mean 63.7451 Std. Dev. 4.9686 n 102 % 21.25 Predicted 63.7451

Node 18 Mean 60.8462 6.9897 Std. Dev. 26 n 5.42 % Predicted 60.8462

Node 19 67.6154 Mean 6.8012 Std. Dev. 13 n 2.71 % Predicted 67.6154

Figure 1 Support for sustainable tourism development decision tree

Byrd and Gustke

support for sustainable tourism, based on the mean SSTI for each group, ranges from 57.90 little or no support (node 11) to 72.85 very supportive (node 8). The best indicator of support for sustainable tourism development is the stakeholder’s perception of existing tourism development impacts in their community ( p>0.001) (see Figure 1). Using their perception of tourism impact, the stakeholders can be divided into three distinct groups or branches representing their level of support for sustainable tourism. Group 1 represents individuals that moderately to neutrally perceive that tourism had a negative impact on their community; represented 18.33 per cent of the respondents in this study (node 1). Group 2 represents individuals that neutrally to moderately perceive that tourism had a positive impact on their community; represented 64.17 per cent of the respondents (node 2). Group 3 represents individuals that moderately to strongly perceive that tourism had a positive impact on their community; represented 17.50 per cent of the respondents (node 3).
Group 1 (Branch 1) — Local worriers The three groups identified by the first decision tree segmentation can each be further segmented by other variables. Group 1, the local worriers, can be segmented first by, if the individual stakeholder had visited the tourism visitor’s center in their community ( p = 0.047) represented by nodes 4 and 5 (see Figure 2). Stakeholders who had not visited the visitor’s center can then be segmented by their participation in gardening activities ( p = 0.016) represented by nodes 10 and 11. Stakeholders who gardened can be segmented by the area, rural or urban, in which they reside ( p>0.001) represented by nodes 18 and 19. Based on this branch of the decision tree, four variables are identified to influence a stakeholder’s support for sustainable tourism development. Those variables are: perception of existing tourism development impact in their community, visitation to the local visitor’s center, participation in gardening

as recreation, and the area in which they reside. Using this information it can be concluded that the group that will be most opposed to sustainable tourism development are individuals who perceive tourism as having a negative impact on the community, have not visited the visitor’s center and do not garden. This group can be described as locals who perceive tourism has a negative impact to the community, are not active in the community, and would not be supportive of sustainable tourism development initiatives.
Group 2 (Branch 2) — Moderates The second group, the moderates, can be segmented into two sub-groups (see Figure 1). This division is made based on the variable of participation in bird watching as a recreational activity. Group 2a engages in bird watching and group 2b does not.Those who have a moderate view of tourism’s impact on the community and bird watch are more supportive (66.77) of sustainable tourism than those who do not bird watch (63.88). Group 2a (Branch 2a) — Physically active moderates Individuals who have a moderate view of tourism’s impact on their community and bird watch can be further divided by their participation in hiking (p = 0.013) represented by nodes 12 and 13 (see Figure 3). Those who do hike are more supportive (69.56) than those who do not (65.42). Hikers can be further divided into those who attended sporting events ( p = 0.007) represented by nodes 20 and 21. Stakeholders who attend sports events are less supportive (66.33) than those who do not attend sports event (72.54). Based on this subbranch of the decision tree, four variables are identified as influencing stakeholders support for sustainable tourism development. Those variables are: perception of existing tourism development impact in their community, participation in bird watching, participation in hiking, and attendance of sporting events. Using this information, it can be concluded that the

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Vol. 7, 3/4, 176–193

Tourism and Hospitality Research

183

Using decision trees

Support for Sustainable Tourism Development Node 0 Mean Std. Dev. n % Predicted 64.9823 6.8898 480 100.00 64.9823

−
Perception of Tourism’s Impact on the Community Adj. P-value=0.0000, F=24.7761, df=2,477

