Adult education as a common good: conceptualisation and measurement

ABSTRACT The article outlines a theoretical framework for conceptualising adult education – and more broadly, lifelong learning – as a common good. It argues that the extent to which adult education as a common good is accomplished in a given society/country reflects its accessibility, availability, affordability and the social commitment to its functioning and that it depends on a country’s specific institutional arrangements. Building on this conceptualisation and using data from the Adult Education Survey (AES), the Labour Force Survey (LFS) and the Continuing Vocational Training Survey (CVTS) for 24 European countries, the authors develop a composite index, based on these four dimensions, which measures the extent to which adult education as a common good is practiced in a given country. This index can be used to assess the effectiveness of national policies in the sphere of adult education across Europe. The results indicate substantial cross-country differences, with North European states and Luxembourg scoring best and Romania scoring worst. Finally, applying cluster analysis, the article identifies six distinctive clusters of countries with regard to the extent of adult education as a common good; the authors designate these cluster categories as reality, feasible, ambiguous, problematic, possible and invisible.


Introduction
Recently, important structural factors (demographic decline and changes in the labour market) have considerably increased the demand for adult education and enhanced its role for social development (Cedefop, 2010;European Commission, 2013). Taking into account these tendencies, the European agenda for adult learning sets the goal to 'enhance the possibilities for adults, regardless of gender and their personal and family circumstances, to access high-quality learning opportunities at any time in their lives, in order to promote personal and professional development, empowerment, adaptability, employability and active participation in society' (European Commission, 2011, p. 3). This goal clearly reflects the idea that adult education should be available to all people, implying that it is regarded as a public good. However, all studies of participation in adult education reveal that it is determined by several factors and that 'the adults most in need of education and training are those with the least access to lifelong learning opportunities: adults with low level or no qualifications, those in low-skilled occupations, the unemployed and economically inactive, older people and the least skilled' (European Commission/EACEA/Eurydice, 2015, p. 8). In addition, data show that fewer public than CONTACT Petya Ilieva-Trichkova petya.ilievat@gmail.com Institute for the Study of Societies and Knowledge, 13A Moskovska str., Sofia 1000, Bulgaria private resources are invested in training and that the 'role of the state is less than that of companies and families ' (European Commission, 2013, pp. 10-14, 63). The trend towards privatisation of adult education is clearly evident in all countries. Such tendencies provide conflicting arguments as to whether adult education is a private, public or common good. In fact, a few studies explore formal education as a common good, such as the research by Rita Locatelli (2016) on schooling and those of Simon Marginson (2016) and Locatelli (2018a) on the case of higher education. However, to the best of our knowledge, no available research is focused on adult education.
Against this background, this article aims: 1) to outline a theoretical framework for conceptualising adult educationand more broadly, lifelong learningas a common good; 2) to develop an index for measuring the extent to which adult education as a common good is accomplished in a given country; and 3) to show whether countries fall into distinct clusters with regard to the extent of practice of adult education as a common good.

Theoretical considerations
Concepts and approaches 'The common good', 'common goods' and 'the commons' are concepts widely discussed in philosophy, economics and political science. Recently, they have attracted the attention of scholars in sociology and educational science as well.
The tradition of philosophical study of the common good dates back to Plato and Aristotle. The concept was further developed in the works of numerous philosophers and political theorists, including Thomas Aquinas, Niccolò Machiavelli, John Locke, Jean-Jacques Rousseau, Adam Smith, Karl Marx, John Stuart Mill, Jacques Maritain, John Rawls and so on. Despite the significant differences in how the common good is conceptualised in the various philosophical and political doctrines, it is generally viewed as a norm that unifies a given (political) community and is closely related to 'public goods' and 'the public interest' (Etzioni, 2015;Hazelkorn & Gibson, 2017;Pusser, 2006). The discussion on common goods has received a new impetus around the notion of the commons, which has gained importance as a result of several ecological challenges arising from the growth of industry, cities and population. The commons' paradigm has brought to the fore the question of how best to govern natural resources used by many individuals in common, and the idea that community governance of resources is an effective vector for development that could counterbalanceor even replacestate intervention and privatisation (Locher, 2016;Ostrom, 1990).
