Knowledge Linkage

Abstract

Research subjects are indicative of problem areas amongst which knowledge traffic occurs and networks form in a scientific community. This article presents a social network analysis of tourism dissertation subjects based on ProQuest Dissertations and Theses–Full Text database (1994-2008). The study suggests an openness and vibrancy of tourism as a domain of knowledge from a social network perspective. Longitudinal examinations revealed a structural change in its knowledge development. Twenty-one subjects were identified as subdomains in this dynamic and interconnected knowledge system. The article also discusses the relationships between subject areas and doctoral program distributions for tourism research at North American institutions. Results of the study contribute to discussions on scientific community and knowledge networks in interdisciplinary tourism studies.

Keywords

knowledge network scientific community social network analysis tourism dissertation

The linkages of research subjects within or across knowledge domains have been topics of interest for academics in a field. A typical and useful approach to the scrutiny of such connections is to look at subject clustering by means of social network analysis (SNA), through which knowledge traffic and subject networks can be mapped and described. Informed by social network and actor–network theories (Granovetter, 1973; Latour, 1987), such perspectives shed light on the formation and structure of knowledge networks in a field and hence contribute to a better understanding of its knowledge development.

A multidisciplinary field of social science, tourism has witnessed a high and sustained enthusiasm amongst its academics in assessing the state of the art of its research. Over the years, such endeavors are undertaken from various perspectives, for example, authorship analysis (Sheldon, 1991), perceptions of journal qualities by the publishing faculty (Pechlaner, Zehrer, Matzler, & Abfalter, 2004), examinations of methodologies for or methodological innovations in tourism research (Riley & Love, 2000), as well as documentation of changes in its research subjects (Swain, Brent, & Long, 1998; Xiao & Smith, 2006). Longitudinal observations on subjects and research techniques have also been made in the field of hospitality (Baloglu & Assante, 1999; Crawford-Welch & McCleary, 1992). In addition, there are reports on the ranking of academics, journals, and programs or institutions in tourism and hospitality based on research productivity (Jamal, Smith, & Watson, 2008; Jogaratnam, Chon, McCleary, Mena, & Yoo, 2005; Severt, Tesone, Bottorff, & Carpenter, 2009; Zhao & Ritchie, 2007), as well as citation analysis of research impacts (Kim, Savage, Howey, & van Hoof, 2009); such discussions have aroused debates amongst academics in these fields (McKercher, 2005). More recently, there have been increasing concerns about the future of tourism studies or new modes of enquiries. Notably, the frontiers of its knowledge development have demonstrated a shift in thoughts toward critical tourism studies or new tourism research (Ateljevic, Morgan, & Pritchard, 2007; Coles, Hall, & Duval, 2006; Tribe, 2005; Tribe & Xiao, 2011).

Among the various perspectives, the clustering and change of research subjects over time through prominent texts such as academic journals and doctoral dissertations present a unique lens for appraising the state-of-the-art and projecting knowledge trends. The purpose of this study is to examine the structure and evolution of tourism knowledge through a SNA of the research subjects identified from tourism doctoral dissertations selected from the ProQuest Dissertations and Theses–Full Text database (1994-2008). The connections between subjects were documented according to their co-appearances in each dissertation and presented in subject correlation matrices for network analysis. Results of the study shed light on the structure of dissertation subjects and on the evolution of tourism as a knowledge domain.

A Social Network Analysis of Dissertation Subjects

Knowledge networks are special types of social networks formed on the premises of research interests or problem areas within a scientific community (Collins, 1974). Often seen in the forms of research groups, task forces, or project teams bound together by research subjects or areas of interest, these networks may change over time as subject areas cluster or break up in the growth of a field. In tourism, the emergence of event studies (or event management) and mobilities (Coles et al., 2006; Coles & Hall, 2006; Urry, 2004) are examples of such evolutions in this domain of knowledge.

Additionally, knowledge networks are characteristic of social connections in varying density and strength in an invisible college (Crane, 1972). In a seminal paper on social networks, Granovetter (1973) refers to these connections as ties with differing strength because of variation in the amount of time, emotional intensity, intimacy, and reciprocity developed around a tie. With respect to diffusions of knowledge, he stresses the cohesive power of weak ties (i.e., nodes connected by fewer links over longer distances) in transmitting information over distances and between groups. Arguably, the tracing of weak ties could define a large network, albeit in lower density, than the tracking of strong and dense ties. With respect to subject clustering in a research field, weak ties are therefore suggestive of large and loose networks whereas strong links correspond to small and intense circles.

Moreover, in the context of power of knowledge (Foucault, 1980), the formations of networks in a scientific community are conducive to interpretations from actor–network perspectives. Anchored within the sociology of knowledge, this body of work was developed by Latour (1986, 1987), Law (1999), and Callon (1986), among others. A constructivist approach by nature, actor–network theory proposes that knowledge is a social product contributing to the examination of power actors and collectives in the process of its creation and utilization (Callon, 1986). Arguably, theoretical observations as such are of particular relevance to the present discussion. Presumably, multidisciplinary by nature, tourism research is characteristic of both strong and weak ties observable from social and actor networks charted by subjects or problem areas.

Research Subjects in Tourism

Subjects or problem areas are often central to state-of-the-art analyses in a field where academic journals and doctoral theses are frequently selected as texts for such exercises. First, in journal subject analysis, Xiao and Smith (2006) present a historical account of published subjects in Annals of Tourism Research (1973-2003) according to their frequencies of subject appearance in the journal. Based on periodic frequency over the first 30-year span of the journal, the rising trends are represented by subjects such as typology of tourists, community and development, alternative tourism experience, social–cultural change, geopolitical region and focus, theory and method, marketing and management, and environment whereas the decline patterns were found in subject areas such as economics, industry, transportation, hospitality, recreation, and the definition of tourism. More recently, newly emerging subjects such as consumer culture, discourse, equity, ethics, governance, mobility, networks, partnership, politics of representation, power, social capital, social tourism, and virtual tourism are likely to represent trends or new problem areas in tourism research (Ballantyne, Packer, & Axelsen, 2009).

Although these patterns could be reflective of the policies and orientations of the journal, such trends were confirmed by Ballantyne et al. (2009) in their recent update based on 12 tourism journals. Notably, the changing foci are represented by increased research on tourists and tourist experiences, a decline in economic and hospitality studies, a rise in marketing and management topics, a gradual erosion of North America dominance, increasing contributions from Australia, New Zealand, and Asia, and an emergence of the interpretive paradigm in certain subject areas (Ballantyne et al., 2009, p. 151).

In terms of tourism and hospitality research, Hu and Racherla (2008) and Racherla and Hu (2010) noted that these domains of knowledge have experienced rapid growth with formation of knowledge networks around major subject areas, and that researchers are linked to other communities, such as marketing and information technology. Based on four hospitality journals, the authors reported nine research streams or subject areas to which published authors were categorized; these include marketing and sales, consumer behavior, finance and accounting, human resources, information technology, customer service and operations, food and beverages, industry studies and education, and strategic management and performance studies (Hu & Racherla, 2008, p. 305). Through an author-by-stream tabulation, they found a pattern characteristic of a power law distribution of the knowledge networks in hospitality.

Second, doctoral dissertations are often regarded as the front edge of research or model of scholarship in a field. In tourism, the growing number of dissertations has long been of interest to its academics and has consequently made it a topic for periodic monitoring and assessments (Jafari & Aaser, 1988). Pioneer studies based on Dissertation Abstracts International can be traced back to the 1970s, for example, Crichton’s (1978) examination of 122 dissertations related to travel, recreation and leisure from 1974 to 1977, and a subsequent undertaking by Pizam and Chacko (1982) of 65 dissertations from 1976 to 1987. Departing from these attempts, Jafari and Aaser (1988) scrutinized a longer period of 1951-1987. Their analysis of 157 dissertations revealed a growing recognition of tourism in the multidisciplinary scientific community, particularly in economics, anthropology, geography, and recreation. Subsequently, an update was reported by Meyer-Arendt and Justice (2002), who mainly replicated Jafari and Aaser’s methodology for the period of 1987-2000, with an analysis of 377 dissertations. Their study documented the numerical, temporal, disciplinary (topical), and institutional features in the production of tourism dissertations in North America.

