TRENDS IN THE USE OF NON-FORMAL ADULT EDUCATION SUPPLY

Although non-formal adult education (NFE) is one of the most widespread forms of lifelong learning today, we still have not enough knowledge about the structure of its supply side and the factors that influence the use of different types of NFE. The paper, therefore, deals with the transformation of the use of NFE supply for adults in the Czech Republic in 2011-2016. The aim of the paper is to (1) show the development dynamics of the use of NFE offer, both in terms of job-oriented and non-job-oriented education and training. Following this, we intend to (2) determine the structure of adults who use different types of NFE. This will allow us to map the groups of adults who benefit most from (a) courses, (b) workshops, (c) private tuition, and (d) guided on-the-job training. We use a secondary analysis of the data from the Adult Education Survey from 2011 (N=3,263) and 2016 (N=4,840) and compare them to meet these goals. The results of both surveys are processed using descriptive and cluster analysis. Based on the data analysis, we demonstrate a significant increase in guided on-the-job training as the primary type of NFE used by adults across all socio-demographic groups. This result suggests a significant neoliberealization and instrumentalization of the NFE offer, which displaces non-job-oriented education and longer-term and more formalized models of education and individualized adult learning.


Introduction
The period after 2010 was one of the most dynamic in terms of the development of non-formal education of adults (NFE). Specifically, it has intensified processes such as globalisation, deindustrialisation, and the subsequent flexibilisation and precarization of work, which resulted in an increased demand for NFE (e.g., Desjardins, 2020a;Desjardins & Ioannidou, 2020;. At the same time, the NFE supply itself began to change more profoundly. It has seen new technological innovations and types of organisation of learning activities, which, on the one hand, increase its flexibility and accessibility, but on the other hand, lead to higher demands on the digital skills of learners (Hämäläinen et al., 2015;Merriam & Baumgartner, 2020;Nygren et al., 2019).
Based on that, the study deals with the trends in the development of use NFE supply after 2010. First, it focuses on determining the critical trends in NFE according to the types of organization of learning activities (e.g., courses, private lessons, guided on-the-job training). Second, it focuses on the change of participants in these activities, determining the factors influencing the involvement of adults in different types of NFE.

NFE
We define NFE following the existing international documents (CEDEFOP, 2008, Eurostat, 2016a, 2016b; UNESCO, 2020) as learning activities outside the formal educational system. It involves structured activities that usually do not result in official certification, according to ISCED (2011). This type of learning includes all organized and planned development activities, such as courses, workshops, and private lessons or training in the workplace. NFE also includes both job-oriented (vocational) as well as non-job-oriented learning.
As to the NFE offer, its overall scope is crucial, as it is the most widespread form of organized lifelong learning for adults, with 30 to 50 % of adults in many Central and Western European countries (Desjardins, 2017(Desjardins, , 2020aDesjardins et al., 2016;Rubenson, 2018). Its importance lies in the fact that it is an important tool for upskilling and reskilling in the field of work, where it complements or replaces vocational education and training and the fact that it has several noneconomic functions. In this regard, it increases adults' civic engagement (Grotlüschen et al., 2019) and improves their quality of life and health (Iňiguez-Berrozpe et al., 2020). It also contributes to community development (Kalenda & Smékalová, 2015) and active population ageing (Desjardins, 2020b).

Types of NFE
As regards its organisation, according to ISCED (2011) and the Classification of learning activities of the European Commission (Eurostat, 2016a(Eurostat, , 2016b, NFE is divided into: (1) Courses that are typically subject-oriented and are taught by one or more people specialised in a specific field. They may occur in one or more settings/environments, including the online/elearning environment. They include learning organised for a group of adults in a classroom or on a digital platform. It is built around a teacher/tutor/instructor passing on knowledge to provide instruction and educate. https://doi.org/10.15405/epiceepsy.21101.3 Corresponding Author: Jan Kalenda Selection and peer-review under

43
(2) Workshops (combined theoretical-practical courses) cover all educational activities that combine classroom-based instruction (theoretical) with practice in real or simulated situations.
It includes vocational apprenticeships, except those covered by formal educational institutions.
Combined theoretical-practical courses also include learning activities that mix classroom instruction with guided-on-the-job training and practice.
(3) Private tuition (private lessons). This type of NFE represents a planned series of learning experiences offered by experts or others who act as experts, selected to deepen knowledge or skills, to learn more intensively, usually undertaken by only one or very few learners. Typically, the tutor is also the education provider.
(4) Guided-on-the-job training. This organized learning activity is characterised by planned periods of training, instruction or practical experience, using normal tools of work, either at the immediate place of work or in a work situation. On-the-job training is usually organised by the employer to make it easier for workers. It may include general training about the company as well as specific job-related instructions.

