Cultural Heritage and Economic Development Measuring Sustainability Over Time PDF

Socio-Economic Planning Sciences 95 (2024) 101998 Contents lists available at ScienceDirect Socio-Economic Planning Sciences...

Socio-Economic Planning Sciences 95 (2024) 101998 Contents lists available at ScienceDirect Socio-Economic Planning Sciences journal homepage: www.elsevier.com/locate/seps Cultural heritage and economic development: measuring sustainability over time Carla Galluccio, Francesca Giambona ∗ Department of Statistics, Computer Science, Applications ‘‘G. Parenti’’, University of Florence, Viale Giovanni Battista Morgagni, 59, Firenze, 50134, Italy ARTICLE INFO ABSTRACT Keywords: This study investigates the role of museums and cultural heritage in local development, highlighting their Cultural heritage ability to attract tourists, generate revenue, and promote inclusion and cultural diversity. According to Sustainability traditional economic theory, cultural heritage provides positive externalities, enhancing employment and Latent transition analysis improving human and social capital, all while adhering to principles of sustainability. In this context, the Economic development Italian Survey on Museums and Other Cultural Institutions offers extensive data on heritage conservation, accessibility, and visitor services. Utilising this longitudinal data, we conduct a latent transition analysis to examine the evolution of the Italian museum sector with a focus on regional differences and museums’ dimension. Our findings classify Italian museums into three homogeneous sustainability states. Additionally, museum size positively affects both the initial and transition probabilities, while the macro-area significantly influences only the initial probability. 1. Introduction valorisation of artistic and cultural heritage takes particular relevance. Beyond that, museums function as dynamic spaces for education, re- In recent decades, a confluence of social, economic, and technolog- search, and cultural exchange. They contribute to enriching societal ical shifts has progressively reshaped the fundamental essence of mu- knowledge and inspire creativity, fostering a sense of continuity and seum organisations. This evolution has propelled them towards greater interconnectedness with the past. engagement in fostering social innovation and promoting territorial From an economic perspective, museums significantly influence a development, all while adhering to sustainability principles. country’s level of civilisation. The presence of well-managed museums The connection between sustainability and heritage preservation enhances a nation’s soft power , attracting global attention and is highlighted in the definition given by The International Council of fostering cultural diplomacy. The economic theory of cultural capital Museums (ICOM), for which ‘‘[...] sustainability is the dynamic process posits that investments in cultural institutions, like museums, yield of museums, based on the recognition and preservation of tangible and returns in terms of enhanced societal well-being and economic pros- intangible heritage with the museums responding to the needs of the commu- perity. Moreover, museums can act as catalysts for urban development, nity. To be sustainable, museums, through their mission, must be an active attracting businesses, residents, and investments to surrounding areas and attractive part of the community by adding value to the heritage and and improving the cities’ reputation. social memory.’’. By this definition, the reference to the temporal In this context, the Italian Survey on Museums and Other Cultural evolution of the term is clearly evident, as well as the close link between Institutions, carried out by the Italian Statistical Institute (Istat), the sustainability, attractiveness and local development (also in terms of Ministry of Cultural Heritage and Activities and Tourism, the Regions social inclusion). and the Autonomous Provinces since 2011, provides a wide range of However, the management of museums is not merely a custodial information for museums and similar structures (https://www.istat.it/ responsibility; the management of museums and cultural heritage can en/archivio/167568). The main research areas analysed are heritage be traced back to some traditional concepts of economic theory. These goods, in fact, satisfy needs that improve a country’s civilisation conservation, accessibility to exhibition spaces, visitor orientation, and level, providing positive externalities, as cultural heritage positively relations with the external context. The availability of longitudinal data affects employment in economic sectors (such as tourism) and deter- allows not only to classify museums based on the available information mines an improvement in human and social capital. In this way, but also to analyse the evolution of museum groupings over time, the ability of museums to achieve their purposes of conservation and paying particular attention to territorial detail and museum size. ∗ Corresponding author. E-mail addresses: [email protected] (C. Galluccio), [email protected] (F. Giambona). https://doi.org/10.1016/j.seps.2024.101998 Received 29 March 2024; Received in revised form 13 June 2024; Accepted 15 June 2024 Available online 21 June 2024 0038-0121/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 Italy is an exemplary case study for examining the role of museums of indicators and classifications, might be important to make a strong and cultural heritage in local development due to its rich and diverse argument for the value of cultural heritage for social and economic cultural history. As a country renowned for its extensive collection of progress. museums, archaeological sites, and cultural institutions, Italy provides Museums are, above all, cultural destinations but they are also a unique context in which to study the impacts of cultural heritage tourist destinations and thus have an impact on economic activity. Mu- on economic and social outcomes. The Italian Survey on Museums and seums represent repositories of a nation’s cultural wealth, safeguarding Other Cultural Institutions offers comprehensive and longitudinal data, its heritage for present and future generations. In this vein, museums allowing for an in-depth analysis of trends and changes over time. are a resource for local development. Indeed, they can help attract Additionally, Italy’s regional diversity, with distinct cultural and eco- tourists , generate revenue, and promote inclusion and cultural nomic profiles across different areas, provides a valuable opportunity to diversity, playing a fundamental role in preserving and promoting cultural heritage. explore how local factors influence the sustainability and effectiveness There is a strong relationship between museum activities and local of museums. By focusing on Italy, we can draw insights that are (regional) tourism, as museum visitors’ trips generate economic activ- relevant not only to other countries with rich cultural heritages but ity, which is related not only