Although following different theoretical perspectives, all discussions on common goods 1 refer to what is shared and beneficial for all, or most, members of a given community or to what could be produced by collective actions and active participation in the public and political sphere; thus, common goods include both goods that serve no particular group and those that will serve members of the yet unborn generations. Rita Locatelli (2016, p. 154) underlines that 'despite the high degree of "flexibility", the term common goods holds a minimum semantic core that can be traced as common in all socio-political claims' and quotes Lorenzo Coccoli, who summarises the main characteristics forming this minimum semantic core as follows: '(1) the opposition of the concept of common goods to the processes of privatisation and marketisation, and more generally to the dynamics of neoliberalism, (2) the re-composition of networks of social solidarity within communities and (3) the development of instruments of participatory democracy'. We think that three additional characteristics of common goods should be emphasised. First, accepting the common good perspective does not mean denying individual freedom. Following Jacques Maritain, it is worth asserting that society exists for the person and the person exists for society in a way that maintains the dignity of free persons. The common good presupposes persons and flows back upon them; it must be good for everyone, and cannot involve trade-offs of one person's good for another's (Maritain, 1985(Maritain, /1947. Second, common goods are not a sum of private goods. They are 'not simply a means for attaining the private good of individuals' (Hollenbach, 2002, pp. 81-2), but have their own value. That is why, from a common goods perspective, 'it is not only the "good life" of discrete individuals that matters, but also the goodness of the life that humans hold in common' (Deneulin & Townsend, 2007, p. 24). Third, common goods are socially embedded. Different communities may have different understandings of the common good/common goods. That is why the common goods can only be defined within diverse contexts and conceptions of well-being and common life (Deneulin & Townsend, 2007;UNESCO, 2015, p. 78).
A specific approach to conceptualising private goods, public goods and common goods is provided by neoclassical economics. Based on two criteriaexcludability and rivalryneoclassical economists distinguish three types of goods: private, public and common. Unlike private goods, public goods are non-excludable and non-rivalrous, i.e. individuals cannot be excluded from the use and benefits of these particular goods, and their use by one individual does not reduce their availability to others. A good is non-rivalrous if one person's consumption of the good does not reduce the benefits someone else obtains from its consumption (Cowen, 2007;Deneulin & Townsend, 2007;Samuelson, 1954). According to Marginson (2016), a common good is nonexcludable but rivalrous. We accept Locatelli's (2018a) understanding that public goods possess non-excludable and non-rivalrous characteristics presented in terms of consumption of a commodity, whereas common goods are characterised with non-excludable and non-rivalrous characteristics presented in terms of participation and generation of the goods themselves.

A common good perspective towards adult education
Taking into account the above-mentioned theoretical considerations, our discussion of adult education as a common good will be guided and organised according to two important perspectives. First, as shown above, the philosophical tradition in the study of the common good refers to both 'the common good' and 'a common good' or 'common goods'. In the following analysis, in attempting to combine ideas from philosophical discussions on common goods with its understanding in neoclassical economics, we will try to ascertain whether adult education could be regarded as a common good. Second, we differentiate between two perspectives that, although closely connected, delineate two distinct relations between adult education and the three types of goods (private, public and common). The first perspective explores whether, and to what extent, adult education itself is implemented, and can be defined, as a private, public or common good. The second one tries to reveal what kind of private, public and common goods are produced by adult education 3 . In this article, we adhere to the first analytical approach.
The discussion of possibilities for understanding adult education as a common good outlines new and fruitful perspectives for rethinking and reimagining the essence of adult education and its role in contemporary societies and thus allows us to go beyond the limits of the concept of public goods (UNESCO, 2015). Although closely related to the notion of public goods, as shown above, the idea of common goods has its own specific meaning 2 . Locatelli's study (Locatelli, 2018b) suggests that the frameworks of education as a public good and as a common good may be seen as a sort of continuum in line with the aim of developing democratic political institutions that enable citizens to have a greater voice in the decisions that affect their well-being. While 'within a public good approach the point of view remains essentially focused on public institutions that should provide the regulatory framework for the development of democratic educational systems, considering education as a common good implies that education is a collective shared endeavor, both in its production and in its benefits' (Locatelli, 2016, p. 158).