45 Node 3 Mean Std. Dev. n % Predicted 69.1667 7.2041 84 17.50 69.1667

−
Visited Visitor Center Adj. P-value=0.0473, F=4.0485, df=1,86

+

+

No Node 4 Mean Std. Dev. n % Predicted 61.2295 8.1943 61 12.71 61.2295

Yes Node 5 Mean Std. Dev. n % Predicted 64.7778 6.1290 27 5.63 64.7778

−
Participated in Gardening Adj. P-value=0.0162, F=6.1334, df=1,59

Yes Node 10 Mean Std. Dev. n % Predicted 63.1026 7.5631 39 8.13 63.1026

No Node 11 Mean Std. Dev. n % Predicted 57.9091 8.3831 22 4.58 57.9091

−
Area of Residency Adj. P-value=0.0066, F=8.2714, df=1,37

Urban Node 18 Mean Std. Dev. n % Predicted 60.8462 6.9897 26 5.42 60.8462

Rural Node 19 Mean Std. Dev. n % Predicted 67.6154 6.8012 13 2.71 67.6154

Figure 2 Branch 1 — local worriers

moderate group that will be most supportive of sustainable tourism development are individuals who perceive tourism as having a moderate impact on the community, bird watch, hike, but do no attend sporting events.
Group 2b — Passively active moderates of sustainable tourism Individuals who have a moderate view of tourism’s impact on their community and do not bird watch can be further divided by their

gender (p = 0.005) represented by nodes 14 and 15 (see Figure 4). Females are more supportive (65.30) than males (62.93). Males can be further divided by their perspective (resident, business owner, government official, tourist) in relation to tourism (p = 0.05). Residents and tourist are more supportive (63.75) than business owners and governmental officials (60.61). Based on this sub-branch of the decision tree, four variables are identified to influence a stakeholder’s support for sustainable tourism development.

184

Tourism and Hospitality Research

Vol. 7, 3/4, 176–193

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Byrd and Gustke

Support for Sustainable Tourism Development Node 0 Mean Std. Dev. n % Predicted 64.9823 6.8898 480 100.00 64.9823

−
Perception of Tourism’s Impact on the Community Adj. P-value=0.0000, F=24.7761, df=2,477

45 Node 3 Mean Std. Dev. n % Predicted 69.1667 7.2041 84 17.50 69.1667
+

Participated in Bird watching Adj. P-value=0.0002, F=13.9196, df=1,306

Yes Node 6 Mean Std.Dev. n % Predicted 66.7662 6.9242 77 16.04 66.7662
−

No Node 7 Mean Std.Dev. n % Predicted 63.8810 5.4871 231 48.13 63.8810
+

Participated in Hiking Adj. P-value=0.0131, F=6.4594, df=1,75

Yes Node 12 Mean Std.Dev. n % Predicted 69.5600 6.0833 25 5.21 69.5600
−

No Node 13 Mean Std.Dev. n % Predicted 65.4231 6.9546 52 10.83 65.4231

Attended a Sports Event Adj. P-value=0.0077, F=8.5292, df=1,23

Yes Node 20 Mean Std. Dev. n % Predicted 66.3333 5.8205 12 2.50 66.3333

No Node 21 Mean Std. Dev. n % Predicted 72.5385 4.7891 13 2.71 72.5385

Figure 3 Branch 2a — physically active moderates

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Vol. 7, 3/4, 176–193

Tourism and Hospitality Research

185

Using decision trees

Support for Sustainable Tourism Development Node 0 Mean Std. Dev. n % Predicted 64.9823 6.8898 480 100.00 64.9823

−
Perception of Tourism’s Impact on the Community Adj. P-value=0.0000, F=24.7761, df=2,477

45 Node 3 Mean Std. Dev. n % Predicted 69.1667 7.2041 84 17.50 69.1667

+

−
Participated in Bird watching Adj. P-value=0.0002, F=13.9196, df=1,306

+

Yes Node 6 Mean Std. Dev. n % Predicted 66.7662 6.9242 77 16.04 66.7662

No Node 7 Mean Std.Dev. n % Predicted 63.8810 5.4871 231 48.13 63.8810

+

−
Gender Adj. P-value=0.0047, F=10.7891, df=1,229

Male Node 14 Mean Std. Dev. n % Predicted 62.9275 5.4083 138 28.75 62.9275

Female, Node 15 Mean Std. Dev. n % Predicted 65.2957 5.3223 93 19.38 65.2957

−
Prespective of Tourism Adj. P-value=0.0501, F=9.4885, df=1,136

Offical;Business, Node 22 Mean Std. Dev. n % Predicted 60.6111 5.9822 36 7.50 60.6111