to museums but also to enterprises in also to regions aiming to leverage cultural assets for local development the tourist industry, retail business, and many other destinations in the and economic regeneration. locality of museums. Therefore, the present work aims to reach the following research Several international initiatives have emphasised the need to create objectives: objective (and synthetic) measures to monitor museums’ sustainability RQ1. Classify museums by sustainability: Define and classify Italian over time and the fundamental role of cultural organisations in a museums based on their sustainability practices. region’s sustainable development process. However, despite the clear relationship between museums, culture, and sustainable develop- RQ2. Analyse temporal changes: Investigate whether membership in ment, little has been done on this topic. the different groups has changed over time. In this vein, research on museums and sustainability has gener- RQ3. Investigate the impact of museums’ features: Understand ally focused on defining the concept of museum sustainability and whether certain ‘‘structural’’ variables, such as museum size and advocating for its integration into museum practices, exploring how geographical location, have played a role in the membership various museum activities contribute to or can enhance sustainable and/or change. development. For instance, the type of museum may influence its RQ4. Map territorial distribution of groups: Visualise the geographic sustainability, as certain museums attract more visitors than others, distribution of classified groups. potentially impacting their financial stability and community rele- vance. Indeed, museums are increasingly acknowledged as fundamental Despite the growing interest in the sustainability of cultural institu- institutions for heritage and cultural preservation and their potential to tions, there are no studies in the literature aiming at classifying Italian advance broader societal goals, including sustainability. museums based on their sustainability levels using quantitative indi- In this regard, a significant contribution comes from , who cators. Our study addresses this gap by employing a latent transition investigated, through a qualitative approach (literature review and analysis to categorise museums into different sustainability classes. It semi-structured interviews), the elements that could be used to define is worth noting that this classification not only offers insights into the and measure a museum’s level of sustainability. current status of the museums analysed but also provides a valuable The authors show the complex interplay of influences affecting mu- tool for classifying museums that do not participate in the census seum sustainability. Highlighting governance structures as foundational survey every year. In other words, by leveraging the characteristics to sustainability, they emphasise the role of organisational frameworks of museums that have been classified by the model, we can infer and decision-making processes in fostering adaptability and innovation. the sustainability status of similar institutions that were not directly They also underscore the significance of financial considerations, advo- surveyed. This approach allows for an ex-post classification, ensuring cating for diversified funding sources and strategic resource allocation that even those museums not consistently captured in the annual survey to ensure long-term growth. can still be evaluated and monitored for sustainability. This capabil- Moreover, sustainability in museums extends beyond financial con- ity is particularly crucial for creating comprehensive and continuous cerns to encompass broader cultural and social dimensions. In this oversight of the museum sector, contributing to more informed and sense, it emerges as essential for sustainable growth to focus on prac- effective policy-making and resource allocation. Finally, another reason tices in collection management, exhibition design, and facility oper- for focusing on the Italian case is the access to official and reliable ations. In other words, it is emphasised that museums must address data collected by Istat. In fact, this dataset offers authoritative and lon- environmental challenges. Therefore, the concept of sustainability has gitudinal information about Italian museums and cultural institutions, become fundamental to encouraging meaningful community relation- ensuring a robust foundation for our analysis. ships and promoting the role of museums as cultural exchange and The remaining part of the paper is organised as follows. In Section 2, dialogue hubs. However, as pointed out, the existing measures used to define we present the theoretical framework. Section 3 describes the statistical and measure museums’ sustainability should have considered com- model. In Section 4, we illustrate the data under analysis and present prehensive assessment frameworks capturing economic, environmen- some descriptive statistics. Section 5 is devoted to the results. Finally, tal, and social indicators. Therefore, they proposed novel indicators the conclusions in Section 7 end the paper. regarding financial stability, environmental stewardship, community engagement, and cultural preservation. These indicators offer museums 2. Theoretical framework deeper insights into their operational strengths and weaknesses and facilitate their informed decision-making and strategic planning for the In recent years, policy makers have underlined the strategic value future. of cultural heritage to territorial development, economic growth and Based on this framework and following the ICOM definition of employment, as stated in the New European Agenda for Culture and museum sustainability, a classification of Italian museums is proposed. in the European Heritage Strategy [7,8]. However, it is difficult to The availability of Istat longitudinal data allows to classify museums measure the magnitude of cultural heritage impact on the economy based on five sustainability dimensions, jointly with the changes in and territories. To properly understand the contribution that cultural museums classification over time, paying particular attention to the heritage makes to the market and society, a common framework must role played by territories and museum size on the membership and be established in Europe for the collection of standardised and compa- transition probabilities. From a methodological point of view, a latent rable data on cultural heritage. Cultural heritage information, in terms transition regression analysis is applied to these goals. 