Trends in NFE
The development of NFE in the last decade is characterized by three trends that may affect the supply of individual learning activities: (1) The development of flexible and agile forms of learning such as e-learning, micro-learning, etc., which lead to a lower level of long-term planning, structuring and organization of NFE, transferring emphasis to learning within the social environment (community of inquiry/learning) and a higher degree of individualization (Kearns, 2015;Messmann et al., 2018;Merriam & Baumgartner, 2020). As a result, we may expect a decrease in the demand for traditional forms of learning activities, such as courses and workshops, which are not suitable tools for agile learning and flexible skills development (Lancaster, 2020), both for the job-oriented and non-joboriented forms of NFE.
(2) An increase in job-oriented training. According to a number of researchers (Desjardins et al., 2016;Rubenson, 2018), in the case of NFE, we may observe both an increase in the share of education and training focused on work skills and an increase in the proportion of learning activities paid for by employers. Due to the NFE's inclination towards the world of work, it can therefore be expected that the share of guided on-the-job training, which will form the central part of all NFE training activities, will increase. In other words, there will be a homogenization of NFE in terms of this learning activity.
(3) An overall increase in NFE participation in most countries (Desjardins, 2017(Desjardins, , 2020aRubenson, 2018) should reduce inequalities in participation in different types of NFE among different social groups. As a result, traditional inequalities such as gender, age, and education should not play a significant role in engaging in different types of NFE. These are forms of private tuition which may have the highest costs in terms of learners and which (economic) inequalities may intervene most strongly.

Problem Statement
Although NFE is one of the most widespread forms of lifelong learning, which underwent dramatic changes in the last decade, we still have only basic knowledge of the changing offer of its various types and factors that influence adult involvement in them. If NFE is researched, the participation of individuals in different educational activities is usually not considered (e.g., Boeren, 2016Boeren, , 2017Lee, 2017;Lee & Desjardins, 2019;Merriam & Baumgartner, 2020;Rubenson, 2018). In representative surveys from the Czech Republic, participation in various NFE types has not yet been addressed (e.g. Kalenda 2015;Vaculíková et al., 2020).

Research Questions
Regarding the identified research problems, the study focuses on answering three interconnected research questions: (1) RQ1: What was the development of the use of various types of NFE in CZ after 2010?
(2) RQ2: How did the development of the use of different types of NFE differ between joboriented and non-job-oriented training?
(3) RQ3: What kind of groups of participants can be identified? What are their key characteristics and how did they change between 2011 and 2016?

Purpose of the Study
Following these questions, the study aims to map out: (1)

Research Methods
We use a secondary analysis of the data from the Adult Education Survey (AES) from 2011 (N=10,253) and 2016 (N=12,468) and compare them to meet these goals. Only NFE participants were included in the analysis, who also stated whether their participation was primarily job-related or personalrelated, namely, n = 3263 (AES 2011) and n = 4840 (AES 2016). The results of both surveys are processed using descriptive and cluster analysis (K-means cluster) focused on the change of crucial variables, i.e., the forms of NFE offered and microsocial factors, including motivation for education. As part of the analysis, weights were applied to ensure the representativeness of both samples for the Czech Republic.

Findings
The results of the development of participation in individual types of NFE in 2011 and 2016 are summarized in Table 1. The first part of the results is devoted to the research questions RQ1 and RQ2. there was less use of courses (-18 p.p.), and we see a strong inclination to use guided-on-the-job training (a total of 52%, +23 p.p.). The increase is only visible among people who declare its job-oriented purpose. In the case of workshops, we see a slight decrease (-6 p.p.). In the case of private tuition, the situation does not change.
For answering the research question RQ3, cluster analysis (K-means cluster) was applied based on preferences for one of the four types of NFE supply. As part of the cluster analysis, we generated solutions for two to five clusters for both years. We assessed all solutions in detail and finally decided on three clusters in both waves of AES. With a higher number of clusters, the solution broke down into small subgroups making up less than 10% of the research population. It was essential for us to identify the basic types. The results are presented in  Cluster analysis was followed by basic descriptive characteristics of the final cluster solutions for 2011 and 2016, also included in Table 3. Cluster I increased the most in 2016, but its socio-demographic composition is almost identical to 2011. In contrast to that, Cluster II got smaller. At the same time, the share of persons participating in the NFE for job-related purposes (-18 p.p.) and persons working full time