Resident;Visitor Node 23 Mean Std.Dev. n % Predicted 63.7451 4.9686 102 21.25 63.7451

Figure 4 Branch 2b — passively active moderates

186

Tourism and Hospitality Research

Vol. 7, 3/4, 176–193

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Byrd and Gustke

Support for Sustainable Tourism Development Node 0 Mean Std. Dev. n % Predicted 64.9823 6.8898 480 100.00 64.9823

−
Perception of Tourism’s Impact on the Community Adj. P-value=0.0000, F=24.7761, df=2,477

45 Node 3 Mean Std. Dev. n % Predicted 69.1667 7.2041 84 17.50 69.1667

+

+

−
Participated in Bird watching Adj. P-value=0.0080, F=7.3903, df=1,82

Yes Node 8 Mean Std. Dev. n % Predicted 72.8500 5.4219 20 4.17 72.8500

No Node 9 Mean Std. Dev. n % Predicted 68.0156 7.3387 64 13.33 68.0156

−
Attended a festival Adj. P-value=0.0341, F=4.6958, df=1,62

Yes Node 16 Mean Std. Dev. n % Predicted 69.3182 6.8396 44 9.17 69.3182

No Node 17 Mean Std. Dev. n % Predicted 65.1500 7.7546 20 4.17 65.1500

Figure 5 Branch 3 — active supporters of sustainable tourism

Those variables are: perception of existing tourism development impact in their community, participation in bird watching, gender, and tourism perspective. Using this information, it can be concluded that the moderate group that will be least supportive to sustainable tourism development are individuals who perceive tourism as having a moderate impact on the community, do not bird watch, are male, and are either a business owner or government official.
Group 3 (Branch 3) — Active supporters The third group, active supporters, can be segmented based on their participation in bird

watching (p = 0.008) represented by nodes 8 and 9 (see Figure 5). Bird watchers are more supportive (72.85) than nonbird watchers (68.02). Stakeholders in this group who do not bird watch can be divided by their attendance at festivals (p = 0.034) represented by nodes 16 and 17. Festival attendees are more supportive (69.32) than nonattendees (65.15). Based on this sub-branch of the decision tree, three variables are identified to influence a stakeholder’s support for sustainable tourism development. Those variables are: perception of existing tourism development impact in their community, participation in bird watching, and festival attendance. Using this information, it can be

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Vol. 7, 3/4, 176–193

Tourism and Hospitality Research

187

Using decision trees

concluded that the group that will be most supportive of sustainable tourism development in a community are individuals who perceive a highly positive impact of existing tourism development on their community and bird watch.
CONCLUSIONS The purpose of this research paper is to explore the use of decision tree analysis in the identification of stakeholders in a community. Four groups were identified on the Support for Sustainable Tourism Development Decision Tree. These four groups were described as local worriers, physically active moderates, passively active moderates, and active supporters. Each group reported some level of support for sustainable tourism development. The key to including stakeholders in the tourism planning process is to understand their interests. The indication that stakeholders’ perception of impact influences their support for sustainable tourism development verifies the results of Perdue et al. (1990) and Vincent and Thompson (2002). Based on these results, it is imperative to understand stakeholders’ perception of tourism impact in order to understand their level of support for sustainable tourism development. A more positive perception of impact indicates a more supportive stakeholder. General demographic characteristics, more specifically gender, and perspective were shown to have an influence on the stakeholders’ level of support for sustainable tourism development. This supports the findings of McFarlane et al. (2000) and Stein et al. (1999). Stakeholders who are female are more supportive than male stakeholders and stakeholders who are either residents or tourists are more supportive than the business owners and government officials. The study reported in this paper found that a stakeholder’s participation in recreational activities influenced their support for sustainable tourism. Stakeholders who participate in recreational activities such as bird watching, gardening, and hiking are more supportive of sustainable