A common good perspective is especially suitable for studying adult education as it corresponds to the specificity of this form of education. Adult education and adult learning are the most important forms of lifelong learning. They include different kinds of knowledge and skills (institutionalised and informal, organised and sporadic, purposeful and unintended) in different perspectives (of the individual and social; of employment and citizenship; of leisure and work), within diverse, public and private, settings. In addition, they involve a wide range of actors (public and private institutions, NGOs, civic and religious organisations) acting at different levels. At first glance, it seems that adult education is mainly a private good and cannot be viewed as a common good. Taking into account Simon Marginson's discussion on higher education (2007,2011), we accept that adult education is intrinsically neither a private nor a publicor commongood. It is potentially rivalrous or non-rivalrous, and potentially excludable or non-excludable, which means that, being nested in the wider social and cultural settings, adult education as a good is policy sensitive and, consequently, varies by time and place.
Defining adult education as a common good implies acknowledging that it is indispensable for human well-being in contemporary societies. If we accept Amartya Sen's definition of human well-being as the freedoms that people have reason to choose and value (Sen, 1999), or Martha Nussbaum's list of central human capabilities (Nussbaum, 2011), it is beyond doubt that human well-being in modern dynamic and liquid societies (Bauman, 2000) would not be possible without the development of adult education as a common good. Adult education as a common good presupposes, and requires, that it develops as an inclusive process shared by, and beneficial to, all or most members of a given community/society. Viewing adult education as a common good means that it unfolds in mutual social relationships, in and through which human beings enhance their well-being; it is therefore a kind of collective endeavour in which are involved different and diverse social institutions. Recently, Simon Marginson (2016, p. 29) has also argued that 'the notion of the common good is a denial that society is and should be composed of atomized individuals living in isolation from one another'.
Adult education is a common good in and for a given community/society providing it 'is immanent within the relationships that bring this community or society into being' (Hollenbach, 2002, p. 9). This understanding of adult education emphasises its complex nature and the plurality of its roles and values, which go beyond its instrumental function, and acknowledges its empowering and transformative mission as well (Boyadjieva & Ilieva-Trichkova, 2018). Thus, adult education as a common good is closely related to concepts such as justice, rights (Walker & Boni, 2013), solidarity and equality (Marginson, 2016).
Our conceptualisation of adult education as a common good is based on an attempt to combine ideas from philosophical discussions on common goods with a foundational approach from neoclassical economics. The philosophical and neoclassical economic perspectives share an emphasis on the inclusive/shared character of common goods, manifested in both their production and use. That is why, we will proceed with a discussion of adult education as a common good from a perspective that focuses on participation, i.e. we will try to reveal who participates in adult education and who is engaged and involved in its provision. This perspective allows us to move from the theoretical conceptualisation of adult education as a common good to its measurement and develop an index for measuring the extent to which adult education has been realised as a common good in a given country.
Access to adult education is the element of the adult education system that is most soaked in politics, for in many countries, the state sets standards for accreditation of institutions providing adult education, regulates the number of available places, provides financial incentives for adults returning to education and so on. That is why we argue that the extent to which adult education is accomplished as a common good in a given society/country reflects its accessibility, availability, affordability and the commitment of society to this education. Adult education is a common good when it is accessible to a growing number of people and when policies have been implemented to reduce inequalities in, and barriers to, its access. The realisation of adult education as a common good depends on the country's specific institutional arrangements.

Data
The empirical bases of our study are macro-level data drawn mainly from the Adult Education Survey (AES) 2011. This survey, conducted via a random sampling procedure, targets people aged 25-64 years who live in private households; it is part of the European Union (EU) statistics on lifelong learning and collects primary data on participation in education and training. The survey takes place every 5 years. However, the micro data from AES 2016 are not yet available. The survey allows us to adopt a wide comparative perspective, as it was conducted in many European countries. The number of these countries in AES 2011 was 30. These data are complemented by data from the Labour Force Survey (LFS) 2011 and the Continuing Vocational Training Survey (CVTS) 2010, which have been extracted from the Eurostat website.