To a large extent, these prior studies were either from an epistemic/disciplinary perspective, or on an institutional standpoint in assessing tourism-related dissertations. Although the multidisciplinary nature of tourism was noted in the above endeavors, the clustering of research subjects and knowledge linkages in the field’s dissertation documents are yet to be examined and reported. As such, this study addresses knowledge linkages from a network perspective through examining research subjects or problem areas in tourism dissertations; observations are discussed from the perspectives of knowledge networks and scientific communities.

Social Network Analyses of Knowledge Linkages

SNA is documented as a useful approach to the description and interpretation of knowledge linkages through connections of research subjects in a field (Scott, 2000). Methodically, such linkages have been subject to SNA for structural interpretations of research collaborations and subject clustering through integrating data on individual attributes with data on interpersonal relations (Liebowitz, 2005; Schonstrom, 2005).

Notably, SNA has increased in popularity in the 1990s as an analytical framework (Scott, Baggio, & Cooper, 2008). It originally evolves from field theory in physics, graph theory in mathematics, and organizational field work in anthropology in 1950s and 1960s (Kilduff & Tsai, 2003). By collecting and organizing relational data into matrices and calculating parameters such as density and centrality, SNA is developed to accurately measure and represent the structure of relations among entities of interest and to explain why relations occur and what their consequences are in a network (Knoke & Yang, 2008). According to Scott et al. (2008), two basic characteristics differentiate SNA from other methods. First, SNA emphasizes on the relations between the actors rather than the attributes of the actors. Second, SNA focuses on the patterns of interactions rather than isolated individual actions. Durland and Fredericks (2005) summarize three major factors for the increased interest in SNA research and application. First, new understandings of relations and interactions have been gained by practical applications. Second, increased knowledge on complexity and the systems approach has led to a greater interest in SNA applications. In addition, the availability of software programs for analyzing data and generating sociograms has technically facilitated its growth.

In SNA, a knowledge domain is often known as a field of scrutiny characteristic of interrelated subject/problem areas (Hu & Racherla, 2008). Over the years, there has been a shift in the knowledge paradigm from an organization and management perspective (Fuller, 2000; Geuna, Salter, & Steinmueller, 2003; Trochim, Marcus, Masse, Moser, & Weld, 2008; Wixted & Holbrook, 2008) to a network approach based on scientific collaborations (Klenk, Hickey, Maclellan, Gonzales, & Cardille, 2009). A key factor in knowledge creation and sharing, social networks have received growing interests from researchers (Lin, 2001; Tsai & Ghoshal, 1998). In such a context, a SNA undertaking can focus on individuals, research projects, universities, journals, or other knowledge repositories (Contractor, Wasserman, & Faust, 2006, p. 682) while relationships (or ties) between the actors may be operationalized as any kind of communication, information exchange, social interaction, publishing, or citation (Klenk et al., 2009). Specifically, SNA examines the entire structure of a network including its subgroups and constituents (Knoke & Kuklinski, 1982). As such, this approach is particularly appropriate for a scrutiny of subject linkages as it allows researchers to visualize knowledge networks through a textured (re)presentation of the making of research in a field.

Because of its tactic nature, interactions through social networks are essential to the creation and sharing of knowledge (Cross, Liedtka, & Weiss, 2005). As argued by Hu and Racherla (2008), strong relationships, interactions, and dialogue among people with varying backgrounds, resources, predispositions, and insights are critical for supporting and sustaining a knowledge domain. Similarly, Melin (2000) and Wagner and Leydesdoff (2005) have noted the contribution of research collaborations to the growth of a domain through facilitation of knowledge sharing among its researchers. One approach in this regard is coauthorship analysis. For example, by observing and visualizing coauthorship networks in publications, it is possible to examine the structure and identify key players in the knowledge networks within a research community (Hu & Racherla, 2008; Newman, 2001, 2004). Co-citation analysis is another approach to exploring knowledge networks in a scientific community. Building on links from the sources of reference, co-citation analysis has been used to examine how diffusions of knowledge can lead to the growth of a domain (Horn, Finholt, Birnholtz, Motwani, & Jayaraman, 2004; Lin, 1995), or conversely, to explore how citation distortions can create unfounded authorities in a scientific community (Greenberg, 2009).

The purpose of this study is to examine the making of tourism research and the evolution of its knowledge through a SNA of the research subjects identified in doctoral dissertations in North American universities over the past 15 years. Five questions are addressed: What is the structure of knowledge networks identifiable from dissertation subjects? How has the structure of knowledge networks evolved over time? What are the major subject areas in this domain of knowledge? What structural changes have resulted from periodic interactions between the major subject areas? What are the associations between time, institutions, and major subjects in dissertation research? To fulfill the above, a SNA is conducted on the subject headwords of tourism dissertations selected from the ProQuest Dissertations and Theses–Full Text database (1994-2008).

Social Network Analyses of Dissertation Subjects: An Application

Data collection

Methodologically, this study followed the steps of Jafari and Aaser (1988) and Meyer-Arendt and Justice (2002). However, instead of Dissertation Abstract International, the search was performed from Dissertations and Theses-Full Text, as the latter is a more comprehensive data source for subject analyses. Overall, it includes more than 2 million dissertation and thesis citations from around the world since 1861, and it contains citations of doctoral dissertations from most accredited North American universities in all academic disciplines (ProQuest, 2009a).

The ProQuest thesaurus, also known as the “subject list” or “controlled vocabulary,” is a list of subject terms or headwords used to index documents and to classify and organize information contained in ProQuest collections (ProQuest, 2009b). In the ProQuest database, a number of subject terms are selected by dissertation authors from the ProQuest thesaurus as “descriptors” to classify their dissertation subjects. Because of its multidisciplinary nature and the scope of its related subjects, a variety of headwords could be relevant descriptors of a tourism dissertation. Subject headwords such as “airline, attraction, aviation, destination, events, hospitality, leisure, parks, and recreation” may all relate to tourism in one way or another. In their seminal work on this topic, Jafari and Aaser (1988, p. 408) discussed extensively the pros and cons of “casting wide nets” in such a search. Furthermore, despite their maintaining a reasonable level of overlaps or linkages, tourism, hospitality, leisure, parks, and recreation are sisters’ fields, and they have gradually developed silos among themselves in their evolution.

To perform a focused search for this analysis, the following five keywords were used: “travel,” “traveler,” “tourism,” “tourist,” and “ecotourism” (and their plurals where applicable). The “ecotourism” headword was added to Jafari and Aaser’s (1988) search terms because of its relatively late appearance as a “keyword” of tourism dissertations. With a span of 15 years (1994-2008), 1,619 results were initially returned. To minimize the chance of missing entries, another round of complementary data search was conducted with the same keywords respecified as document title, and 1,901 results were found. Overlaps in the two search results were removed; further examination of dissertation abstracts also eliminated those nontourism-focused theses (e.g., “travel” as a keyword was also used in many dissertations on transportation or civil engineering). Eventually, 816 entries were identified as tourism-focused doctoral dissertations completed during 1994 to 2008. For each entry, information was collected on dissertation title, author, affiliated institution, year of completion, and subject terms. Because of the absence of needed subject terms specified above, four entries were removed from the final set. Therefore, for this study, a total of 304 terms were selected as identifiers or subject indices from 812 tourism dissertations included in the subsequent SNA. Notably, these iterations are hoped to result in the type of data validity and reliability needed for a longitudinal observation.