tourism development than those who do not participate in recreational activities. This supports the findings by McFarlane et al. (2000).
Support for sustainable tourism stakeholders To develop support for a sustainable tourism policy, the planners should be informed and educate the stakeholders about the development. The results imply that if the stakeholders perceive that a tourism development plan is sustainable they will generally support the plan and the development. If, however, they are left to try to interpret the tourism plan on their own, misunderstandings may occur resulting in the stakeholders perceiving that the tourism development is not sustainable. If stakeholders perceive that tourism development is not sustainable, they may take actions, legal and illegal, to stop or hinder the policy, plan, or development as indicated by Markwick (2000) and Ioannides (1995). The stakeholders who were identified as local worriers were low in their support for sustainable tourism development and perceived tourism had a negative impact on the community. For the development to precede sustainability and be successful this group should be identified and the reasons behind their perceptions of negative impacts understood, addressed and if possible resolved. One of the possible reasons for their negative perception could be a general lack of knowledge about the tourism industry. There are indications from the results that local worriers may not be familiar with the tourism industry in their county, for example at least 50 per cent of the respondents had never visited the local visitor’s center. Unfamiliarity and a lack of awareness by local worriers may result in a lack of support in the community. Awareness should increase their understanding, and thus their level of support, which will allow for tourism development to be more sustainable. Local worriers should also have their interests included in any discussion on future tourism development in the county,

188

Tourism and Hospitality Research

Vol. 7, 3/4, 176–193

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Byrd and Gustke

to address any future negative impacts that might be anticipated by this group. Moderates were moderate in their support for sustainable tourism development.They were neutral, to moderately positive in their perceptions of tourism impact on the community. Moderates are the largest group identified in this study, representing 64 per cent of the respondents. Moderates are a very important stakeholder group in the community because of their numbers. Moderates should be kept aware of all the tourism issues in the community and informed about the direct and indirect benefits received from the tourism industry in the county. Facilitating an awareness of tourism among moderates will assist in maintaining their level of support and will allow for the moderates to have an avenue to communicate their concerns to tourism planners. The moderates were divided into two main sub-groups, physically active moderates and passively active moderates. The physically active moderates are the most supportive of the two moderate groups. Physically active moderates can be identified and described mainly by their recreational activities. This group would be concerned about how current and new tourism development would impact their recreational activities in the community. For example, would the proposed development of tourism reduce their opportunities to participate in hiking or bird watching? The passively active moderates are less supportive of sustainable tourism development in their community. Passively active moderates can be identified and described by their demographic characteristics (gender, perspective) in the community.These individuals are concerned with tourism’s impact to the community on a social or economic level. If the development would be beneficial economically or socially to the community then they would be more likely to support the development. Active supporters indicated high levels of support for sustainable tourism development and perceived that tourism had a positive impact on the community. This group had

strong positive ideas about tourism and would generally support sustainable tourism development. Active supporters’ positive ideas and beliefs towards tourism may be reflected in their participation in at least one recreational activity during 2002. Slightly over 76 per cent (76.19 per cent) of the respondents in this group participated in at least one recreational activity in 2002. The respondents’ participation in recreational activities implies that active supporters are active and many participate in either natural resource-based activities such as bird watching or tourism activities such as attending a festival. The respondents’ participation in recreation activities indicates that these stakeholders enjoyed some of the activities and resources that can be described as tourism related. Based on this participation, active supporters may be more aware than the other two groups of the positive impacts that tourism had on the county.
Summary of support for sustainable tourism Understanding stakeholder support can assist tourism planners and policy developers in identifying stakeholder groups and their interests. An individual’s perception of tourism’s impacts was indicated to be the strongest predicator of an individuals support for sustainable tourism development. The more positive the impacts that the respondent perceived tourism had on the community, the more supportive they were to sustainable tourism.Therefore, there is a need to insure that stakeholders have an understanding of the actual impacts, both positive and negative, tourism has on a community. This understanding can be accomplished through education and information initiatives designed to facilitate awareness among the different groups about the positive and negative impacts, as well as, the value of tourism. As previously indicated, the planner must understand the stakeholders’ perceptions before education initiatives can be started. Also, if they have valid reasons for their negative perceptions, the reasons should be addressed.

