**5. Data Analysis**

The variables used have different units of measurement. Thus, a normalisation process was needed. The eigenvalues of the variance and covariance matrix of the transformed variables are shown in Table 5. The first principal component alone summarises more than 40% of the total variability, namely the information contained in the five variables used in the analysis, while the second is more than 33%. The data analysis has some limitations due to the secondary data delivered from the Observatory of Tourism for Islands Economy and compared with Eurostat ones. The variable considered to better describe the two EU island's tourist models is the only one available for all the EU islands. Sometimes the statistical indicators are different for each country at a sub-regional level; thus, the number of variables considered in this article is comparable but limited.

**Table 5.** Principal Component Analysis, Extracted Components.


Source: data analysis on OTIE islands database.

The factor analysis summarised the five variables into two components. In the first, the more critical in terms of expressed variability, we find the variables of social development (population variation), and economic development (interpretation of employers), together with the variable of the development of industrial tourism (variation of hotels). We can define the first component as relating to the product in various forms (social, economic, and tourist).

Thus, we move to a more in-depth analysis of the results by calculating the factor scores (FAC) resulting from the FA calculation and expressing the link between the cases and the extracted components. By placing the ingredients in hierarchical order concerning FAC1 (from the strongest to the weakest link), it is possible to understand the island's "behaviour" (Table 6) and make some reflections on the characteristics of the tourist models on these islands.


**Table 6.** Rank islands order and variables value considering FAC\_1 (value 2010–2011 vs. 2018–2019).

Source: data analysis on OTIE islands database.

First, it is evident that, compared to the other islands, Malta has a different tourism development model, strongly influenced by economic and social factors. The other islands in Table 6 have a less intense but evident development trend. In some cases, this economic development is measured only on the basis of the number of hotels. (La Réunion, Corse, Illes Balears, Região Autónoma dos Açores).

On the contrary, on the last five islands (the "marginal" islands), it is noted that despite a population loss and a decrease in hotel and non-hotel facilities, the trend of international arrivals is consistent. It, therefore, seems that the economic conditions towards which this group of islands is moving do not affect the international tourist attraction. The marginality is also evident from the non-growth of hotel structures, contrary to non-hotel systems (which grow in almost all the "marginal" islands).