Data matrix

A series of data matrix constructions and transformations were carried out in the data analysis. First, a binary dissertation-by-subject table matrix was constructed, with dissertation in rows and subject in columns. In this matrix, x_ij = 1 if the ith dissertation entry used the jth subject, or x_ij = 0 otherwise. The dissertation-by-subject matrix was then transposed into a new subject-by-dissertation matrix, where the rows of the original matrix were then turned into columns and the original columns into rows. In the third step, a new subject-by-subject data matrix was built through multiplying the transposed data matrix X′ by the original data matrix X. The product (X′X) of these two matrices contained information on the relationships between each pair of the 304 dissertation subjects. In this subject-by-subject matrix, the value in the ijth (when i ≠ j) cell indicates how many times the ith and the jth subjects co-appeared in the selected dissertations. The diagonal values of the matrix represent the characteristics of each subject on its own, which, in this case, refers to the total number of appearances a given subject has had in this data set. To examine how the relationships among subjects changed over time, five 3-year subject-by-subject data matrices for 1994-1996, 1997-1999, 2000-2002, 2003-2005, and 2006-2008 periods were constructed using the same procedure.

Data analysis

Data analysis consisted of five major phases. First, descriptive overviews were presented on the production of tourism dissertations in North America from 1994 to 2008; the use of subjects for these dissertations was also reviewed. Second, the relationships among subjects were visualized through the program of Netdraw, and a series of mathematical network measures were conducted using UCINET (Borgatti, Everett, & Freeman, 2002) for a structural understanding of how dissertation subjects were connected to one another. Third, a temporal factor was introduced by breaking down the overall subject network into five 3-year periodical networks, and the longitudinal structural change of subject networks was examined. Fourth, a set of key players in the subject networks was identified for further examinations on their ego-network structures as well as their roles within the entire subject network. A reduced network was also constructed for these identified key subjects where analyses were run to understand how they were associated to each other over time in tourism dissertation research. At last, an institutional factor was introduced to understand tourism dissertation research at different doctoral programs.

A Structure of Dissertation Subject Networks

Descriptive overview

It is found that annual production of tourism dissertations in North America has experienced a nearly 300% increase from 1994 (n = 31) to 2008 (n = 90). The number of subjects for each dissertation ranged from 1 to 10, with an average of 3 subjects per thesis. Among these dissertations, about 15% had only one subject, 26% had two, and another 26% contained three subjects. More than 80% of the dissertations had four subjects, and only a small portion (about 10%) had more than five subjects. The average number of subjects per dissertation has changed over time. Gradually increasing from two subjects in 1994, the average number reached its peak in 2002 with more than four subjects per dissertation. Despite the exceptionally large number for 2005, the average has started to drop since 2002 back to the late 1990s level and reached its lowest point in 2008.

Subject network construction and visualization

Facilitated by Netdraw (Borgatti et al., 2002), the overall subject-by-subject data matrix can be visualized through a subject network (Figure 1), in which the nodes stand for subjects and the ties denote their connections. The lower left part of this diagram is an overview of the entire subject network. Recreation, the most connected node, sits at the center of the network. The isolated node lying in the upper left corner represents the subject of Physical Anthropology. Using a spring-embedding algorithm based on geodesic distance, the Netdraw program repositioned all the nodes to reveal the structure of the subject network. Therefore, the distance of one node to another in this diagram indicates the extent of connectedness of the subjects in tourism dissertation research. Several node clusters are noticeable by visually examining their network structures. For example, Cultural Anthropology, Sociology, Minority and Ethnic Groups, Geography, History, American Studies, American History, and Women Studies appear more connected among themselves than with other subjects. A similar structure is also found among subjects such as Tourism, Marketing, Management, Consumer Behavior, and Economics. Travel, Demographics, Personality, Transportation, Hotel and Motel, and Economic Impact form another cohesive group.

Figure 1

The Structure of the Overall Subject Network

Longitudinal network analysis

A possible way to examine how networks evolve over time is to develop a series of subject networks with a given interval of time and then compare their network structures. In this study, five 3-year periodical subject networks were constructed for longitudinal analyses. The structures of these periodical networks are depicted in Figure 2.

Figure 2

Three-Year Periodical Subject Networks

For each period, the number of completed dissertations, network size, number of ties among subjects, network density, and the betweenness and centralization of a network were measured and presented in Table 1. The periodical production of dissertations has been growing over the years, with a relatively steady increment ranging from low 20 to high 30.

Table 1

Longitudinal Subject Network Measures

	Time period
	Overall	1994-1996	1997-1999	2000-2002	2003-2005	2006-2008
No. of dissertations	812	101	130	168	188	225
Network size	304	56	86	171	164	131
No. of ties	3,650	290	586	1,552	1,584	852
Density	.0396	.0942	.0802	.0534	.0593	.05
Normalized density	.3573	.291	.3291	.3414	.3589	.2836
Betweenness centralization index (%)	52.75	67.63	64.63	49.77	48.67	65.48

Indicated by the number of nodes in a network, network size is a basic demographic measure in SNA. Compared with stable increment in the five periods, a nearly twofold increase in the size of the periodical subject networks was observed as it turned from the 1998-1999 period (size = 86) to the 2000-2002 period (size = 171). It indicates a substantial number of new subjects have emerged in tourism dissertation research since 2000. However, the growth did not last long as the network size started to drop after the 2000-2002 period, even though dissertation production still kept growing and reached its new height in 2006-2008.

A widely used notion in graph theory, network density measures the extent to which all possible relations in a network are actually present (Mitchell, 1969). Ranging from “0” (where nodes are isolated from one another) to “1” (where nodes are connected to one another), network density is calculated as the number of actual connections between nodes divided by the number of possible connections (Scott, 2000). As this analysis focused on the presence rather than the quality of connections among subjects, the previous values of subject networks were dichotomized for network density measurement. In general, a higher network density indicates a higher degree of connection among the subjects in tourism dissertations. The densities of all the periodical networks remained low; none of them has reached a density of 10% in terms of subject connections in their respective network. In this study, a low density may indicate the diversity in the subject areas of tourism research and may also imply the large potential to link to other knowledge domains in future tourism studies. Notably, density is very sensitive to network size (Borgatti et al., 2002). Following Smith’s (2007) method by taking the logarithm of network density with a base of “N/2” and adding “1,” a normalized version of network density was calculated for periodical networks to make them comparable. The results show that connections among subjects have been growing denser until the 2006-2008 period, when its subject network encountered a significant decrease in density and reached its lowest point for the past 15 years. It suggests a decline in the incorporation of subject areas in tourism dissertation studies from 2006-2008.

Betweenness centrality refers to the extent to which a particular point lies “between” the various other points in a network (Scott 2000) while betweenness centralization describes the betweenness existing in the entire network by calculating the ratio of the actual sum of betweeness centrality for each node to the maximum possible sum (Freeman, 1979). A high betweenness centralization score indicates a hierarchical network structure, where a single or a small number of nodes in the network tend to be more central than other nodes. The group betweenness centralization scores show that, overall, there is a substantial degree of concentration or centralization in the five periodical subject networks, as even the lowest score for the 2003-2005 network was around 50%. This indicates that certain subjects have been influential throughout the years. From 1994 to 2005, the betweenness centralization index kept decreasing, suggesting a growing diversity in the subject areas of tourism dissertations. However, this diversifying trend did not last long, as the betweenness centralization for the 2006-2008 period has risen to the 1990s level.

