Tourism trends in central region of Portugal

Abstract

With continuous growth over the last few decades, tourism in Portugal has consolidated itself as a fundamental activity for generating wealth and employment, and there has been a diversification of destinations, where in addition to the historically best-known city regions, other less explored regions are gaining popularity in the search for authentic experiences. In this context, it is essential to highlight the importance of the Central Region, which is not only rich in historical and cultural heritage, but also has natural surroundings conducive to outdoor activities, ecological initiatives, and responsible tourism. The region comprises coastal and inland municipalities, marked by various diversities, potentialities, but also weaknesses in their NUTS III.

Objectives | The aim was to identify distinct groups of municipalities with similar patterns, both in terms of value and the profile (correlation) of the time series, as well as spatial correlations of municipalities with high, low and outliers. Was also to evaluate the existence of differences between the region's NUTS III and to study the relational structure of the indicators under analysis.

Methodology | We used data on tourism indicators (lodging capacity (LC), lodging capacity per 1000 inhabitants (LC1000), nights (N), nights per 100 inhabitants (N100), average stay (AS), guests (G), proportion of non-resident guests (PNRG), lodging incomes (LI), total incomes (TI), Rooms (R) and Bed occupancy net rate (BONR) for 2017 and 2022, provided by the National Statistics Institute, as well as the Official Administrative Map of Portugal, version 2023, from the Directorate General for Territory, to delimit the territorial units of the study area, namely municipalities and NUTS III. Statistical analysis was performed using IBM SPSS 29.0.1 and R 4.3.2, integrated with the GIS application ArcGIS Pro 3.2.

In the cluster and outlier analysis, the Global Moran index (I) (Anselin, 1995) was used as a local indicator of the spatial association of municipalities, conceptualizing as neighbors those municipalities whose geometric boundaries share at least one vertex, and the euclidean distance method was used.

In the time series clustering, when the characteristic of interest parameter was the value of the indicator, the similarity of the series was obtained by taking the square root of the sum of the squares of the differences of the respective annual values. When studying the profile of the time series, similarity was measured based on the statistical correlation of the series over the years. The number of clusters was based on the highest value of the Pseudo-F statistic.

Correlation, ANOVA and the Kruskal-Wallis test followed by post-hoc tests were used to evaluate the existence of differences in the indicators between NUTS III and the relational structure of the indicators was evaluated by Factor Analysis.

Main Results and Contributions | There was great heterogeneity in the distribution of the indicators in the different NUTS III, 50% of the tourist accommodation establishments in the Centro Region in the years under study had a LC greater than or equal to 256 and a number of nights greater than or equal to 20201; a number of rooms greater than or equal to 120.17; a AS greater than or equal to 1.74 nights; a total number of guests greater than or equal to 11944 and a PNRG of 19.33; LI greater than or equal to 662.58 thousand €, TI greater than or equal to 898.67 thousand € and a BONR greater than or equal to 22.59 %. Tábua and S. Pedro do Sul with more than 3 nights of AS, Aveiro and Nazaré with more than 42% of BONR and Ourém followed by Coimbra in the other indicators.

From the of cluster and outlier analysis in 2022, it is important to mention the municipalities of: Leiria for having high values in the LC, N, G, LI, TI and R indicators, as do the neighboring municipalities; Lousã and Sertã for having low values in the same indicators, as do the neighboring municipalities; Batalha and Pombal for having low values in these indicators, while the neighboring municipalities have high values; Figueira da Foz for having high values in the LC1000 and N100 indicators, while the neighboring municipalities have low values.

The time series clustering showed that for the study in terms of value, the municipalities were grouped into 2 clusters in the analysis of the AS and BONR indicators, 9 clusters for TI and N100, 3 clusters for the PNRG indicator and 10 clusters for the remaining indicators. Regarding the analysis of the series profile, the municipalities were grouped into 2 clusters for the R indicator, 9 clusters for AS and G, 6 clusters for the BONR indicator and 10 clusters for the remaining indicators.

Factor analysis showed strong and significant correlations between LC, N, G, R, LI and TI and low correlations, almost zero, between AS and PNRG with the other indicators. The relational structure of the indicators is explained by three latent factors that explain 86 % of the total variance (component 1: LC, N, G, R, LI and TI; component 2: AS, PNRG and BONR; component 3: LC1000 and N100).

Regarding the NUTS III in the Central Region, there are significant differences for PNRG between Beira Baixa and the other NUTS III, except with Médio Tejo and Beiras e Serra da Estrela and between Coimbra and Beiras e Serra da Estrela for the LI and TI indicators.

Limitations | In addition to the time constraints, which prevented contact with the promoters of the respective tourist accommodation, for a more robust time series analysis, it would have been desirable for all the indicators to have annual data for a longer range, 10 or more years. Furthermore, in the 6 years of data analyzed, not all municipalities provided data for each year.

Conclusions | Since the transformation of the Central Region into a major tourist destination necessarily involves defining appropriate strategies and plans to develop the entire region in an integrated manner (Eusébio et al., 2008), it is essential to identify the municipalities with the highest and lowest tourist visibility. In this sense, it was verified that the methodologies adopted, in addition to analyzing tourism indicator data, are an essential tool for identifying and mapping hot spots, cold spots and abnormal space values with agglomeration and significance, as well as making it possible to identify the most similar municipalities, enhancing the analysis and definition of management and marketing strategies for the region's tourism potential.

References

Anselin, L. (1995). Local Indicators of Spatial Association—LISA. Geographical Analysis, 27(2), 93-115. https://doi.org/https://doi.org/10.1111/j.1538-4632.1995.tb00338.x

Eusébio, C. A., Castro, E. A., Costa, C. (2008). Diversidade no Mercado Turístico da Região Centro de Portugal. Revista Turismo e Desenvolvimento, 10, 10-24. DOI: https://doi.org/10.34624/rtd.v0i10.13555