Index calculation
We developed an Index of Adult Education as Common Good (IAECG) to capture the extent to which adult education has been realised as a common good in a given country. The Index includes indicators, referring to four dimensions of adult education: accessibility, availability, affordability and social commitment to adult education.
Accessibility of adult education refers to the ability of people from all backgrounds to access and benefit from adult education on a reasonably equal basis (for higher education, see Usher & Medow, 2010, p. 1). It is measured with two indicators: participation rate and Adult education equity index (AEEI).
Availability of adult education relates to resources (both institutional and individual) that are suitable for adult education and enable people's participation in it. It is measured through the following indicators: distance to place of education, distance learning, suitable offers, access to information and IT equipment.
Affordability of adult education captures the cost of adult education in relation to people's financial means and the support they receive to overcome obstacles to participation in adult education. It is measured with the following indicators: enterprise expenditure on continuing and vocational training (CVT) courses, acceptable cost of education and employer's support.
Social commitment to adult education refers to the engagement of various actors (public, private, institutions of civil society, religious organisations) in the provision and funding of adult education and is measured with two indicators: engagement of various institutions with payment and engagement of various institutions with provision. Table 1 reports the name and includes description of the source and the time it refers to for each indicator used to measure a given dimension of adult education as a common good. Some of our indicators are based on objective data (such as expenditure or participation rate) and some, on subjective data (respondents' assessment/personal opinion of the state of affairs, e.g. whether the costs are acceptable). We think that both types of indicators should be taken into account when assessing whether adult education is or is not realised as a common good in a given country.
For the calculation of the IAECG, we have followed the methodology used in the report 'Don't Panic: Findings of the European Catch-Up Index 2015' (Lessenski, 2016), prepared by the Open Society Institute -Sofia. This methodology has been chosen for two main reasons. First, it offers very clear guidelines for all statistical procedures. Second, it allows exploring the dynamics of the indexes over time as a next step of the analysis. More specifically, we standardised the values of different indicators described above according to a statistical procedure, which recalculates them on one and the same scale while simultaneously preserving the order and proportions between them. The standardising was done following the normalisation method of z-scores, which uses mean weighed score and standard deviation. Using this procedure, the distribution of the values in the countries for each of the indicators was translated and the mean 0 and dispersion 1 were calculated, while the order and proportions between the values for the different countries were preserved. Then, we transformed the standardised values into scores ranging from 0 to 100. Values smaller than 0 and bigger than 100 ('extreme values') received scores 0 and 100, respectively. Own calculations based on data from the AES (2011) 5.00% • Distance learning % of people who reported distance learning using online or offline computer and/or distance learning using traditional teaching material for the most recent formal activity and for the three non-formal activities, out of the total number of people who participated in formal and non-formal education and training Own calculations based on data from the AES (2011) 5.00% • Suitable offers % of people who have not marked 'No suitable education or training activity (offer)' as an obstacle to participation in lifelong learning Own calculations based on data from the AES (2011) 5.00% • Access to information % of adults who reported they had access to information on learning possibilities Own calculations based on data from the AES (2011) 8.33%

Social commitment
• Engagement of various institutions with payment % of those who have partial or full payment for the first randomly selected non-formal education and training activity during the last 12 months paid by employer or prospective employer, public employment services or other public institutions Own calculations based on data from the AES (2011) 12.50% • Engagement of various institutions with provision % of those who reported institutions other than formal education and non-formal education and training institutions as being providers of the first randomly selected non-formal education and training activity during the last 12 months.
Own calculations based on data from the AES (2011) 12.50% Each of the four dimensionsaccessibility, availability, affordability and social commitment contains different numbers of basic indicators. We assign equal importance to each one of the four dimensions. The level of importance of each indicator within a given dimension is calculated by dividing the overall weight of the dimension by the number of indicators. For the sake of transparency, we are prepared to send the indicators themselves on request so that users can use weights other than the ones proposed in the article. The weights are also provided in Table 1.