Structural correlations of periodical subject networks

To investigate the structural correlations among these periods, the periodical subject networks were reconstructed based on the complete list of 304 subjects. For those subjects that were actually absent in a certain period, their connections to others in the corresponding periodical data matrix was simply valued as “0.” The five reconstructed periodical subject networks were then joined into a single data file in UCINET for calculating their metric correlations (Table 2). With each network represented by a square spot, the structural similarities and dissimilarities among these periodical networks are presented in a two-dimensional space (Figure 3) based on a metric multidimensional scaling. In this scaled scatter plot of proximity/similarity, the closer the subject network spots, the more structural similarities these networks share, and vice versa. The five periodical network spots in the plot are connected with an arrow line in chronological order. It is noticed that the structural distinctions among the five periodical networks are mainly at the horizontal level.

Table 2

Structural Correlations Among Periodical Subject Networks

	Time period
	1994-1996	1997-1999	2000-2002	2003-2005	2006-2008
1994-1996	1	—	—	—	—
1997-1999	0.735	1	—	—	—
2000-2002	0.557	0.635	1	—	—
2003-2005	0.628	0.703	0.773	1	—
2006-2008	0.745	0.757	0.649	0.724	1

Figure 3

Multidimensional Scale of Periodical Subject Networks

Several implications can be derived through visually examining the spatial positions of these periodical networks. First, the five subject networks could be divided into two clusters based on their similarities in network structure. One cluster consists of the networks for 2000-2002 and 2003-2005 periods, and the other includes the networks for the rest of the three periods. This clustering indicates a relatively different pattern in the diffusion of knowledge for dissertation research from 2000 to 2005 compared with those completed in other time periods. Second, the structure of the subject network has encountered its most substantial variation when it evolved from the 1997-1999 period to the 2000-2002 period, suggesting that there has been a primary transition at the turn of the century in either research emphasis or subject patterns of tourism dissertations. Third, the evolution of network structure from 1994 to 2008 has not followed a static direction but has displayed a U-shape pattern. The turning point was in the 2000-2002 period, after which the network structure tended to evolve in the opposite direction. This “U-turn” in the evolution of subject network structure indicates that tourism dissertation research may be turning back to its old fashion as in the early 1990s.

Network reduction and key subject identification

The connections among subjects appeared to display a “core/periphery” structure in both the overall and periodical networks. “Core/periphery” is a structural notion in SNA. Borgatti and Everett (1999) described a core/periphery structure as

. . . consist[ing] of two classes of nodes, namely a cohesive subgroup (the core) in which actors are connected to each other in some maximal sense and a class of actors that are more loosely connected to the cohesive subgroup but lack any maximal cohesion with the core. (p. 377)

Using the UCINET 6.0 program, a core/periphery analysis was performed on both the overall and periodical subject networks. For each network, the program used genetic algorithm to fit a core/periphery model to the corresponding subject data matrix and classified each subject based on its membership to either the core or the periphery partition.

An ideal core/periphery partition will divide all the subjects in two groups in a way that the densities of connections among “core” subjects, between “core” and “periphery” subjects, and among “periphery” subjects vary significantly from each other. Table 3 provides the core/periphery measures for the overall and periodical subject networks. In SNA, there is no absolute benchmark for determining whether a network has a good or poor core/periphery model fit because of the variation of networks under study (Borgatti et al., 2002). Unlike social networks derived from a coherent group of actors, for which a relatively high core/periphery model fit is expected (e.g., “0.8”), the subject networks in this study were constructed in an open-ended manner by accumulating connections among dissertation subjects emerged in a study period, and therefore a model fit of around “0.4” is considered as acceptable (Borgatti et al., 2002). Notably, the overall subject network did not show a clear core/periphery structure (model fit = 0.126), although connection densities of the three subject partitions seemed to vary significantly. As to the subject networks for the 1997-1999, 2000-2002, and 2006-2008 periods, the core/periphery model fit was at an acceptable level, in which substantially distinctive densities of connections were found among different subject partitions, and a small number of them were identified as “core” subjects. However, the subject networks for the 1994-1996 and 2003-2005 periods have inadequate core/periphery model fits. The connection densities for different partitions in these two networks did not show a significant variation; more than half the subjects in these two networks were grouped into the “core” partition.

Table 3

Core/Periphery Model Measures for Periodical Subject Networks

	Density
Time period	Core partition	Core/periphery partition	Periphery partition	Model fit	No. of core subjects	Percentage of core subjects
Overall	.487	.039	.005	.126	112	36.8
1994-1996	.61	.075	.037	.22	27	48.2
1997-1999	2.086	.212	.04	.465	15	17.4
2000-2002	2.288	.223	.023	.394	18	10.5
2003-2005	.137	.016	.023	.041	128	78
2006-2008	1.385	.215	.031	.409	13	9.9

Based on the above analyses, 21 subjects have emerged as subdomains of knowledge in tourism dissertation research:

Agricultural Economics, American Studies, Area Planning and Development, Behavioral Science, Communication, Cultural Anthropology, Economics, Environmental Science, Geography, Management, Marketing, Minority and Ethnic Groups, Political Science, Public Administration, Recreation, Sociology, Social Psychology, Social Structure, Tourism, Urban Planning, and Women Studies.

Ego-network analysis of major subjects

Focusing on the structure of the entire network, it is hard to see, from previous studies, how and to what extent a specific subject was embedded in the overall subject network. To have a better understanding of the opportunities and constraints a subject has had over time for connections with other subjects (or subdomains) in tourism research, it is necessary to examine a subject’s “close neighborhood,” or its egocentric networks. By definition, an egocentric network consists of a focal actor (or ego), a set of alters having direct ties to the ego, and ties among these alters (Wasserman & Faust, 1994). In this study, ego-network analysis was performed on the 21 major subjects with measures presented in Table 4.

Table 4

Ego-Network Measures for Major Subjects in Overall Network

				Periodical ego-network size					Periodical ego-network tie changes
Subject	Overall size	Density	Normalized ego-betweeness	1994-1996	1997-1999	2000-2002	2003-2005	2006-2008	T1	T2	T3	T4	Mean	SD
Recreation	249	4.55	61.22	42	69	141	129	90	47	108	138	123	104	39.92
Tourism	157	7.59	38.72	0	7	87	87	34	6	84	110	89	72.25	45.58
Marketing	135	8.46	35.33	12	23	78	61	36	23	73	85	71	63	27.37
Economics	85	16.42	31.41	6	13	47	39	13	11	50	62	42	41.5	21.77
Management	75	14.38	35.12	8	9	27	42	18	11	30	49	48	34.5	17.93
Cultural anthropology	63	21.52	32.54	11	28	33	30	26	21	23	33	28	26.25	5.377
Urban planning	54	27.95	15.52	8	19	24	31	20	13	21	31	31	24	8.717
Area planning and development	53	28.16	15.48	8	19	24	31	18	13	21	31	29	23.5	8.225
Environmental science	51	24.16	28.05	7	16	31	28	6	13	31	31	24	24.75	8.5
Geography	45	26.36	26.79	22	13	21	20	9	17	20	21	19	19.25	1.707
American studies	42	28.34	21.76	11	18	16	19	18	17	18	19	17	17.75	0.95
Sociology	40	32.56	17.85	13	15	9	25	16	10	12	18	23	15.75	5.909
Social structure	38	43.24	10.63	11	16	22	17	8	15	16	23	17	17.75	3.59
Agricultural economics	35	28.75	26.9	5	8	19	12	6	9	21	25	12	16.74	7.5
Minority and ethnic groups	33	40.15	14.3	8	12	14	18	9	8	14	18	17	14.25	4.5
Political science	29	39.9	16.59	4	8	9	17	12	8	13	14	19	13.5	4.5
Women studies	22	38.53	24.01	0	11	11	9	11	10	10	12	14	11.5	1.914
Public administration	21	47.14	17	3	10	11	6	8	9	11	9	6	8.75	2.06
Social psychology	21	44.76	16.85	3	6	9	5	10	5	7	10	11	8.25	2.754
Behavioral science	21	39.56	72.88	4	3	6	5	8	3	5	3	9	5	2.828
Communication	17	45.59	20.45	3	9	0	11	3	8	8	10	10	9	1.155

Note: T1, T2, T3, T4 = Ego-network tie change from one to the next period (1994-1996, 1997-1999, 2000-2002, 2003-2005, and 2006-2008).