2 C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 3. Method probabilities of membership in each latent status at Time 1 (𝛿𝑠1 ) af- fected by the covariate (x), the probabilities of transitioning to a latent Latent transition analysis (LTA) serves as a variant of the latent class status at a particular time conditional on latent status membership at model designed not only to capture latent class membership but also the immediately previous time (𝜏), and the probabilities of observing transitions in such membership over time. each response at each time conditional on latent status membership Unlike in latent class analysis (LCA), where latent classes denote (𝜌). The parameters 𝜌 express the relationship between each manifest stable sets of characteristics or behavioural states, in LTA, individu- variable (or indicator) and each latent class — which is to say that als may transition between latent classes across time. Hence, within item response probabilities indicate how units can be classified into the this framework, the term ‘‘latent status’’ is utilised instead of ‘‘latent specified latent classes, given their manifest variable values. class’’ , acknowledging that subgroup membership is not presumed Item response probabilities and latent class membership can help to remain constant over time. Consequently, the model is termed a accurately categorise museums into a specific class (or status). Using latent transition model. the LTA also allowed us to observe stability and change in the latent In LTA three sets of parameters are estimated: classes, which was useful for identifying stayers (those in the same class at each wave) and the number of movers. 1. The probabilities of latent status membership are estimated for In the process of model specification, a major concern in LTA (as in each time point; LCA) is choosing the number of latent states to retain. This is done by 2. The transition probabilities denoting the probability of transi- considering the parsimony and interpretability of the possible solutions, tioning from a particular latent status at time 𝑡 to another latent while seeking to give substantial meaning to the identified latent status at time 𝑡 + 1, typically displayed in a matrix with the rows status [19,20]. In order to determine the optimal number of latent corresponding to the earlier time and the columns corresponding states, it is useful to estimate different models specifying a different to the later time. These transition probabilities illustrate the number of states. The final is the model with the lowest BIC (Bayesian probability of transitioning to the latent status indicated in the information criterion) value or with the lowest AIC (Akaike information column, given prior membership in the latent status indicated criterion). in the row. Notably, the diagonal elements of this transition To perform the model, we used the LMest package imple- probability matrix signify the probability of remaining in a particular latent status at a given time, conditioned on being in mented in the R statistical software. that same latent status at the previous time; 4. Data 3. A series of item-response probabilities delineates the relationship between the observed indicators of the latent variable at each time point and latent status membership, mirroring the manner The Italian Survey on Museums and Other Cultural Institutions, in which factor loadings establish links between observed indi- carried out by the Italian Statistical Institute (Istat), the Ministry of cators and latent variables in factor analysis. That is, in addition Cultural Heritage and Activities and Tourism, the Regions and the Au- to the number of classes and the size of classes being subject to tonomous Provinces since 2011, provides a wide range of information change, it is interesting to locate the museums that are stayers for museums and similar structures useful to describe the development (in the same class at each time) and those who are movers. path of cultural heritage on sustainable socio-economic growth. In the following a brief description of data used in our analysis. That is, further to the number of classes (and their sizes) being sub- ject to change, we deemed it noteworthy to locate museums classified 4.1. The Italian survey on museums and other cultural institutions as stayers (in the same class at each time) and those who classified as movers, in terms of sustainability. The Italian Survey on Museums and Other Cultural Institutions is Moreover, akin to LCA, incorporating covariates into a latent tran- a census survey that aims to obtain and release data on museums sition model is essential. The aim of integrating covariates into the and cultural institutions. Each year, the data collected describes about model is to discern attributes that forecast membership in various latent 4,500 museums, with a response rate ranging from approximately 90% statuses and/or forecast transitions between latent statuses. to 95%. Let 𝐿𝑡 represents the latent variable at Time t with S latent statuses, Following the initial editions based on a four-year rotation principle where 𝑠1 = 1... S at Time 1, 𝑠2 = 1... S at Time 2, and so on, up (e.g., 2007, 2011, and 2015), the survey has transitioned to an annual to 𝑠𝑇 = 1... S at Time T. In addition, there is a covariate 𝑋 used to conduct since 2017, supported by the project for regional and sectoral predict latent status membership at Time 1 and transitions between statistics for the 2014–2020 cohesion policies, funded through EU latent statuses at any two adjacent times. Then the latent transition cohesion policy financing. model can be expressed as follows: The survey conducted annually by Istat offers an updated and ∑ 𝑆 ∑ 𝑆 detailed description of all the museums and other museum-related 𝑃 (𝑌 = 𝑦|𝑋 = 𝑥) = ⋯ 𝛿𝑠1 (𝑥)𝜏𝑠2 |𝑠1 (𝑥) … 𝜏𝑠𝑇 |𝑠𝑇 −1 (𝑥) structures present in Italy, that is, of all those permanent structures 𝑠1 =1 𝑠𝑇 =1 open to the public that acquire, conserve, communicate and exhibit, 𝑅 (1) ∏ 𝑇 ∏ 𝐽 ∏𝑗 𝐼(𝑦 =𝑟 ) non-profit, for study, education and pleasure, goods and collections of 𝜌𝑗,𝑟 𝑗,𝑡|𝑠 𝑗,𝑡 , cultural interest, whether public or private, state or non-state, as long 𝑗,𝑡 𝑡 𝑡=1 𝑗=1 𝑟𝑗,𝑡 =1 as they are equipped with services organised for use. where 𝑌 represents the response variables, 𝛿𝑠1 (𝑥) is the marginal The data pertain to museums, galleries, collections, archaeological probability of class membership at the initial time 𝑠 = 1, 𝜏𝑠2|𝑠1 (𝑥) sites and parks, monuments, and monumental complexes (both public denotes the probabilities of transition to a latent state conditionally and private) surveyed annually through the Istat survey. on the previous latent state membership, 𝑟𝑗,𝑡 = 1, … , 𝑅𝑗 refers to the The data is gathered using a web questionnaire and published categories of item 𝑗 at time 𝑡, with 𝑗 = 1, … , 𝐽 indicating the items and on the Istat’s website. The questionnaire covers different dimensions 𝑡 = 1, … , 𝑇 indicating the times, and 𝜌𝑗,𝑟𝑗,𝑡 |𝑠𝑡 represents the probability related to the museums, such as what kind of supports (audioguide of response 𝑟𝑗,𝑡 to item 𝑗 at time 𝑡 conditionally on the membership to and videoguide, QR code, application for smartphones or tablets) and latent state 𝑠 at time 𝑡. 