After the weighting, we constructed the composite IAECG, where the index comprises all four categories; we then calculated it for 24 countries. This was the number of countries for which we found information regarding the indicators of interest. The Index ranges between 0 and 100, where a value of 0 means that adult education is not at all realised as a common good in a given country and a value of 100 means that adult education is realised to the greatest extent as a common good in a given country. The same interpretation refers to all four dimensions: accessibility, availability, affordability and social commitment to adult education.

Cluster analysis
After calculating the overall index scores and the scores in each dimension, we applied cluster analysis. In general, this type of analysis consists in a collection of algorithms used to classify objects such as countries, species and individuals in order to reduce the dimensionality of a data set by exploiting the similarities/dissimilarities between cases (OECD, 2008). In this case, the cluster analysis was made with regard to the categories comprising the overall score of the index of adult education as a common good. More specifically, agglomerative hierarchical cluster analysis of the Complete Linkage (the furthest neighbour) with the help of the Stata 14 statistical package was used to discover the number of clusters of countries. This type of analysis tends to produce very compact clusters, where each member of a cluster must be close to every other member of the same cluster (Kaufman & Rousseeuw, 2005). The similarity of cases was measured using the standard Euclidean distance between them. Having in mind that cluster solutions are sensitive to the specific clustering algorithm used (Kantardzic, 2011;Xu & Wunsch, 2009), we also employed another hierarchical clustering algorithm, i.e. the Ward Linkage. The results of the second algorithm were consistent with the results produced using the first algorithm. Table 2 shows the overall Index scores and the respective country rank for adult education as a common good. The countries are sorted according to their overall IAECG score. Table 2 highlights the significant differences between countries with regard to the extent to which adult education as a common good has been accomplished, with the North European countries and Luxembourg scoring the best and Romania, the worst. Specifically, the country with the highest overall index score is Luxembourg, followed by Sweden, Denmark, Finland, Norway and the Netherlands: this means that adult education as a common good has been realised to the greatest extent in these countries. However, as of 2011, it seems that in none of the countries studied has adult education as a common good been practiced completely. In most of the countries, substantial developments are accompanied by problems in reaching this goal. These countries include Norway, France, Belgium, Slovakia, Spain, Austria, Malta, Cyprus, Hungary, Germany, the Czech Republic, Slovenia and Estonia. Countries at the bottom of the table -Italy, Latvia, Greece, Lithuania and Polandhave a long way to go before adult education becomes accessible, available and affordable for most citizens. Romania is a country that definitely is far from realising adult education as a common good. It is important to emphasise that in calculating the index, we have assigned equal weight to all four dimensions distinguished for adult education as a common good. However, depending on the theoretical considerations or the policy purposes, researchers or policymakers may prefer different weights. To facilitate further use and transparency of the IAECG, we present the indexes for each of the dimensions of adult education as a common good in Table 3. Table 3 demonstrates that differences also occur between the country ranks depending on the four dimensions of adult education as a common good. For example, Denmark, which occupies the third place according to its overall score, ranks fifth on accessibility, first on availability, fifteenth on affordability and fourth on social commitment. These differences indicate that adult education as a common good is a complex phenomenon and that, in this respect, problematic aspects appear in all countries, which should be improved upon. The results also show that the four dimensions of the index of adult education as a common good have different differentiating power. Thus, the countries differ considerably in relation to accessibilitythe two highest scores being 84.63 and 83.47, and the two lowest scores being 18.24 and 6.7. The differences between countries in relation to availability of adult education, however, are not so pronouncedthe two highest scores are 73.21 and 68.8, and the two lowest, 41.16 and 23.47.

Results
To deepen our analysis, at a next step, we carried out cluster analysis of the four dimensions of the common good, thereby trying to capture similarities and differences between countries in relation to the way the dimensions interact in each country. The algorithm of the cluster analysis used allowed us to identify six distinctive groups of countries based on their overall score. We have given the characteristics of each cluster in Table 4. More specifically, this table provides an overview of the intersection between the levels of the four dimensions of adult education as a common good as these have emerged from the cluster analysis.