Ego-network size measures the total number of alter subjects directly connected to the ego subject during the past 15 years. The results suggest a dominance of Recreation (size = 249), Tourism (size = 157), and Marketing (size = 135) in dissertation research, as each of these ego subjects embraced more than half the subjects identified in this study. Connecting to at least 20% of the total subjects, Economics (size = 85), Management (size = 75), and Cultural Anthropology (size = 63) also had a prominent influence in the subject networks. Other important subjects include Urban Planning (size = 54), Area Planning and Development (size = 53), Environmental Science (size = 51), Geography (size = 45), American Studies (size = 42), and Sociology (size = 40).

Ego-network density measures the degree of cohesion among the subjects directly connected to a given focal subject through calculating the number of actual ties divided by the number of all potential ties, times “100” (Borgatti et al., 2002). Despite its sensitivity to network size, which renders it unreliable to compare densities among ego-networks with different sizes (Friedkin, 1981), valuable information can still be drawn from this measure. To some extent, a relatively higher density in this study suggests a more stable and enclosed knowledge system for doctoral tourism research on some subject areas (e.g., Public Administration, Communication, Social Psychology, Social Structure, Minority and Ethnic Groups, Political Science, Behavioral Science, and Women Studies) whereas a relatively lower density indicated a more open and vibrant knowledge setting for studies on subject areas, such as Recreation, Tourism, Marketing, Economics, and Management, where there were lots of potential for existing subjects to further collaborate with others.

Normalized ego-betweenness measures a focal subject’s potential to engage other subjects through calculating the extent to which an ego subject is a part of the relationships among the alter subjects. In this analysis, Recreation had the highest normalized ego-betweenness index, followed by Tourism, Marketing, Management, Cultural Anthropology, and Economics, suggesting a large potential for these subjects to engage other knowledge domains in tourism studies. The relatively lower ego-betweenness for Social Structure, Minority and Ethnic Groups, Social Psychology, Planning, Area Planning and Development, Political Science, Public Administration, and Sociology indicates that these subjects have a much flatter ego-network structure, where the alter subjects have a higher chance to be connected in tourism research even without the presence of an (the) ego.

The periodical ego-network sizes were also presented in this study through counting the number of alter subjects that an ego has had for periodical ego-networks. The majority of the 21 subjects have experienced a substantial increase in their ego-network size for the 6 years from 2000 to 2005, except for Geography whose ego-network size has been shrinking ever since the 1994-1996 period. Tourism and Women Studies did not show up in the first period either because these two subjects had not been included in ProQuest’s controlled vocabulary in those early years or because no relevant dissertation was found for that period. Communication was absent for the 2000-2002 period, indicating this dissertation subject was not classified during that period.

There could also have been other alter subjects involved in the periodic ego-networks. This overtime subject variation of the ego-networks could not be uncovered by simply looking at the periodical changes in ego-network size. Therefore, the change of ties in the periodical ego-networks were examined through summing up the number of old subject ties lost and the number of new subject ties emerged when an ego-network evolved from one period to another. The results showed that for the major subjects, the largest change in their ego-network ties occurred when their networks evolved from the 2000-2002 to the 2003-2005 period. This means 2000-2005 is the most active period for major subjects to incorporate different subdomains of knowledge into their corresponding networks. In addition, the average periodical ego-network tie change and its standard deviation (last two columns in Table 3) also provide information on how structurally stable the major subjects’ ego-networks have been over time. A higher mean of the periodical ego-network tie change indicates a less stable ego-network structure for a subject (e.g., Recreation, Tourism, Marketing, Economics, and Management), as there tend to be more periodical changes in alter subjects involved with a focal subject while a smaller mean of the ego-network tie changes denotes a more stable knowledge network for tourism studies in some subject areas (e.g., Behavioral Science, Social Psychology, Public Administration, and Communications).

Correspondence analysis of subject associations

To understand how and to what extent these major subjects have been associated to one another over the years, a correspondence analysis was conducted based on a two-way data table constructed with 758 dissertations by the 21 major subjects; 54 dissertations were excluded due to their absence of these subjects as identifier terms. In SNA, correspondence analysis is a useful tool for assessing the structural equivalence of the nodes, particularly in a two-mode network (Borgatti & Everett, 1997; Roberts, 2000).

Figure 4 presents the structural associations among the 21 major subjects (i.e., the column scores of the correspondence analysis). The first five dimensions accounted for 8.80%, 7.62%, 6.85%, 6.58%, and 6.29%, respectively of the total variance. In Figure 4, the first (horizontal) dimension tends to differentiate the subjects based on their applied/theoretic orientation in tourism research. The subjects on the left of the plot are more applied while those on the right tend to be more theoretical. A humanistic/nonhumanistic distinction among the 21 subjects is observed on the second (vertical) dimension. The subjects in the upper half of the plot are relatively more humanistic than those in the lower half, which focus more on nonhumanistic issues in tourism research.

Figure 4

Correspondence Analysis of Key Subjects

It is found that research subjects in tourism dissertations have clustered into a number of subdomains of knowledge. Recreation is very close to the origin point, indicating its weak dimensionality with other subjects. Located in the upper left quadrant, the first major cluster is economics- and business-oriented tourism subjects, including Agricultural Economics, Economics, Marketing, Management, Behavioral Science, Social Psychology, and Communication. Sitting across the border of the two left quadrants, the second cluster embraces theoretical and humanistic subjects such as Cultural Anthropology, Minority and Ethnic Groups, Sociology, and American Studies. The third cluster is found in the lower left quadrant, which consists of Political Science, Environmental Science, Public Administration, Urban Planning, and Area Planning and Development. It is not surprising to see an overlap of Urban Planning and Area Planning and Development, as they represent highly similar research areas in dissertation research. Geography, Social Structure, and Women Studies form their own clusters, suggestive of limited connections with other subjects and of relatively distinct areas in tourism dissertation research.

Associations between years and major subjects

By aggregating the connections, a correspondence analysis was conducted to examine the association between the five periods and the 21 major subjects. The distribution in Figure 5 shows how the appearance of subjects has changed over time. Notably, dissertations in the 1997-1999 and 2006-2008 periods shared more structural similarities than they had with those in other periods regarding their citation of major subjects. To some extent, the distribution of major subjects in tourism dissertations was periodically skewed. Compared with subjects lying within the five periodic square spots, those falling beyond the area (e.g., Political Science, Communication, and Behavior Science) tended to be more anecdotal or temporal with respect to their appearances in tourism dissertations. In addition, Geography was the most indexed subject among the 1994-1996 dissertations while Tourism as a subject headword appeared more often in 2000-2005 than in the previous periods, indicating a relatively recent “sovereignty” of PhD-in-tourism programs in North American institutions. The majority of American Studies, Women Studies, and Social Psychology studies in tourism dissertations were completed in the more recent period of 2006-2008.

Figure 5

Correspondence Analysis of Association Between Years and Subjects

Associations between institutions and major subjects

A total of 79 North American universities were identified for having produced at least three tourism doctoral dissertations during 1994-2008. The top 10 institutions were Michigan State (34), Texas A&M (34), Purdue (27), Clemson (25), Penn State (24), Illinois (22), Waterloo (17), Virginia Tech (17), University of Florida (14), Oklahoma State (11), and University of Texas at Austin (11).