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Vol. 7, 3/4, 176–193

Tourism and Hospitality Research

189

Using decision trees

Differing perceptions of tourism’s impact requires communication between all community stakeholders: residents, business owners, tourists, and government officials.This communication must be two ways. Stakeholders need to be informed about the positive and negative impacts tourism has on the community and ways of controlling or reducing the negative impacts. This will give the stakeholders a better understanding of tourism’s role in the community, which should increase their understanding of the impacts, positive and negative, that tourism has on the community. By doing this, tourism development can be conducted in a more sustainable manner. If the stakeholders are not involved in the tourism development, there is a higher chance that the development would fail.
Discussion of decision tree application The use of decision trees in tourism is a new way to approach many concepts that are integral to tourism such as marketing, planning, and creating and maintaining quality of life. Decision trees have been used in the past with patient research such as, Quality of Life in HIV/ AIDS Patients and Patient Satisfaction with Management Care (The Measurement Group, 2003), and marketing research such as consumer profiling (Eherler and Lehmann, n.d.). Decisions tree analysis can be used to identify stakeholder groups in a community. Unlike the traditional method of stakeholder mapping that relies simply on the researchers or planners’ knowledge and experiences, the decision tree analysis is grounded in statistical analysis and is a more robust tool for stakeholder identification and management. Decision trees are a parsimonious technique that can facilitate interpretation and understanding for tourism professionals about many tourism concepts and behaviors. The parsimonious nature of decision trees is based on the descriptive and visual nature of the output. Variable interaction can be visually shown to tourism professionals such as DMO executives, state travel directors and community planners,

which facilitate the explanation and application of the results of research. The use of decision trees to segment stakeholders has the four characteristics Kotler et al. (2003) states need to be present in segmentation. Decision trees use standard measures to develop the groups and each group is identifiable so that it can be measured by multiple variables including size. Each group identified by a decision tree should be accessible because the tree identifies multiple characteristics of each group. Each node on the decision tree can be used as a stakeholder group, therefore the size of the group is only dependent on the planners’ needs and resources. Finally, the decision tree eliminates much of the bias that exists in other forms of segmentation. Decision trees rely on data to split the sample based on specific variables. The results of this study can assist tourism planners with developing tourism in a manner such that stakeholder interests can be incorporated into the tourism plan. Identifying stakeholder groups is only one step in the sustainable development of tourism in an area. There are many other requirements that must be researched and addressed. Each of requirements needs to be fully addressed if true sustainable development is to be accomplished. Future research should focus on methods of identifying the efficient and effective methods for identifying and communicating with the stakeholder groups that are identified in a community.The communication to each group should be specific to that group. Clearly, stakeholder identification and segmentation using decision trees is not the panacea for solving stakeholder involvement; however, it is a more objective and systematic way of determining the interests of the groups and composition of the groups. The results of decision tree analysis can assist planners in developing more confidence about identifying stakeholders and understanding the interests that motivate them. Future research should focus on these motivations and how they are linked to the complex interactions of the environments, economic, and social elements of sustainable tourism.