In Table 4, we have arranged these clusters in an order showing the extent to which they represent different levels of realisation of adult education as a common good. We have designated the six distinctive clusters as 1) reality, 2) feasible, 3) possible, 4) ambiguous, 5) problematic and 6) invisible. The reality cluster of adult education is formed mainly by Northern European countries (Denmark, Finland, Sweden) and Luxembourg. In all these countries, adult education is characterised by high levels of participation, a wide range of available opportunities for adult learning and strong commitment of society towards adult education provision. The largest cluster of countries is that in which adult education as a common good is feasible; it includes Austria, Table 2. Index of adult education as a common good (overall score and rank) for 24 countries.

Country
Overall score Cyprus, Estonia, France, Germany, Hungary, The Netherlands, Norway, Spain. These countries have achieved a real progress with regard to making adult education accessible to all and involving different stakeholders in their provision and have shown potential to improve the affordability and availability of adult education. The countries in which adult education as a common good is possible are the Czech Republic, Lithuania, Poland and Slovenia. Although adult education has not yet been realised as an inclusive process to which the society is strongly committed, these countries demonstrate a potential to reach this goal in the future. The next cluster refers to countries in which realisation of adult education as a common good is ambiguous -Belgium, Malta and Slovakia. These countries have both achievements and significant drawbacks in the provision of adult education. The last two clusters include countries that are far from achieving adult education as a common good. The problematic one is formed by two Mediterranean countries, Italy and Greece, and one post-Communist country, Latvia. The invisible cluster comprises only one country, Romania.

Discussion of the results
The conceptualisation and measurement of adult education as a common good provide a new perspective for understanding and assessing how adult education develops in different countries and in different institutional, socio-economic and cultural contexts. Against the trend of increasing privatisation of adult education, viewing adult education as a common good allows us to reimagine it as indispensable for human well-being in contemporary knowledge-based and highly dynamic societies. The developed IAECG captures the complexity of adult education as a phenomenon, the social embeddedness and the multifaced determination of its implementation. The Index also highlights that the importance of adult education for individual and social wellbeing and its complexity require a wider social commitment to adult education provision. Undoubtedly, the understanding of adult education as a common good calls for a humanistic view on adult education as an inclusive process that promotes social solidarity and justice (Boyadjieva & Ilieva-Trichkova, 2017). The suggested IAECG can be used as an instrument for evaluating the feasibility of such a perspective and how it will unfold in the future. The results presented in the paper suggest that, although not without tensions, adult education as a common good is not a utopian chimera. It remains to be seen how challenges stemming from the most recent political, demographic and cultural processes in Europe will affect adult education policies and practices.
The analysis reveals six distinct clusters of countries in Europe with regard to the extent to which adult education is realised as a common good. In a recent study, Eve-Liis Roosmaa and Ellu Saar (2017) theoretically distinguished seven adult-learning country types, which correspond to the typology of countries based on varieties of capitalism, welfare state regimes and their extensions. The types are: liberal, social democratic, conservative/continental, Southern Europe, post-Socialist neoliberal, post-Socialist embedded neoliberal and post-Socialist Balkan. Their study and ours capture different aspects of adult education. Eve-Liis Roosmaa and Ellu Saar focus mainly on its accessibility, captured in terms of the level of participation and inequality in participation, whereas we add indicators on the availability and affordability of adult education and on social commitment to it. Moreover, the overlap of countries between the two studies is not full. These differences explain to a certain extent why both similarities and differences are to be seen between our analysis and the typology proposed by Eve-Liis Roosmaa and Ellu Saar. Thus, for instance, Estonia displays more features of a Northern or a Western European country than its neighbours Latvia and Lithuania. Another example refers to the observed heterogeneity within the Visegrad countries with regard to the realisation of adult education as a common good there. Using sets of demographic and socio-economic variables, Peter Robert (2012) has identified, respectively, four and three clusters of countries, which also seem to differ from the mainstream lifelong learning typologies.