𝐼(𝑦𝑗,𝑡 = 𝑟𝑗,𝑡 ) is an indicator function equal to 1 services (such as the possibility to book the ticket online or if there if 𝑦𝑗,𝑡 = 𝑟𝑗,𝑡 at time 𝑡, and equals to 0 otherwise. is a car park) they can provide to the visitors, financial aspects (if the Therefore, Eq. (1) expresses how the probability of observing a museum has received public or private funding), information about the particular vector of responses, conditioning to 𝑋 is a function of the collections exhibited (like, how many goods are exposed), and so on. 3 C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 five indicators per year, while Tables 8–10 in the Appendix show the Table 1 variables which were synthesised per year. Distribution of the museums that answered the questionnaire in 2018, About the covariates which could influence the initial and transition 2019, and 2021. probabilities, we included in the analysis the macro-area, dividing the Region 2018 2019 2021 regions into five geographical areas, that is, North-West (Liguria, Lom- Abruzzo 108 110 84 bardia, Piemonte, Valle d’Aosta), North-East (Emilia-Romagna, Friuli- Basilicata 48 49 42 Calabria 166 163 134 Venezia Giulia, Trentino-Alto Adige, Veneto), Center (Lazio, Marche, Campania 233 227 199 Toscana, Umbria), South (Abruzzo, Basilicata, Calabria, Campania, Emilia-Romagna 454 458 424 Molise, Puglia), and Island (Sardegna and Sicilia), and the size of the Friuli-Venezia Giulia 175 170 145 museums, a variable with three levels (small, medium, and large) Lazio 357 349 298 obtained by considering the quartiles of the distribution of the numbers Liguria 194 197 156 Lombardia 433 419 373 of workers employed in the museums. The variable macro-area, whose Marche 291 282 254 structure into five categories is consistent with the Eurostat NUTS- Molise 41 43 34 1 classification (https://ec.europa.eu/eurostat/web/nuts), is constant Piemonte 411 414 352 over the years, while the variable size can vary over time. Puglia 164 142 131 The selection of the macro-area of belonging and museums’ size Sardegna 290 307 265 Sicilia 260 241 220 as covariates is essential due to their significant influence on the Toscana 553 580 511 sustainability of museums. The macro-area of belonging, which divides Trentino-Alto Adige 201 200 183 Italy into geographical regions, captures regional differences in culture, Umbria 165 170 161 economy, and administration that affect museum operations, funding, Valle d’Aosta 60 60 46 Veneto 304 299 280 visitor demographics, and strategic priorities. Museums in wealthier regions may start with higher sustainability due to better funding and support, while regions with strong economic growth and effective cul- tural policies might see more museums improving their sustainability 4.2. Descriptive statistics over time. On the other hand, museums’ size influences resources, operational capacity, and strategic capabilities. Larger museums are ex- For our aims, we use the data carried out by the Survey on Museums pected to start with higher sustainability levels due to greater financial and Other Cultural Institutions for the years 2018, 2019 and 2021. The resources and extensive support networks, and they are more likely to year 2020 is not included in the analysis as it mainly focused on the improve their sustainability further through investments in infrastruc- Pandemic sanitary measures adopted by the museums. ture and programming. Smaller museums may have difficulties initially Table 1 shows the distribution of the museums that answered the but can benefit from targeted support and strategic partnerships to questionnaire in the years under analysis. enhance their sustainability. Including these covariates helps under- As it is possible to observe, the number of museums and cultural stand the impacts of regional context and institutional capacity on the institutions that answered the questionnaire is similar for all the years sustainability trajectories of museums, showing if the improvement in considered. More specifically, the data collected documents an Italian the museums’ sustainability practice could be ascribed to exogenous or heritage quantifiable in about 4 thousand museums (from 78% to 80% endogenous factors, providing insights for tailored policy interventions in the years considered), archaeological areas (about 7% of the total and support strategies. each year), monuments (from 13% to 15% in the years considered) and The number of museums interviewed in every year analysed, iden- ecomuseums (recorded only in 2018 and representing about 1.4% of tified by their name and address, equals 1,936. Among these, the the total) open to the public. majority of the museums coded as small and medium are located in the It is a heritage spread throughout the territory: in more or less one Center of Italy (about 23% and 35%, respectively), while the majority Italian municipality in three, there is at least one museum structure. of the museums coded as large are located in the North-West of Italy Generally, the majority are museums, galleries or collections, followed (about 30%). These results are generally consistent each year. by monuments and monumental complexes, archaeological areas and parks and ecomuseum structures. 5. Results Toscana, Emilia-Romagna, Lombardia, Piemonte, Lazio, and Veneto are the regions with the highest concentration of structures including In this paragraph empirical findings from the latent transition model museums, archaeological areas and monuments while Roma, Firenze, are discussed. Torino, Milano, Bologna, Trieste, Genova, Napoli, Venezia and Siena First of all, to select a suitable number of latent states to retain, are the top 10 cities with the highest number of the historical-cultural, we estimated the model without covariates and with homogeneous architectural, and archaeological richness of Italy. However, the terri- transition probabilities, and selected the model with the lowest BIC and torial differences and the type of structure are significant. AIC values. Fig. 1 shows the comparison between BIC and AIC for a From the three surveys, we detected five dimensions of the sus- number of latent states ranging from 1 to 7, based on which we selected tainability (following the definition proposed in ) of the museums: the model with 3 latent states. services, related to the facilities made available for the public; supports, Then, we estimated the model, including the covariates and keeping regarding supports made available to visitors for the visit; activities, that the number of latent states fixed at 3. Tables 3–5 show the estimated is, the activities organised by the museums for the visitors; web, namely conditional response means of the dimensions identified under this museum products and services available on the web; digitalisation, model, initial probabilities, and transition probabilities. The latter are related to the process of digitalisation of the assets the museums own. also shown in Fig. 2. We did not report the graph of the transition These dimensions were measured using different items every year. probabilities obtained under the null model as very similar to the one Therefore, to obtain a singular indicator which we could use to mea- obtained from the model with the covariates. sure them over the years, we aggregated the items composing each Considering that it is always possible to order the states according to dimension by summing the answers and dividing the result by the the informative content of the application, based on Table 3 we ordered number of items composing the dimensions. We were able to do that the states according to increasing sustainability levels, represented by because all the variables were indicator variables (or were transformed the dimensions identified. Therefore, museums in the first group show into indicator variables). Table 2 shows the descriptive statistics of the lower sustainability levels, depicted by lower values in each dimension. 