190

Tourism and Hospitality Research

Vol. 7, 3/4, 176–193

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Byrd and Gustke

REFERENCES
Allen, L., Hafer, H., Long, P. and Perdue, R. (1993) ‘Rural Residents’ Attitudes Toward Recreation and Tourism Development’, Journal of Travel Research, 31, 4, 27–33. Andereck, K. L. and Vogt, C. A. (2000) ‘The Relationship Between Residents’ Attitudes Toward Tourism and Tourism Development Options’, Journal of Travel Research, 39, 27–36. Ansari, W. E. and Phillips, C. J. (2001) ‘Interprofessional Collaboration: A stakeholder approach to evaluation of voluntary participation in community partnerships’, Journal of Interprofessional Care, 15, 4, 351–368. Arimond, G. and Elfessi, A. (2001) ‘A Clustering Method for Categorical Data in Tourism Market Segmentation Research’, Journal of Travel Research, 39, 391–397. Arnstein, S. R. (1969) ‘A Ladder of Citizen Participation’, Journal of the American Institute of Planners, 35, 4, 216–224. Babbie, E. (1995) ‘The Practice of Social Research’, 7th edn, Wadsworth Publishing Company, Belmont. Beierle, T. C. (1998) ‘Public Participation in Environmental Decisions: An evaluation framework using social goals’, Resources for the Future, Washington, DC. Bramwell, B. and Sharman, A. (1999) ‘Collaboration in Local Tourism Policymaking’, Annals of Tourism Research, 26, 2, 392–415. Briassoulis, H. (2002) ‘Sustainable Tourism and the Question of the Commons’, Annals of Tourism Research, 29, 4, 1065–1085. Brunt, P. and Courtney, P. (1999) ‘Host Perceptions of Sociocultural Impacts’, Annals of Tourism Research, 26, 3, 493–515. Byrd, E. T. (1997) ‘Barriers to Rural Tourism: A comparison of the perceptions of the host community, local business owners, and tourists’, Unpublished master’s thesis, North Carolina State University, United States. Carmin, J., Darnall, N. and Mil-Homens, J. (2003) ‘Stakeholder Involvement in the Design of U.S.Voluntary Environmental Programs: Does sponsorship matter?’ Policy Studies Journal, 31, 4, 527–543. Carter, N. and Darlow, A. (1997) ‘Local Agenda 21 and Developers: Are we better equipped to build a consensus in the 1990s?’ Planning Practice and Research, 12, 1, 45–57.

Clarkson, M. B. E. (1995) ‘A Stakeholder Framework for Analyzing and Evaluation Corporate Social Performance’, The Academy of Management Review, 20, 1, 92–117. Cottrell, S. P. (2001) ‘A Dutch International Development Approach: Sustainable tourism development’, Parks and Recreation, 36, 9, 86–93. Crosby, N., Kelly, J. and Schaefer, P. (1986) ‘Citizens Panels: A new approach to citizen participation’, Public Management Forum, 46, 170–178. Curry, N. (2001) ‘Community Participation and Rural Policy: Representativeness in the development of Millennium Greens’, Journal of Environmental Planning and Management, 44, 4, 561–576. Davis, J. S. and Morais, D. B. (2004) ‘Factions and Enclaves: Small towns and socially unsustainable tourism development’, Journal of Travel Research, 43, 3–10. De Lopez, T. T (2001) ‘Stakeholder Management for Conservation Projects: A case study of Ream National Park, Cambodia’, Environmental Management, 28, 1, 47–60. De Oliveira, J. A. P. (2003) ‘Governmental Responses to Tourism Development: Three Brazilian case studies’, Tourism Management, 24, 97–110. Dillman, D. (2000) ‘Mail and Internet Surveys: The tailored design method’, 2nd edn, John Wiley and Sons, Inc, New York. Donaldson, T. and Preston, L. E. (1995) ‘The Stakeholder Theory of the Corporation: Concepts, evidence, and implications’, The Academy of Management Review, 20, 1, 65–91. Eherler, D. and Lehmann, T. (n.d.) ‘Responder Profiling with CHIAD and Dependency Analysis’, Institute of Applied Economic Research, Limburgs Universitair Centrum, http://www.luc.ac.be/iteo/articles/lehmann. pdf (accessed 30th May, 2003). Fiorino, D. J. (1990) ‘Citizen Participation and Environmental Risk: A survey of institutional mechanisms’, Science, Technology, and Human Values, 15, 2, 226–243. Freeman, R. E. (1984) ‘Strategic Management: A stakeholder approach’, Pitman, Boston. Goeldner, C. R. and Ritchie, J. R. B. (2002) ‘Tourism: Principles, practices, philosophies’, 9th edn, John Wiley and Sons, Inc, New Jersey. Gunn, C. A. (1994) ‘Tourism Planning: Basic concepts cases’, 3rd edn, Taylor & Francis: Washington, DC.