Conclusion
The first of the two main aims of this paper was to analyse the development of different types of NFE between 2011 and 2016. Importantly, we found that the decline of participation in traditional types of organised learning -i.e., courses is typical for this phenomenon. The number of participants in them shattered to almost half during the investigated period. This empirical finding is in line with the current literature (e.g., Kearns, 2015;Lancaster, 2020;Messmann et al., 2018;Merriam & Baumgartner, 2020), emphasising a higher demand for more flexible, experience-based and agile forms of organised learning.
The same tendencies, but to a lesser degree, are also common for workshops, while private tuition did not change.
Another key identified trend is a rapid increase in the involvement of adults in organised on-the-job training, which almost doubles its size. According to some scholars (Desjardins, 2017;Desjardins & Ioannidou, 2020;Rubenson, 2018), this result is compatible with an increased supply of job-related training characterised the previous decade. In the case of participation in NFE, it results not only in higher participation for work-related reasons and employer-supported training but also in higher utilisation of guided on-the-job training as a primary organisational type of NFE. On the one hand, it leads to the homogenisation of different forms of NFE towards one type. On the other hand, it makes NFE more economically and labour market centred than before. In other words, the results confirm neoliberal tendencies inside adult education (Milana et al., 2016) and its instrumentalisation, where learning practice increasingly depends on the private sector.
During the following second aim, analysis and depiction of groups of participants in different types of NFE, we identified three crucial social groups: (1) Cluster I. (mostly on-the-job training participants). These are adults that dominantly learn for job-related reasons and are highly active in the labour market. Common features of the persons in this cluster are a higher proportion of males over females, adults between age 30 to 50 years, and the high occurrence of individuals with secondary education (ISCED 3) that form two-thirds of this group.
(2) Cluster II. (mostly workshop participants). The second cluster has a lower orientation to training dedicated to work-related skills. Furthermore, it has a higher proportion of older adults, many times retired persons. In addition, the tertiary level of education (ISCED 5-6) is typical for almost half of the members of this group. Finally, females are more frequent participants of this type of NFE than males.
(3) Cluster III. (mostly course participants). The participants of the courses gained their knowledge and skills for non-job-related reasons with higher occurrence than in previous types of NFE.
Based on that, courses are the primary type of non-work-related education in our sample.
Compared to other clusters, this one has a higher proportion of adults with just primary or low secondary education (ISCED 0-2 and ISCED 3c). Almost one-third of its members are low- Interestingly, for this typology, it is essential that it has not changed too much since 2011 in its internal structure (i.e., the composition of the adults forming different participant groups) but in the size of the particular clusters. The cluster of participants learning through guided on-the-job training almost doubled its size, while the other two dwindled to half.
These findings confirm the homogenisation tendencies related to the development of different types of NFE elaborated above. The pattern found here corresponds to increased employer-sponsored and joboriented education (Desjardins, 2020a;Rubenson, 2018).
Our results also support thesis about existence of deep inequality inside participation in NFE (Boeren, 2016(Boeren, , 2017Boyadjieva & Ilieva-Trichkova, 2017, 2018Desjardins et al., 2006;Lee, 2017;Lee & Desjardins, 2019). Notably, the results reveal that inequalities exist not just in overall participation or number of training hours but also according to involvement in different types of NFE.
Although high-educated adults, actively involved in the labour market with full-time job contracts, are more frequent participants of workshops, low-educated people outside the labour market and with timerestricted job contracts participate more in courses. Similar to Busemeyer´s (2015) work, the present study shows that adults at higher social risk also participate less in work-related NFE, mainly guided on-the-job training. This situation is not just typical for low-educated adults, unemployed, or elderly but also for females. In this regard, the same conclusions were drawn by several scholars (Dämmrich et al., 2015;Vaculíková et al., 2020), who reported higher support to males from employers.

Study Limitations
These results need to be interpreted with caution because they are based on the data from only one country -the Czech Republic. Its adult education system is characterised by a high level of skill mismatch (Hamplová & Simonová, 2014), an orientation towards job-oriented learning  as well as high inequality in NFE according to integration into the labour market and educational attainment . For these reasons, our findings might not represent trends in the use of different types of NFE and the structure of their participants. As a consequence, this study failed to account for the international comparative dimension. The second downside of the used methodology lies in the secondary nature of our data sources (AES 2011 and2016), enabling us to analyse only data from the first half of the previous decade. Due to this limitation, questions about the most recent trends affecting the world pandemic situation remain unanswered.
Even though this article provides a springboard for understanding trends in the use of different types of NFE among adults, more studies are needed to be done in this area. They should consider not only additional features of organized learning (e.g., online/offline organization platforms, involvement of tutors, or overall quality or length of learning), but also international comparative dimension, both in a spatial and temporal sense.