Figure 6 presents the correspondence analysis on the association between degree granting institutions and the 21 major subjects. Three major clusters emerged from among the 79 universities based on areas of emphasis in their PhD programs. The first cluster, represented by UC Berkley, New York University, University of Chicago, Emory University, and University of Texas at Austin, is centered around sociological and anthropological approaches to tourism, particularly in relation to Minority and Ethnic Groups, Women Studies, and American Studies in their doctoral tourism research. Second, Clemson, Virginia Tech, Penn State, Taxes A&M, Purdue, Michigan State, and Oklahoma State formed another cluster with focuses on Recreation, Tourism, and Marketing. The third group had a relatively general and broad range of interests in subjects such as Urban and Area Planning, Geography, Environmental Science, Economics, Political Science, and Public Administration. Typical in offering such programs are University of Minnesota, Washington, Florida, Waterloo, and University of Western Ontario.

Figure 6

Correspondence Analysis of Association Between Institutions and Subjects

Knowledge Linkages Seen Through Tourism Dissertations

Research subjects are indicative of problem areas around which knowledge linkages or networks form in the growth of a scientific specialty (Ben-David, 1964; Mulkay, 1977; Mullins, 1972). This article presents a SNA of the structure of tourism research subjects and periodic shifts of its problem areas observable from doctoral dissertations derived from ProQuest Dissertations and Theses–Full Text (1994-2008).

As a domain of knowledge, tourism has been a popular subject for doctoral studies in North American institutions. The increase of dissertations over the past 15 years has been phenomenal, marking an exponential growth of its knowledge base from this source. There was also a notable increase by network size at the turn of the century, accompanied by the emergence of new subjects and knowledge linkages in dissertation research. Nonetheless, it is also notable that despite the growth of dissertation numbers, the network size of its research subjects has been on a slight decline in the period of 2003-2008. By linkages among research subjects, the consistent low density across the study periods (<0.1) denotes a potentially poor state of connection among the subject areas documented in North American tourism dissertation research.

As a multidisciplinary field, tourism has been expanding its knowledge base with more focused problem areas explored. Interestingly, the recent decline in network size and its current state of low density of subject connections may indicate a healthy state of its scientific community in terms of relatively stable problem areas and the level of openness to be potentially linked to other knowledge domains. Viewed historically, such variations in the range of subject areas are results of the knowledge traffic among its researchers, which is typical in the early growth of a field (Ben-David, 1964; Mullins, 1972).

From an evolution standpoint, the structural similarities (and dissimilarities) of subject networks in different periods provide a useful perspective on the change of foci in dissertation research. Collectively, based on similarities in network structures, the subject clusters for the periods of 2000-2003 and 2003-2005 were distinct from those in the other three periods. There has been a radical transition at the turn of the century in terms of dissertation subjects. Notably, the evolution of subject clusters displayed a U-shape pattern with a turning point in the 2000-2002 period when (or where) a retreat into the past has occurred in the network structure. In view of associations between years and subjects, the turn of the century (2000-2005) has witnessed knowledge linkages centered around Political Science, Cultural Anthropology, Environmental Science, Social Structure, Economics, and Minority and Ethnic Groups. From an actor–network perspective, results from the correspondence analysis of associations could be attributable to the relational effects created by the complex and intricate linkages resulting from a multiplicity of actors, such as dissertation researchers, programs, and institutions (Latour, 1986, 1987; Law, 1999), and to the growth of tourism as an inter-/postdisciplinary field of study beyond narrow boundaries at an institutional level (Coles, et al., 2006; Tribe, 2005; Tribe & Xiao, 2011).

The core/periphery analysis generates a list of 21 key subjects from tourism dissertations, which is complementary to the core subjects identified from prior state-of-the-art analyses of field journals (Ballantyne et al., 2009; Xiao & Smith, 2006). Interestingly, in light of theory versus application and humanistic versus nonhumanistic distinctions, dissertation subjects have clustered into several subdomains of knowledge. Notably, Recreation stands out as a subdomain with weak dimensionalities with others; a business-oriented subdomain is characteristic of knowledge linkages to Agricultural Economics, Economics, Marketing, Management, Behavioral Science, Social Psychology, and Communication; a theoretical and humanistic subdomain is formed around Cultural Anthropology, Minority and Ethnic Groups, Sociology, and American Studies; and another policy-/development-oriented subdomain grows around Political Science, Environmental Science, Public Administration, Urban Planning, and Area Planning and Development. In addition, Geography, Social Structure, and Women Studies have formed relatively distinct subdomains of their own in tourism dissertation research.

The study also finds that Recreation, Tourism, and Marketing are most frequently connected with other subjects to form dominant knowledge networks in dissertation research. Prominent and important knowledge linkages have grown around Economics, Management, Cultural Anthropology, Planning and Development, Environmental Science, Geography, American Studies, and Sociology.

Additionally, the growth and linkage of its knowledge are reflective of the origin of its education and research. As noted by Smith and Godbey (1991), leisure, recreation, and tourism have been of close and subtle relationships with embeddedness and separations over the years. In the late 1960s and early 70s, travel and tourism were newcomers in the spectrum of North American recreation and park administration and leisure studies (van Doren & Heit, 1973). As the field grew, tourism has quickly emerged as a major domain in recreation departments and brought with it a clear bearing on business and marketing. In the 1980s, Jackson and Burton (1989) noted tourism as the third most dominant theme in recreation and leisure studies. Similarly, such an embeddedness was noted by Perdue, Couglin, and Valerius (1987), who reported that faculty in recreation and leisure studies contributed almost one third to the published research in Annals of Tourism Research and Journal of Travel Research. Interestingly, according to the recently published subjects in Journal of Leisure Research, Leisure Sciences, and Leisure Studies (e.g., the past 10 volume years), tourism as a dominant subject has almost faded away in these top three recreation and leisure studies journals. To some extent, the recent evolution of tourism and recreation and leisure studies into relatively isolated domains (or “silos”) in North America could partly explain the return of subject foci in doctoral tourism research.

Furthermore, a dynamic structure of knowledge linkages is notable in doctoral tourism research. Relatively stable and enclosed knowledge networks have formed round subject areas, such as Public Administration, Communication, Social Psychology, Social Structure, Minority and Ethnic Groups, Political Science, Behavioral Science, and Women Studies while knowledge networks around Recreation, Tourism, Marketing, Economics, and Management have remained relatively low in density, suggesting a degree of openness and vibrancy of the field for multidisciplinary collaborations.

In sum, although observations from this study are still largely reflective of a multidisciplinary nature of the field (Graburn & Jafari, 1991), the notable interrelatedness of subjects across knowledge domains and the increasing blurring of boundaries are indeed characteristic of a trend for tourism to evolve into an inter-/postdisciplinary state (or stage) of knowledge development (Coles et al., 2006; Coles & Hall, 2006; Hall & Page, 2009; Tribe & Xiao, 2011) or of a turn toward new modes of critical tourism enquiries (Ateljevic et al., 2007; Ren, Pritchard, & Morgan, 2010; Tribe, 2005). Complementary to prior state-of-the-art appraisals and perspectives, results from this analysis contribute to recent discussions on scientific communities, actor–network theories, and institutionalization in (or of) tourism research (Hall, Williams, & Lew, 2004; Paget, Dimanche, & Mounet, 2010; Ren et al., 2010; Rodger, Moore, & Newsome, 2009; Tribe, 2010; van der Duim, 2007; Xiao, 2010). Methodically, doctoral dissertations are typical texts (or manifest contents) for examining knowledge linkages and network formations, and SNA serves as a useful analytic approach for this scrutiny. The study adds a perspective on knowledge development and intellectual connections in inter-/transdisciplinary tourism studies.