4 C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 Table 2 Descriptive statistics of the five indicators services, supports, activities, web, and digitalisation per year. Year 2018 2019 2021 Item Min. Max. Mean Dev. Std. Min. Max. Mean Dev. Std. Min. Max. Mean Dev. Std. services 0 1 0.34 0.23 0 1 0.40 0.32 0 1 0.47 0.37 supports 0 1 0.44 0.22 0 1 0.40 0.22 0 1 0.23 0.28 activities 0 1 0.63 0.34 0 1 0.64 0.35 0 1 0.50 0.32 web 0 1 0.34 0.26 0 1 0.38 0.30 0 1 0.28 0.26 digitalisation 0 1 0.29 0.34 0 1 0.54 0.50 0 1 0.41 0.45 Table 3 Estimated conditional response means under the model with covariates with k = 3 latent states. Latent states 1 2 3 Item services 0.1836 0.3661 0.6269 supports 0.1996 0.3327 0.5453 activities 0.3449 0.6113 0.8067 web 0.1601 0.2979 0.5428 digitalisation 0.5753 0.7197 0.8142 Table 4 Estimated initial probabilities under the model with covariates with k = 3 latent states. Latent states 1 2 3 Initial probabilities 0.3217 0.3590 0.3194 Table 5 Estimated transition probabilities under the model with covariates with k = 3 latent states. Latent states 1 2 3 1 0.8448 0.0876 0.0675 2 0.0929 0.8488 0.0583 3 0.1086 0.0777 0.8137 Fig. 1. AIC and BIC criteria for a number of latent states ranging from 1 to 7 in the model without covariates. have higher levels in each dimension, showing higher effort in investing in competitive advantage. From Table 4, the probability of belonging to a latent state is homogeneous, with about 30% of the museums in each state. In Table 5 the probability of transition between states is reported. There is a high probability of remaining in the same latent state (about 85% for the first two states and 82% for the third). Noteworthy, a slight shift is found from state 3 to state 1 (about 11%), whilst museums characterised by a low-medium level of sustainability (states 1 and 2) mainly move into the closest categories (low versus medium, medium versus low). Table 6 provides the estimated regression parameters of the initial probabilities. In this context, the estimated North-East macro-area pa- rameter is positive and significant for latent states 2 and 3, indicating that the probability of being in one of these latent states at the begin- ning of the study is higher for museums located in the North-East of Italy with respect to the museums located in the North-West. On the contrary, museums located in the South of Italy have a lower initial probability of being in the latent state 2 compared to those located in the North-West of Italy. About the size, all the estimated regression coefficients are positive and significant, meaning that the probability Fig. 2. Averaged transition probabilities under the model with the covariates with k = 3 latent states. for medium and large museums to be in latent states 2 and 3 at the study’s beginning is higher than in latent state 1 with respect to small museums. Regarding the covariates that affect the transition through the latent states, Table 7 shows that only the size of the museums influences these Conversely, states 2 and 3 are those of museums with a medium and probabilities. More specifically, we observed that the increase in the high level of sustainability, respectively. In particular, those in state 3 size of the museums corresponds to higher transition probabilities from 5 C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 Table 6 Estimates of the regression parameters of the initial probability to belong to the other latent states with respect to the first state under the model with 3 latent states. In bold are the significant regression parameters for 𝛼 = 0.05. Effect 𝛽̂12 𝛽̂13 North-East 0.3596 0.6034 Macro-area Center 0.2065 0.2936 ref: North-West South −0.3684 −0.4129 Island −0.2358 −0.4192 Size Medium 0.9522 1.6127 ref: Small Large 1.4343 3.1067 Fig. 4. Cartogram of the museums belonging to state 2 in 2018 per province. Fig. 3. Cartogram of the museums belonging to state 1 in 2018 per province. state 1 to states 2 and 3. The contrary can be stated for the transition probabilities from state 3 to states 2 and 1 and the transition probability from latent state 2 to state 1. Generally, about 16% of the museums show a movement through the latent states in the years considered. In particular, the museums in the South of Italy and the Islands show higher transitions, with a percentage of 21% and 24%, respectively. Follow the North-West and North-East of Italy, with a percentage of museums that transit to an- Fig. 5. Cartogram of the museums belonging to state 3 in 2018 per province. other latent state of about 14%. In the Center of Italy, the percentage is 13%. In particular, the museums located in the North-East, Center, and South of Italy show, in percentage, higher transitions from lower states to higher ones in the years 2018–2019, while in the years 2019–2021, the highest percentage of movements through lower states to higher Piemonte, Puglia, and Sardegna. Table 11 in Appendix illustrates the ones are shown by the museums located in North-East, North-West, percentage of museums that show a transition between the latent Center, and South Italy. states per region for the years 2018–2019 and 2019–2021. Examples Regarding the transitions from higher to lower states, in the years of museums that show significant transitions through the states, for 2018–2019, the museums located in the South of Italy and the Island example, from state 3 to state 1, are the Museo del Bergamotto in show the higher percentage of the transitions. In the years 2019– Calabria or the Casa Museo ‘‘Gaia da Camino’’ in Veneto. On the other 2021, museums with a higher percentage of transitions between higher hand, some examples of museums that show movements from state 1 to lower states were located in the North-West, Center of Italy, and to state 3 are the Museo Diocesiano in Campania and the Museo Civico the Island. In more depth, the regions that showed the majority of di Bari in Puglia. movements in 2018–2019 from lower latent states to higher are Emilia- Figs. 3–5 show the provincial distribution (with ten elements in each Romagna, Lombardia, Toscana, Piemonte, Puglia, and Sardegna. In class to make the representation more straightforward) of the museums 2019–2021 are Calabria and Piemonte. On the contrary, the regions for each latent state in 2018. The museums clustered in the first latent that move significantly to a lower state in 2018–2019 are Emilia- state are distributed throughout the national territory. Conversely, the Romagna and Sardegna, while in 2019–2021, they