*3.1. Overlap of Databases*

The linear correlation coefficient is 0.73, which indicates a strong and direct correlation between Scopus and WoS. Of the 135 articles (84 Scopus and 51 WoS), it was observed that 32 articles were indexed in both databases, which represents 38% of the articles from Scopus and 63% of those from WoS. Consequently, the remaining 52 articles from Scopus and 19 from WoS were classified as single documents as they are present in only one of the two bases. Table 3 shows the results related to the singularity of the databases, measured through the Meyer index (MI), with Scopus being the database with the highest singularity index with a MI = 0.81, while in WoS it reaches MI = 0.69.


Source: own elaboration.

The traditional overlapping (TO) percentage between Scopus and WoS determined a similarity of 31.07% between the databases, which is also understood as the existence of a 68.93% disparity between them. Likewise, in addition to the previous calculations, it is necessary to determine the percentage of coverage that Scopus shows in relation to WoS and vice versa [35], for which relative overlapping (RO) was applied. The percentages obtained show that 38.10% of Scopus was covered by WoS, while the RO

of WoS shows that 62.75% of this database was covered by Scopus. As a result, Scopus has a lower overlap than WoS.

### *3.2. Productivity Per Years*

The joint database consists of 103 articles (32 duplicated articles were eliminated). Figure 2 shows that scientific production covers 25 years (1994–2018), the first indexed study being Market-Based Product Development in Heritage Tourism, by Light and Prentice [37]. During this period, 2017 is consolidated as the year of greatest productivity, with 21 published articles.

**Figure 2.** Trend of publications Scopus ∪ WoS.

Considering the four stages of Price [38] in the evolution of scientific production, precursors, exponential growth, linear growth, and collapse of the scientific field, the presence of two of them in the growth process are observed here. The first part or stage of precursors lasts from 1994 to 2003 and corresponds to 0.50 articles/year (60% with a single signature), with slow growth being evident. The second stage goes from 2004 to 2018, with a ratio of 6.53 articles/year, and most of the production involving a single author per article. It is observed that the Price law is fulfilled: between 10 and 15 years after the first publication, the information developed on the subject at a global level is duplicated [38].

Finally, Figure 3 shows an average correlation between Scopus and WoS regarding articles that have been indexed, with R<sup>2</sup> = 0.5364.

**Figure 3.** Correlation between the numbers of articles published in Scopus and WoS.

### *3.3. Citations*

During the 25 years of study, a cumulative total of 1135 citations (13.51 citations/article) are identified regarding the 84 Scopus articles indexed in Scopus. In WoS, 687 citations (51 articles), 13.47 citations/article, are recorded. Scopus has an h = 14, which means that at least 14 of the total articles identified have obtained 14 citations or more, and WoS an h = 8. The year that reaches the highest number of citations in both databases is 1998, with 378 citations in Scopus and 317 in WoS, concentrating 33% and 47% of the total citations, respectively.

Of the Scopus and WoS articles, 65% and 51%, respectively, obtained between 24 and 1 citations, while 27% and 43% of the articles, respectively, do not have any citations recorded during the analysis period. On the other hand, more than 100 citations were recorded in 3 articles. It was observed that documents published in recent years have not received a significant number of citations; surely because these studies did not achieve the necessary dissemination to be consolidated as referents of the subject, a fact that limits the amount of citations they can receive [39].

The most-cited articles on the subject were Local Development and Heritage: Traditional Food and Cuisine as Tourist Attractions in Rural Areas, by Bessiere [20], with 378 citations in Scopus (18.0 citations per year) and 314 in WoS (15.1 citations per year), Tourism Development of World Heritage Sites in China: A Geographic Perspective, by Li et al. [40] with 156 citations (14.2) in Scopus and 138 in WoS (12.5), and The "Vicious Circle" of Tourism Development in Heritage Cities, by Russo [41], with 130 in Scopus (7.6) and 106 in WoS (6.2). In addition, three articles that have received a considerable number of citations were identified, but these are only indexed within Scopus: Industrial Heritage: A Nexus for Sustainable Tourism Development, by Jonsen-Verbeke [42], Resident Attitudes towards Heritage Tourism Development, by Chen and Chen [43], and "Heritagisation", a Challenge for Tourism Promotion and Regional Development: An example of Food Heritage, by Bessiere [44].

### *3.4. Authors*

A total of 231 authors were identified in the scientific productions analyzed in the joint analysis matrix (103 articles), which corresponds to a productivity index per author of 1.04 articles. The most productive authors in the subject were Rasoolimanesh and Jaafar, with four articles belonging to the Universiti Sains Malaysia (Malaysia). Both authors have an average citations/article of 15.33 in Scopus and 8.25 in WoS. The second most productive author was Bessiere, of the University of Toulouse II (France), with 2 articles; however, he has a better citation average of 202.5 in Scopus and 317 in WoS. Authors such as Huibin, Marzuki, Razak, Min, and Sun are also in this second position, with 2 articles, but they are indexed only in the Scopus database and the number of citations is very small.

Total author productivity can be analyzed by means of different types of processes, which enables them to be classified according to the contribution that each author provides within the subject of study. The classification of Crane [45] is used to fulfill this purpose in this study, in which the production by authors can be explained by four groups of authors: (1) large producers—those who have a production greater than 10 articles, (2) moderate producers—authors who have produced between 5 and 9 documents, (3) aspiring authors—authors who have between 2 and 4 studies, and (4) transition authors—authors who have only produced one work. There are no large or moderate producers in this area; 223 are transition and 8 are aspiring.

The total transience index (TI = [PI (productivity index)= 0]) of the total set of documents under study is 96.5%; that is, this index is the same or represents the total of occasional authors who only arise once within the subject and who do not continue making contributions throughout the rest of the line of evolution, a figure which is based on the classification of Crane [45].

Another important element of analysis is the collaboration trend in scientific production, which makes it possible to analyze current relationships. In this sense, Berelson [46] determines that the more varied and the greater the collaboration shown within the development of documents, the greater the maturity of the subject of study [47]. This study revealed that 35% (36) of the articles are of a single authorship, and the remaining 65% (67) are the product of collaboration. In the case of collaboration, 29% (30) are signed by two authors, 17% (18) of papers are signed by three authors, 12% (12) by four authors, and 7% (7) are signed by five or more authors. With these data, the co-authorship index is 2.34 authors/article.