Our results raise the question as to the explanatory power of the typology of welfare regimes in international comparative studies of adult education. This question is in line with existing critiques of the use of typologies in lifelong learning research. On the one hand, some criticisms refer to the limitations of working with country-level variables, given that governance is also located at regional and transnational levels (Riddell & Weedon, 2012). On the other hand, some of these typologies have been constructed without statistical testing or validation (Boeren, 2016). Our typology of countries based on the extent to which adult education has been realised as a common good does not correspond fully to the typology of countries based on their specific variety of capitalism and welfare state regime; this fact suggests that additional dimensions of cross-country differences that are not captured by the second typology (e.g. cultural dimensions) should be taken into account to better explain the development of adult education in different countries. It is also worth discussing whether adult education as a common good would be better understood by combining macro and meso levels of analysis.
Finally, our findings chime with recent challenges to make lifelong learning a reality, which is one of the goals of the Strategic Framework 'Education and Training 2020'. More specifically, our findings highlight that this challenge cannot be reduced simply to boosting the participation rate in adult education, but also requires a much wider vision of how this type of education can become a common good.
We see several directions for future research within two broad perspectives. The first is related to further deepening the conceptualisation of adult education as a common good. We agree with Rita Locatelli (2016, p. 153) that the apparent vagueness of the definition of concepts such as common goods, 'which may be seen as a limit, constitutes in reality their strength', and that 'common goods exemplify what Lévi-Strauss called "significant flottant", the meaning of which, although imprecise, enables a concept to function as a point of attraction of different meanings'. However, to grasp these different meanings, important topics at the theoretical level need to be comprehensively addressed, e.g. public interests and adult education as a common good; the public-private divide in the provision of adult education and adult education as a common good; adult education as a common good and governance in adult education, and so on. It is also worth analysing how adult education itself contributes to achieving other common goods. For instance, Melanie Walker and Monica Mclean (2013) focus on how university-based professional education in South Africa might contribute to the public good of poverty-reduction and thus to achieving more justice and less inequality. To date, the contribution of adult education to the common goods has not been explored.
The second perspective points to different directions for developing a methodology for exploring the dynamics of how adult education unfolds as a common good in specific national contexts over time. The developed IAECG has some limitations, which derive from the quality and availability of data for measuring different dimensions of adult education as a common good. A real challenge is finding reliable indicators for measuring social commitment to the production and governance of adult education. The set of indicators used needs to be enriched both with subjective and objective indicators and with indicators capturing the qualitative aspects of adult education. It is also important to widen the range of countries studied. Special attention should be focused on how different factors (e.g. the country's specific institutional arrangements) influence the extent to which adult education functions as a common good.

Conclusion
The present article contains both theoretical and methodological contributions. At the theoretical level, it outlines a theoretical framework for conceptualising adult education as a common good. Evident signs of dismantling of the welfare state model and the accelerated processes of privatisation in the sphere of education put to the fore the need to rethink the nature of all forms of education. A common-good perspective provides grounds for a humanistic approach to adult education, centred around the issues of accessibility and inclusion, and promotes the values of solidarity and justice in the educational sphere and its governance. More specifically, our analysis demonstrates that the extent to which adult education as a common good is accomplished in a given society/country reflects the accessibility, availability, affordability of and social commitment to adult education. At the methodological level, the article involves the designing of an index for measuring adult education as a common good, an index allowing countries to be ranked according to the extent to which adult education as a common good has been realised in them.
The article also has clear political implications as it provides theoretical conceptualisations and develops a methodological instrumentan index of adult education as a common goodfor assessing the effectiveness of national polices in the sphere of adult education across Europe. The new global Education 2030 Agenda has adopted a perspective based on lifelong learning. Adult education could substantially contribute to the Education 2030 Agenda (Milana, Holford, Hodge, Waller, & Webb, 2017) only if it is understood, governed and practiced, from a humanistic perspective that takes into account both individual and public interests, and that calls for the pursuit of common goods. Notes 1. In the discussion on common goods, we refer to contemporary societies, which, to various degrees, adhere to the principles of democracy and market economy. 2. For a further discussion on the similarities and differences between public good and common good, see Deneulin and Townsend (2007), p. 32. 3. See, for example, Simon Marginson's discussion (2016) on the public goods that higher education produces and how it can contribute to the common good, and Ellen Hazelkorn and Andrew Gibson's discussion (2017) on how higher education serves the public good.