are Emilia-Romagna, museums belonging to latent states 2 and 3 are mainly concentrated in 6 C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 Table 7 Estimates of the regression parameters of the transition probabilities under the model with 3 latent states. In bold are the significant regression parameters for 𝛼 = 0.05. Effect 𝛾̂12 𝛾̂13 𝛾̂21 𝛾̂23 𝛾̂31 𝛾̂32 North-East 0.171 −0.273 −0.162 0.040 −0.316 −0.228 Macro-area Center 0.073 −0.226 −0.081 0.040 −0.358 −0.237 ref: North-West South 0.292 0.654 0.058 0.330 0.386 −0.254 Island −0.115 −0.566 0.449 −0.370 1.096 0.364 Size Medium 0.929 1.740 −0.937 0.570 −2.195 −0.776 ref: Small Large 1.446 2.993 −1.268 1.080 −3.391 −1.432 the Center and North of Italy. Similar results were observed in 2019 values and evaluate (possible) changes over time. To this goal, latent and 2021 (Fig. 6 in Appendix). transition models help us group museums into specific classes (or Finally, regarding the size of the museums, about 17% of the status) and understand stability and change within these sustainabil- museums coded as small show transitions across the latent states over ity latent classes over time. This involves identifying museums that the years. Similarly, about 14% of the museums coded as medium, remain in the same state across years and those that move between and 16% of museums identified as large show movements through the states. On the other hand, this classification could be employed to latent states in the years under analysis. classify a-posteriori museums not involved in all the questionnaire administrations, representing in this way a valuable tool for inferring 6. Discussion the sustainability status of similar institutions that were not directly surveyed. Museums and cultural heritage sites represent a powerful resource Therefore, by using data carried out by the Italian Survey on Mu- for local development. They can inspire and help regenerate local seums and Other Cultural Institutions, for the years 2018, 2019, and economies, attract visitors, and bring in revenue. There is also growing 2021, we found that Italian museums are divided into three homo- evidence that they can contribute to social cohesion, civic engagement, geneous classes. After having defined and computed five (possible) health, and well-being. For several decades now, cities and regions dimensions of museums’ sustainability, the main empirical findings have drawn on these resources to implement heritage-based actions suggest (i) that museums show a tendency to remain in the same state as part of their broader economic development strategies. National, in the period selected, (ii) the territorial area in which the museum municipal, and regional governments, the museum community, and is located and its size both affect the probability to being in a specific other stakeholders are increasingly interested in these issues. class at the beginning of the study and (iii) only the size of the museum Sustainability, and all its variations, is an open challenge with affects the probability to move from a state to another in the period which many museums and places of culture have yet to deal system- here considered. Although some of our empirical findings are notewor- atically. It is a process that began a few years ago but is struggling to thy, it could be interesting to consider additional dimensions and/or find widespread application. The development of an overall strategy other museum covariates, jointly with a greater level of territorial through public policies at local, regional, and national levels is still at disaggregation (e.g., at the provincial level) which allows for a more an early discussion stage. in-depth analysis of local development. As the International Council of Museums (ICOM) has pointed out, the contribution of museums to sustainable development is now an 7. Conclusions essential element of its agenda. According to , ‘‘[...] sustainability is the dynamic process of museums based on the recognition and preservation of tangible and intangible heritage with the museums responding to the needs This paper proposes a classification of Italian museums to group of the community. To be sustainable, museums, through their mission, must them by different sustainability values and evaluate possible changes be an active and attractive part of the community by adding value to the over time. By employing a latent transition analysis, we were able heritage and social memory.’’. to group museums into specific classes and understand stability and Culture and the museum sector can make a contribution to nu- change within these sustainability latent classes over time. Using data merous Sustainable Development Goals (SDGs) for the United Na- from the Italian Survey on Museums and Other Cultural Institutions for tions 2030 Agenda related to the main challenges facing contemporary the years 2018, 2019, and 2021, we found that Italian museums can society, starting with Target 11.4, within Goal 11 ‘‘Cities and sus- be divided into three homogeneous classes. In addition, our findings tainable communities’’ which concerns the strengthening of efforts to suggest that museums tend to remain in the same state over the selected safeguard cultural and natural heritage and its connection with socio- period, the territorial area and museum size both affect the probability economic development (above all) at a territorial level. Thanks to their of being in a specific class at the beginning of the study, and only the widespread presence in the territory, museums constitute a cultural size of the museum affects the probability of moving from one state to infrastructure and are in relation to the urban and territorial context. another during the period considered. It is precisely this last aspect, of integration with territorial policies, However, it should be noted that to measure the five dimensions of that requires in-depth work and where a great opportunity for growth sustainability, different items were used for each year due to changes could open up, with the possibility of museums interacting with the in survey questionnaires. This discrepancy means that the dimensions bodies responsible for defining policies for sustainable development do not completely overlap over time, representing a limitation of our oriented towards local development. But, to date, there is considerable study. While the core aspects of sustainability are consistently targeted, difficulty in finding quantitative information on museums and their the inconsistency in specific survey items may introduce some vari- characterisation at the territorial level. In this sense, the Italian Survey ability in the measurements. Acknowledging this limitation is crucial on Museums and Other Cultural Institutions represents a valuable for interpreting our findings and highlights the need for developing source of reliable information that can be used, on the one hand, more standardised and consistent measurement frameworks in future to classify Italian museums to group them by different sustainability research. 