Conclusion

This SNA examined the structure and evolution of knowledge development based on tourism dissertations. The size and density of its subject networks indicate the openness and vibrancy of tourism as a research community. Longitudinal examinations suggest that its diffusion patterns have been evolving, and 21 subjects are identified as subdomains in its knowledge development seen through its dissertations. The study also demonstrates the dimensionalities of, and associations among, these subdomains. The results both confirm and lend to discussions on notions such as scientific community (Collins, 1974; Crane, 1972), social networks (Granovetter, 1973), actor–network theory (Latour, 1987; Law, 1999), and permeable theoretical boundaries in social sciences (Abbott, 2001). A gradual evolving of tourism into an inter-/postdisciplinary state (or stage) of enquiry is notable in its knowledge development. The study sheds light on intellectual connections in tourism knowledge networks. Methodologically, this application of SNA also contributes to approaches or techniques for examining knowledge networks.

Notwithstanding, a number of limitations should be acknowledged. First, data collection was restricted by the use of only five keywords for dissertation document search. Some tourism-related dissertations may not have been included due to their use of other keywords. Second, the analysis is of a relatively short span (1994-2008) for assessing subject linkages and knowledge development in tourism dissertations. Despite a brief history of systematic research, tourism as a domain of knowledge could benefit from comprehensive examinations over a longer period.

In addition although this study reported a U-pattern or structural change in the evolution of knowledge networks, the dynamism and causes and consequences of the shift are yet to be seen in (doctoral) tourism research. Likewise, although the analysis identified 21 subdomains of knowledge and examined their linkages, how connections among these subdomains collectively evolve remains largely unknown in tourism studies. Notably, such trends and network evolution scenarios could serve as foci for future studies. Moreover, in terms of the strength of knowledge ties (or density of links among subject areas), it would be interesting to find out the respective contribution of strong versus weak ties to the structural change in tourism knowledge networks.

Another issue for future inquiries has to do with longitudinal observation of the interactions among selected subdomains of tourism knowledge. For example, departing from this endeavor, it will be interesting that future tourism knowledge research look into how sociology/anthropology and business interact to form new networks in its scientific community, and how such interactions and collaborations will have resulted in greater interdisciplinarity and consequently enhanced theory development in tourism.

Footnotes

Tianyu Ying, PhD (email: yingtianyu@gmail.com), is an assistant professor in the Department of Tourism at the University of Otago, Dunedin, New Zealand.

Honggen Xiao, PhD (email: hmhgxiao@polyu.edu.hk), is an assistant professor in the School of Hotel and Tourism Management at The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong SAR. Both authors share interests in knowledge networks, social network analysis, and tourism knowledge development.

References

Abbott

(2001). Chaos of disciplines. Chicago, IL: University of Chicago Press.

Ateljevic

Morgan

Pritchard

(2007). Editor’s introduction: Promoting an academy of hope in tourism enquiry. In Ateljevic

Pritchard

Morgan

(Eds.), The critical turn in tourism studies: Innovative research methodologies (pp. 1-8). Oxford, England: Elsevier.

Ballantyne

Packer

Axelsen

(2009). Trends in tourism research. Annals of Tourism Research, 36, 149-152.

Baloglu

Assante

(1999). A content analysis of subject areas and research methods used in five hospitality management journals. Journal of Hospitality & Tourism Research, 23, 53-70.

Ben-David

(1964). Scientific growth: A sociological view. Minerva, 2, 455-476.

Borgatti

Everett

(1997). Network analysis of 2-mode data. Social Networks, 19, 243-269.

Borgatti

Everett

(1999). Models of core/periphery structures. Social Networks, 21, 375-395.

Borgatti

Everett

Freeman

(2002). UCINET for Windows: Software for social network analysis. Boston, MA: Harvard Analytic Technologies.

Callon

(1986). Some elements of a sociology of translation: Domestication of the scallops and the fishermen of St Brieuc Bay. In Law

(Ed.), Power, action and belief: A new sociology of knowledge (pp.196-233). London, England: Routledge.

10.

Coles

Hall

C. M.

(2006). The geography of tourism is dead. Long live geographies of tourism and mobility. Current Issues in Tourism, 9, 289-292.

11.

Coles

Hall

C. M.

Duval

(2006). Tourism and post-disciplinary enquiry. Current Issues in Tourism, 9, 293-319.

12.

Collins

(1974). The TEA set: Tacit knowledge and scientific networks. Science Studies, 4, 165-185.

13.

Contractor

Wasserman

Faust

(2006). Testing multitheoretical, multilevel hypotheses about organizational networks: An analytic framework and empirical example. Academy of Management Review, 31, 681-703.

14.

Crane

(1972). Invisible colleges: Diffusion of knowledge in scientific communities. Chicago, IL: University of Chicago Press.

15.

Crawford-Welch

McCleary

(1992). An identification of the subject areas and research techniques used in five hospitality related journals. International Journal of Hospitality Management, 11, 155-167.

16.

Crichton

(1978). Dissertations on travel, recreation, and leisure. Journal of Travel Research, 17, 14-22.

17.

Cross

Liedtka

Weiss

(2005). A practical guide to social networks. Harvard Business Review, 83, 124-132.

18.

Durland

Fredericks

(2005). An introduction to social network analysis. In Durland

Fredericks

(Eds.), Social network analysis in program evaluation: New directions of evaluation (No. 107, pp. 5-13). Hoboken, NJ: Wiley.

19.

Foucault

(1980). Power/knowledge: Selected interviews and other writings 1972–77. Brighton, Sussex: Harvester Press.

20.

Freeman

(1979). Centrality in social networks: Conceptual clarification. Social Networks, 1, 215-239.

21.

Friedkin

(1981). The development of structure in random networks: An analysis of the effects of increasing network density on five measures of structure. Social Networks, 3, 41-52.

22.

Fuller

(2000). The governance of science. Philadelphia, PA: Open University Press.

23.

Geuna

Salter

Steinmueller

(2003). Science and innovation: Rethinking the rationales for funding and governance. Cheltenham, England: Edward Elgar.

24.

Graburn

Jafari

(1991). Introduction: Tourism social science. Annals of Tourism Research, 18, 1-11.

25.

Granovetter

(1973). The strength of weak ties. American Journal of Sociology, 78, 1360-1380.

26.

Greenberg

(2009). How citation distortions create unfounded authority: Analysis of a citation network. British Medical Journal (Clinical Research Edition), 339, b2680.

27.

Hall

C. M.

Page

(2009). Progress in tourism management: From the geography of tourism to geographies of tourism—A review. Tourism Management, 30, 3-16.

28.

Hall

Williams

Lew

(2004). Tourism: Conceptualizations, institutions, and issues. In Lew

Hall

Williams

(Eds.), A companion to tourism (pp.3-21). Malden, MA: Blackwell.

29.

Harcourt

. (2002). Measured interpretation: Introducing the method of correspondence analysis to legal studies. University of Illinois Law Review, 4, 979-1018.

30.

Horn

Finholt

Birnholtz

Motwani

Jayaraman

(2004, November). Six degrees of Jonathan Grudin: A social network analysis of the evolution and impact of CSCW research. Paper presented at the Computer Supported Cooperative Work, Chicago, IL.

31.

Racherla

(2008). Visual representation of knowledge networks: A social network analysis of hospitality research domain. International Journal of Hospitality Management, 27, 302-312.

32.

Jackson

Burton

(1989). Mapping the past. In Jackson

Burton

(Eds.), Understanding leisure and recreation: Mapping the past, charting the future (pp. 2-28). State College, PA: Venture.

33.

Jafari

Aaser

(1988). Tourism as the subject of doctoral dissertations. Annals of Tourism Research, 15, 407-429.