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Vol. 7, 3/4, 176–193

Tourism and Hospitality Research

191

Using decision trees

Hardy, A. L. and Beeton, R. J. S. (2001) ‘Sustainable Tourism or Maintainable Tourism: Managing resources for more than average outcomes’, Journal of Sustainable Tourism, 9, 3, 168–192. Hassan, S. S. (2000) ‘Determinants of Market Competitiveness in an Environmentally Sustainable Tourism Industry’, Journal of Travel Research, 38, 3, 239–245. Huba, G. J. (2003) ‘CHAID. The Measurement Group’, http://www.themeasurementgroup. com/Definitions/CHAID.htm (accessed 30th May, 2003). Hunt, L and Haider, W. (2001) ‘Fair and Effective Decision Making in Forest Management Planning’, Society of Natural Resources, 14, 873–887. Inskeep, E. (1991) ‘Tourism Planning: An integrated and sustainable development approach’, Van Nostrand Reinhold, New York. Ioannides, D. (1995) ‘A Flawed Implementation of Sustainable Tourism:The experience of Akamas, Cyprus’, Tourism Management, 16, 8, 583–592. Jamal, T. B. and Getz, D. (1995) ‘Collaboration Theory and Community Tourism Planning’, Annals of Tourism Research, 22, 1, 186–204. Johnson, G. and Scholes, K. (1999) ‘Exploring Corporate Strategy’, 5th edn, Prentice-Hall, Europe, London. Jones, T. M. (1995) ‘Instrumental Stakeholder Theory: A synthesis of ethics and economics’, The Academy of Management Review, 20, 2, 404–437. Kass, G. V. (1980) ‘An Exploratory Technique for Investigating Large Quantities of Categorical Data’, Applied Statistics, 29, 2, 119–127. Kavallinis, I. and Pizam,A. (1994) ‘The Environmental Impact of Tourism: Whose responsibility is it anyway? The Case Study of Mykonos’, Journal of Travel Research, 33, 2, 26–32. Koskela, T. (1999) ‘Decision Trees’, http://www.hut. fi/u/timoko/treeprogs.html (accessed 30th May, 2003). Kotler, P., Bowen, J. and Makens, J. (2003) ‘Marketing for Hospitality and Tourism’, 3rd edn, Prentice-Hall, Upper Saddle River, NJ. Lankford, S. V. (1994) ‘Attitudes and Perceptions Toward Tourism and Rural Regional Development’, Journal of Travel Research, 32, 3, 35–43. Lankford, S. V. and Howard, D. R. (1994) ‘Developing a Tourism Impact Attitude Scale’, Annals of Tourism Research, 21, 1, 121–139.

Long, P., Perdue, R. and Allen, L. (1990) ‘Rural Residents Perceptions and Attitudes by Community Level of Tourism’, Journal of Travel Research, 28, 3, 3–9. MacKay, K. J., Andereck, K. L. and Vogt, C. A. (2002) ‘Understanding Vacationing Motorist Niche Markets’, Journal of Travel Research, 40, 356–363. Maddox, R. N. (1985) ‘Measuring Satisfaction with Tourism’, Journal of Travel Research, 23, 3, 2–5. Markwick, M. C. (2000) ‘Golf Tourism Development, Stakeholders, Differing Discourses, and Alternative Agendas: The case of Malta’, Tourism Management, 21, 515–524. Martin, S. R. (1995) ‘Montanans’ Attitudes and Behavioral Intentions Toward Tourism: Implications for sustainability’, in McCool, S.F. and Watson, A.E. (eds), ‘Linking Tourism, The Environment, and Sustainability — Topical volume of compiled papers from a special session of the annual meeting of the National Recreation and Park Association’, Intermountain Research Station, Ogden, UT, pp. 69–76. Mason, P. and Cheyne, J. (2000) ‘Resident’s Attitudes to Proposed Tourism Development’, Annals of Tourism Research, 27, 2, 391–411. Mathieson, A. and Wall, G. (1982) ‘Tourism: Economic, physical, and social impacts’, John Wiley and Sons, Inc, New York. McFarlane, B. L. and Boxall, P. C. (2000) ‘Factors Influencing Forest Values and Attitudes of Two Stakeholder Groups: The case of the Foothills Model Forest, Alberta, Canada’, Society and Natural Resources, 13, 649–661. The Measurement Group. (2003) ‘Data Mining and Statistical Modeling’, http://www.themeasurementgroup.com/datamining.htm (accessed 15th June, 2003). Murphy, P. E. (1983) ‘Perceptions and Attitudes of Decision Making Groups in Tourism Centers’, Journal of Travel Research, 21, 3, 8–12. Murphy, P. E. (1985) ‘Tourism: A Community Approach’, Methuen, London. Owen, J. T. (1998) ‘Market Segmentation in Hospitality Research: No longer a sequential process’, International Journal of Contemporary Hospitality Management, 10, 7, 289–296. Perdue, R., Long, P. and Allen, L. (1990) ‘Resident Support for Sustainable Tourism development’, Annals of Tourism Research, 17, 586–599.