7 C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 Table 8 Variables measured in the context of the Italian Survey on Museums and Other Cultural Institutions conducted by Istat in 2018 and used to create the indicators services, supports, activities, web, and digitalisation. Dimension 2018 Services Booking of tickets and visits Parking The following services were Cloakroom available to the public: Cafeteria and restaurant Food and drink vending machines Furnished spaces for visitors to park Bookshop Free Wi-Fi Reception and entertainment for children (playrooms, etc.) Assistance for disabled visitors Facilities for disabled visitors Other Supports Reception point for information and orientation Information panel and/or map of the visit routes at the entrance The following visiting supports Signage to indicate the visit routes were available: Paper information material (brochures, leaflets, mobile cards, etc.) Panels and/or captions for describing the individual works Audio and/or video guides Applications for smartphones and tablets Interactive displays and/or virtual reconstructions (touch screen, video...) QR Code and/or proximity systems (Bluetooth, Wifi, etc.) Tablets available to the public Video/multimedia room Routes and information materials dedicated to children Materials and information support to encourage use by disabled people Complete name of the institute outside the headquarters Indication of the opening hours outside the office Signage – informative, directional and identifying – outside the premises, on the roads approaching the structure Activities Exhibitions and/or temporary exhibitions Educational workshops The museum/institute carried out: Guided tours Web Dedicated website Catalogue accessible online for visitors The following web services were available: Online ticketing service Possibility to visit virtually the museum/institute via the Internet Social media accounts (Facebook, Twitter, Instagram, Pinterest, Foursquare, etc.) Links to digital maps and/or geographic coordinates for location Digitalisation A digital inventory A digital scientific catalogue The museum/institute has: The findings of this study provide valuable insights that can sig- CRediT authorship contribution statement nificantly inform museum policy from local, regional, and national perspectives. By categorising Italian museums based on their sustain- Carla Galluccio: Formal analysis, Data curation. Francesca Gi- ability levels and examining changes over time, policymakers can ambona: Supervision, Project administration, Methodology, Conceptu- better understand the stability and dynamics within the museum sec- alization. tor. This understanding allows for targeted interventions to enhance sustainability practices across museums. Specifically, the study’s results Data availability highlight the influence of geographic location and museum size on sustainability, suggesting that tailored strategies may be necessary Data will be made available on request. for different regions and museum sizes. Recognising the tendency of museums to remain in the same sustainability state underscores the Acknowledgement need for consistent and long-term policy support to encourage progress. Integrating these findings into policy frameworks can help develop We acknowledge funding from Next Generation EU, in the con- more effective programmes that preserve cultural heritage and promote text of the project PE5 CHANGES - Spoke9 ‘‘Cultural Heritage Active local development, economic regeneration, and social cohesion. Addi- Innovation for Sustainable Society’’ (CUP B53C22004010006). tionally, expanding the analysis to include additional dimensions and finer territorial disaggregation could provide more insights, enabling Appendix policymakers to elaborate more specific and significant cultural policies that align with the Sustainable Development Goals (SDGs) and foster See Tables 8–11 and Fig. 6. sustainable community growth. 8 C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 Table 9 Variables measured in the context of the Italian Survey on Museums and Other Cultural Institutions conducted by Istat in 2019 and used to create the indicators services, supports, activities, web, and digitalisation. Dimension 2019 Services Free Wi-Fi connection in the exhibition area Spaces and/or facilities for children (reception and/or The following services were entertainment, playrooms, changing tables, etc.) available to the public: Spaces and/or facilities for disabled visitors (e.g. equipped bathrooms, ramps, elevators, etc.) Supports Paper information material (brochures, leaflets, mobile cards, etc.) Information panel and/or map of the visit routes at the entrance The following visiting supports Signage to indicate the visit routes were available: Panels and/or captions for the description of the individual works Audio guides and/or video guides Applications for smartphones and tablets Video and/or touch screen Multimedia supports (interactive displays, virtual reconstructions, augmented reality, etc.) QR Code and/or proximity systems (Bluetooth, Wifi, etc.) Tablets available to the public Routes and information materials dedicated to children Information supports to facilitate visits by disabled people Information outside the office on opening hours Signage outside the headquarters on the roads approaching the structure Activities Guided tours Educational workshops (e.g. activities for children, teenagers and school groups) The museum/institute carried out: Assistance for disabled visitors Web General information on access methods (e.g. address, timetables, fares, routes, etc.) The following web services were available: Information on educational activities/workshops Information on events and/or temporary exhibitions User services (e.g. purchase of tickets, booking visits, purchase of gadgets and/or books, etc.) Links to access social media Link to digital maps and/or geographical coordinates for locating the office Catalogues in digital format of the assets owned (photos, videos, databases, etc.) Information on the research activities carried out (e.g. publications and research projects) Information on collaboration or partnership projects with third parties (e.g. museum or tourist networks and/or circuits) Exhibitions or virtual tours Social media account (Facebook, Twitter, YouTube, Instagram, Flikr, Linkedln, Pinterest, Foursquare, etc.) Digitalisation A digital collections (partially o totally) The museum/institute has: 9 C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 Table 10 Variables measured in the context of the Italian Survey on Museums and Other Cultural Institutions conducted by Istat in 2021 and used to create the indicators services, supports, activities, web, and digitalisation. Dimension 2021 Services Room/laboratory for teaching, study, research and/or conference activities Video/multimedia room The following services were Free Wi-Fi connection available to the public: Supports Applications for smartphones and tablets Video and/or touch screen The following visiting supports Multimedia supports (interactive displays, virtual were