34.

Jamal

Smith

Watson

(2008). Ranking, rating and scoring of tourism journals: Interdisciplinary challenges and innovations. Tourism Management, 29, 66-78.

35.

Jogaratnam

Chon

McCleary

Mena

Yoo

(2005). An analysis of institutional contributors to three major academic tourism journals: 1992-2001. Tourism Management, 26, 641-648.

36.

Kilduff

Tsai

(2003). Social networks and organizations. Thousand Oaks, CA: Sage.

37.

Kim

Savage

Howey

van Hoof

(2009). Academic foundations for hospitality and tourism research: A reexamination of citations. Tourism Management, 30, 752-758.

38.

Klenk

Hickey

Maclellan

Gonzales

Cardille

(2009). Social networks analysis: A useful tool for visualizing and evaluating forestry research. International Forestry Review, 11, 134-140.

39.

Knoke

Kuklinski

(1982). Network analysis. Beverly Hills, CA: Sage.

40.

Knoke

Yang

(2008). Social network analysis (2nd ed.). Thousand Oaks, CA: Sage.

41.

Latour

(1986). The powers of association. In Law

(Ed.), Power, action and belief: A new sociology of knowledge (pp. 264-280). London, England: Routledge.

42.

Latour

(1987). Science in action: How to follow scientists and engineers through society. Cambridge, England: Harvard University Press.

43.

Law

(1999). After ANT: Complexity, naming and topology. In Law

Hassard

(Eds.), Actor network theory and after (pp. 1-14). Oxford, England: Blackwell.

44.

Liebowitz

(2005). Linking social network analysis with the analytic hierarchy process for knowledge mapping in organizations. Journal of Knowledge Management, 9, 76-86.

45.

Lin

(1995). The cross-citation analysis of selected marketing journals. Journal of Management, 12, 456-489.

46.

Lin

(2001). Social capital: A theory of social structure and action. Cambridge, England: Cambridge University Press.

47.

McKercher

(2005). A case for ranking tourism journals. Tourism Management, 26, 649-651.

48.

Melin

(2000). Pragmatism and self-organization: Research collaboration on the individual level. Research Policy, 29, 34-40.

49.

Meyer-Arendt

Justice

(2002). Tourism as the subject of North American doctoral dissertations 1987-2000. Annals of Tourism Research, 29, 1171-1174.

50.

Mitchell

(1969). Social networks in urban situations. Manchester, England: Manchester University Press.

51.

Mulkay

(1977). Sociology of the scientific research community. In Spiegel-Rosing

Price

(Eds.), Science, technology, and society: A cross-disciplinary perspective (pp. 93-148). London, England: Sage.

52.

Mullins

(1972). The development of a scientific specialty: The phage group and the origins of molecular biology. Minerva, 10, 51-82.

53.

Newman

(2001). Scientific collaboration networks (I): Network, construction and fundamental results. Physics Review E, 64, 0161311-0161318.

54.

Newman

(2004). Co-authorship networks and patterns of scientific collaboration. Proceedings of the National Academy of Sciences U S A, 98, 404-409.

55.

Paget

Dimanche

Mounet

(2010). A tourism innovation case: An actor-network approach. Annals of Tourism Research, 37, 828-847.

56.

Pechlaner

Zehrer

Matzler

Abfalter

(2004). A ranking of tourism and hospitality journals. Journal of Travel Research, 42, 328-332.

57.

Perdue

Couglin

Valerius

(1987). Tourism and commercial recreation: Past, present, and future research. In Barnett

(Ed.), Research about leisure: Past, present, and future (pp.161-174). Champaign, IL: Sagamore.

58.

Pizam

Chacko

(1982). Tourism and hospitality related dissertations 1976-1980. Annals of Tourism Research, 9, 587-620.

59.

ProQuest. (2009a). ProQuest dissertations and theses database. Retrieved from http://www.proquest.com/en-US/catalogs/databases/detail/pqdt.shtml

60.

ProQuest. (2009b). Thesaurus. Retrieved from http://proquest.umi.com/i-std/en/lcd/tools/thesaurus.htm

61.

Racherla

(2010). A social network perspective on tourism research collaborations. Annals of Tourism Research, 37, 1012-1034.

62.

Ren

Pritchard

Morgan

(2010). Constructing tourism research: A critical inquiry. Annals of Tourism Research, 37, 885-904.

63.

Riley

Love

(2000). The state of qualitative tourism research. Annals of Tourism Research, 27, 164-187.

64.

Roberts

Jr. (2000). Correspondence analysis of two-mode network data. Social Networks, 22, 65-72.

65.

Rodger

Moore

Newsome

(2009). Wildlife tourism, science and actor-network theory. Annals of Tourism Research, 36, 645-666.

66.

Schonstrom

(2005). Creating knowledge networks: Lessons from practice. Journal of Knowledge Management, 9, 17-29.

67.

Scott

(2000). Social network analysis (2nd ed.). London, England: Sage.

68.

Scott

Baggio

Cooper

(2008). Network analysis and tourism: From theory to practice. Clevedon, England: Channel View.

69.

Severt

Tesone

Bottorff

Carpenter

(2009). A world ranking of the top 100 hospitality and tourism programs. Journal of Hospitality & Tourism Research, 33, 451-470.

70.

Sheldon

(1991). An authorship analysis of tourism research. Annals of Tourism Research, 18, 473-484.

71.

Smith

(2007). Average path length in complex networks: Patterns and predictions. Cornell University Library. Retrieved from http://arxiv.org/abs/0710.2947

72.

Smith

Godbey

(1991). Leisure, recreation and tourism. Annals of Tourism Research, 18, 85-100.

73.

Swain

Brent

Long

(1998). Annals and tourism evolving: Indexing 25 years of publication. Annals of Tourism Research, 25, 991-1014.

74.

Tribe

(2005). New tourism research. Tourism Recreation Research, 30, 5-8.

75.

Tribe

(2010). Tribes, territories and networks in the tourism academy. Annals of Tourism Research, 37, 7-33.

76.

Tribe

Xiao

(2011). Developments in tourism social science. Annals of Tourism Research, 38, 7-26.

77.

Trochim

Marcus

Masse

Moser

Weld

(2008). The evaluation of large research initiatives: A participatory integrative mixed-methods approach. American Journal of Evaluation, 29, 8-28.

78.

Tsai

Ghoshal

(1998). Social capitals and value creation: The role of inter-firm networks. Academy of Management Journal, 41, 464-476.

79.

Urry

(2004). Small worlds and the new social physics. Global Networks, 4, 109-130.

80.

van der Duim

(2007). Tourismscapes: An actor-network perspective. Annals of Tourism Research, 34, 961-976.

81.

van Doren

Heit

(1973). Where it’s at: A content analysis and appraisal of the Journal of Leisure Research. Journal of Leisure Research, 5, 67-73.

82.

Wagner

Leydesdoff

(2005). Network structure, self-organization, and the growth of international collaboration in science. Research Policy, 34, 1608-1618.

83.

Wasserman

Faust

(1994). Social network analysis: Methods and applications. Cambridge, England: Cambridge University Press.

84.

Wixted

Holbrook

(2008). Conceptual issues in the evaluation of formal research networks. Vancouver, British Columbia, Canada: CPROST.

85.

Xiao

(2010). The capacity of a scientific community: A study of the Travel and Tourism Research Association. Journal of Hospitality & Tourism Research. Advance online publication. doi:10.1177/1096348010380597

86.

Xiao

Smith

(2006). The making of tourism research. Annals of Tourism Research, 33, 490-507.

87.

Zhao

Ritchie

J. R. B.

(2007). An investigation of academic leadership in tourism research 1985-2004. Tourism Management, 28, 476-490.