192

Tourism and Hospitality Research

Vol. 7, 3/4, 176–193

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Byrd and Gustke

Pizam, A. (1978) ‘Tourism’s Impacts: The social costs to the destination community as perceived by its residents’, Journal of Travel Research, 16, 4, 8–12. Pizam, A., Uriely, N. and Reichel, A. (2000) ‘The Intensity of Tourist-Host Social Relationship and its Effects on Satisfaction and Change of Attitudes: The case of working tourist in Israel’, Tourism Management, 21, 395–406. Reed, M. G. (1997) ‘Power Relations and Community Based Planning’, Annals of Tourism Research, 24, 3, 566–591. Robson, J and Robson, I. (1996) ‘From Shareholders to Stakeholder: Critical issues for tourism marketers’, Tourism Management, 17, 7, 533–540. Ryan, C. (2002) ‘Equity, Management, Power Sharing and Sustainability — Issues of the ‘New Tourism’’, Tourism Management, 23, 17–26. Sautter, E. T. and Leisen, B. (1999) ‘Managing Stakeholders: A tourism planning model’, Annuals of Tourism Research, 26, 2, 312–328. Simrell King, C. and Feltey, K. M. (1998) ‘The Question of Participation: Toward authentic public participation in public administration’, Public Administration Review, 58, 4, 317–326. Southgate, C. and Sharpley, R. (2002) ‘Tourism, Development and the Environment’, in Sharpley, R. and Telfer, D.J. (eds), ‘Tourism and Development: Concepts and issues’, Channel View Publications, Cleveland, pp. 231–262.

SPSS. (2002) ‘Introduction to Answer Tree’, SPSS, Chicago, IL. SPSS. (2003) ‘How Decision Tree Results are Different in Answer Tree 3.1/3.0’, http:// www.spss.com/spssbi/answertree/decisiontrees.htm (accessed 30th May, 2003). Steelman, T. A. (2001) ‘Elite and Participatory Policymaking: Finding balance in a case of national forest planning’, Policy Studies Journal, 29, 1, 71–89. Stein, T. V., Anderson, D. H. and Kelly, T. (1999) ‘Using Stakeholders’ Values to Apply Ecosystem Management in an Upper Midwest Landscape’, Environmental Management, 24, 3, 399–413. Stoney, C. and Winstanley, D. (2001) ‘Stakeholding: Confusion or Utopia? Mapping the Conceptual Terrain’, Journal of Management Studies, 38, 5, 603–626. Vincent,V. C. and Thompson, W. T. (2002) ‘Assessing Community Support and Sustainability for Ecotourism Development’, Journal of Travel Research, 41, 153–160. Williams, J. and Lawson, R. (2001) ‘Community Issues and Resident Opinions of Tourism’, Annals of Tourism Research, 28, 2, 269–290. Yuksel, F., Bramwell, B. and Yuksel, A. (1999) ‘Stakeholder Interviews and Tourism Planning at Pamukkale, Turkey’, Tourism Management, 20, 351–360.

© 2007 Palgrave Macmillan Ltd. 1467-3584 $30.00

Vol. 7, 3/4, 176–193

Tourism and Hospitality Research

193