available: reconstructions, augmented reality, etc.) QR Code and/or proximity systems (Bluetooth, WiFi, etc.) Tablets available to the public Activities Educational workshops (for children, teenagers and school groups) Thematic and/or educational tours specifically dedicated to children The museum/institute carried out: Guided tours Conferences, conferences and seminars Live shows and/or cultural entertainment initiatives Exhibitions and/or temporary exhibitions Web Online ticketing service (for booking visits, purchasing tickets, etc.) Social media accounts (Facebook, Twitter, YouTube, Instagram, Pinterest, etc.) The following web services were available: Online virtual tours Online guided tours and/or alternative online methods of visiting the museum/institute Online educational workshops (for children, teenagers and school groups) Online conferences, conferences and seminars Catalogues of the heritage owned in digital format (photos, videos, databases, etc.) Digitalisation Digital assets displayed to the public It movable assets digitalised The museum/institute has: Table 11 Percentage of museums that show a transition between the latent states per region for the years 2018–2019 and 2019–2021. The percentages are computed with respect to the total of museums that show a movement (and not the total of museums under analysis). Here, ‘‘S’’ means ‘‘latent state’’, while ‘‘S1-S2’’, for example, means ‘‘transition from latent state 1 to latent state 2’’. Year 2018–2019 2019–2021 Latent states S1-S2 S2-S3 S1-S3 S2-S1 S3-S2 S3-S1 S1-S2 S2-S3 S1-S3 S2-S1 S3-S2 S3-S1 Region Abruzzo 3.3 0.0 5.3 0.0 0.0 7.1 2.9 4.5 0.0 0.0 0.0 0.0 Basilicata 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Calabria 0.0 5.0 5.3 5.0 0.0 14.3 14.7 0.0 0.0 1.6 0.0 5.0 Campania 0.0 0.0 5.3 7.5 0.0 0.0 8.8 4.5 0.0 3.2 2.8 10.0 Emilia-Romagna 6.7 20.0 0.0 10.0 8.3 14.3 8.8 13.6 16.7 11.1 11.1 15.0 Friuli-Venezia Giulia 3.3 5.0 0.0 5.0 8.3 0.0 8.8 0.0 8.3 4.8 0.0 0.0 Lazio 0.0 5.0 10.5 10.0 8.3 7.1 5.9 0.0 25.0 6.3 8.3 5.0 Liguria 6.7 0.0 21.1 2.5 8.3 0.0 0.0 9.1 0.0 0.0 11.1 0.0 Lombardia 0.0 20.0 21.1 5.0 8.3 7.1 5.9 13.6 8.3 9.5 11.1 0.0 Marche 6.7 0.0 0.0 5.0 0.0 14.3 5.9 4.5 0.0 4.8 0.0 5.0 Molise 0.0 0.0 0.0 0.0 0.0 0.0 2.9 0.0 0.0 0.0 0.0 0.0 Piemonte 16.7 15.0 26.3 17.5 0.0 7.1 11.8 9.1 8.3 6.3 11.1 15.0 Puglia 13.3 0.0 10.5 5.0 0.0 7.1 2.9 4.5 16.7 6.3 0.0 5.0 Sardegna 13.3 5.0 10.5 7.5 33.3 7.1 2.9 0.0 0.0 19.0 11.1 20.0 Sicilia 3.3 0.0 5.3 2.5 8.3 7.1 8.8 4.5 8.3 12.7 0.0 10.0 Toscana 20.0 20.0 5.3 10.0 8.3 0.0 5.9 22.7 0.0 12.7 5.6 0.0 Trentino-Alto Adige 3.3 0.0 0.0 2.5 0.0 0.0 0.0 0.0 0.0 1.6 2.8 0.0 Umbria 0.0 5.0 0.0 0.0 0.0 0.0 0.0 9.1 8.3 4.8 8.3 0.0 Valle d’Aosta 0.0 0.0 0.0 2.5 0.0 0.0 0.0 0.0 0.0 1.6 0.0 0.0 Veneto 3.3 0.0 0.0 2.5 8.3 7.1 2.9 0.0 0.0 3.2 16.7 10.0 10 C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 Fig. 6. Cartograms of the museums belonging to states 1 to 3 in 2019 and 2021 per province. References UNESCO. Synthetic report submitted to the intergovernmental committee for the intangible cultural heritage of humanity at its eighth session in Baku, Azerbaijan, 2-7 December 2013. 2013. International Council of Museums. Museums and sustainable development: How Throsby D. Cultural capital. J Cultural Econ 1999;23:3–12. can ICOM support, in concrete terms, the museum community’s sustainable Snowball JD. The economic, social and cultural impact of cultural heritage: development projects?. 2011, URL. methods and examples. In: Rizzo I, Mignosa A, editors. Handbook on the 11 C. Galluccio and F. Giambona Socio-Economic Planning Sciences 95 (2024) 101998 economics of cultural heritage. Cheltenham: Edward Elgar Publishing; 2013, p. Collins LM, Lanza ST. Latent class and latent transition analysis: with applications 438–55. in the social, behavioral, and health sciences. New York: John Wiley & Sons; Grincheva N. Museums as actors of city diplomacy: From ‘‘hard’’ assets to ‘‘soft’’ 2010. power. In: Amiri S, Sevin E, editors. Palgrave mac-millan series in global public Magidson J, Vermunt JK. Latent class models. In: Kaplan D, editor. The sage diplomacy. Cham: City Diplomacy; 2020, p. 111–36. handbook of quantitative methodology for the social sciences. Thousand Oaks: Beccherle P, Oliva S, Lazzeretti L. Brescia as a cultural city: The role of museums Sage Publications; 2004, p. 175–98. and digital technologies. Sci Regionali 2023;00:1–26. Lazarsfeld PF. The logical and mathematical foundation of latent structure anal- of Europe C. Recommendation of the committee of ministers to member states ysis. In: Stouffer SA, editor. Measurement and prediction. Princeton: Princeton on the European cultural heritage strategy for the 21st century. 2017, URL. University Press; 1950, p. 361–412. European Commission. Final communication from the commission to the Euro- Collins LM, Graham JW, Rousculp SS, Fidler PL, Pan J, Hansen WB. Latent pean parliament, the European council, the council, the European economic and transition analysis and how it can address prevention research questions. NIDA social committee and the committee Of The Regions. a new European agenda research monograph, vol. 142, 1994, p. 81–111. for culture. 2018, URL. Magidson J, Vermunt JK. Latent class factor and cluster models, bi-plots, and Sacco P, Ferilli G, Blessi GT. From culture 1.0 to culture 3.0: Three socio- related graphical displays. Sociol Methodol 2001;31(1):223–64. technical regimes of social and economic value creation through culture, and Nylund KL, Asparouhov T, Muthén BO. Deciding on the number of classes in their impact on European cohesion policies. Sustainability 2018;10(3923):1–23. latent class analysis and growth mixture modeling: A Monte Carlo simulation Florida R. Who’s your city? how the creative economy is making where to live study. Struct Equation Model Multidiscipl J 2007;14(4):535–69. the most important decision of your life. New York: Basic Books; 2007. Bartolucci F, Pandolfi S, Pennoni F. LMest: An r package for latent Markov Panzera E. Cultural heritage and territorial identity: synergies and develop- models for longitudinal categorical data. J Statist Soft 2017;81(4):1–38. ment impact on European regions. advances in spatial science series. New York: Springer Publishing; 2022. Falatooni E, Selen W, Kerr D. A new framework for selecting composite indicators to assess sustainability of a destination. Athens J Tourism 2016;3(1):7–24. Francesca Giambona is Associate Professor in Economic Statistics at the Department Cerquetti M, Montela MM. Museum networks and sustainable tourism man- of Statistics, Computer Science, Applications ‘‘G. Parenti’’, Florence (Italy). She has agement. The case study of marche region’s museums (ITALY). Enlight Tour Ph.D. in applied statistics and the main focus of her academic research is on, skill Pathmaking J 2015;5(1):100–25. mismatch, well-being, economic uncertainty, tourism, culturale heritage, statistical Pop IL, Borza A. Factors influencing museum sustainability and indicators for models specification in economics museum sustainability measurement. Sustainability 2016;8(1):101–23. 12