Understanding and Measuring Child Well-being in the Region of Attica, Greece: Round Five

Abstract

This research paper aims to present the results of the implementation of the C.W.-SMILE tool that recorded child well-being in the second semester of the 2020–2021 school year. This is the fifth round of this ongoing diachronic research. The C.W.-SMILE tool consists of six dimensions: home conditions (D.1), nutrition (D.2), unemployment of guardians (D.3), free healthcare (D.4), moral education (D.5), and leisure (D.6). The first three dimensions concern children’s economic well-being, while the latter three dimensions determine children’s non-economic well-being. Based on welfare economics, the combination of economic well-being and non-economic well-being constitutes children’s general (social) well-being. Each dimension consists of Simple Indicators. The paper also presents the results of the school year as a whole, to help investigate the evolving impact of the COVID-19 pandemic on children’s well-being for more periods. The tool was used to measure the well-being of children in Attica, through questionnaires that were circulated in 22 public schools and three support centers of the organization ‘The Smile of the Child’ (25 units in total). The sample consisted of 509 children, belonging to three distinct school categories. The results of the second semester are mapped in seven clusters (relating to seven socio-economically homogenous groups of municipalities in Attica). Analysis of the results of the school year 2020–2021 was done based on the data collected from a sample of 1,623 children; in other words, it took into account the data relating to the entire samples that were surveyed in the first and second semesters. The central outcomes of all the previous rounds of the research are verified through a principal component analysis (PCA), and a multiple correspondence analysis (MCA) which are employed for the fifth round. Finally, the paper suggests an action plan of social welfare based on a ChoiCo game designed for the needs of the fifth round of the C.W.-SMILE research.

1 Introduction

Nowadays, the global impact of COVID-19 is negative in many aspects of the economy; this pertains not just to tourism (Rodousakis & Soklis, 2021, 2022) but also to the society as a whole (Erden et al., 2021; Giovanis & Ozdamar, 2022; Gozgor, 2021; Ikeda et al., 2021; Midões & Seré, 2021; Morrison et al., 2021; Shek, 2021). However, the most alarming effect of the global crisis is on child well-being (Dirwan et al., 2021; OECD, 2020, 2021; UNICEF, 2020, 2021; United Nations, 2020). Taking into account that children’s quality of life is very important for a successful life as adults, for example, in terms of academic accomplishment (Kaya & Erdem, 2021), the primary target of societies in the light of the pandemic should be the achievement of child well-being (Guillén-Fernández & Vargas-Chanes, 2021; Leriou et al., 2021, pp. 3–7).

Many tools have been developed for the measurement of child well-being all over the world. The Middle Years Development Instrument (MDI) of Canada is the most important tool on a global level (Guhn et al., 2012; Schonert-Reichl et al., 2011, 2013). Additionally, other tools to measure the subjective (Michalos & Land, 2018; Michalos & Hatch, 2020; Sirgy, 2021, pp.132–133; Voukelatou et al., 2021) well-being of children have been developed, with a strong focus on life satisfaction (Aymerich et al., 2021; Borualogo & Casas, 2021; Caqueo-Urízar et al., 2022; Casas et al., 2013; Casas & González-Carrasco, 2021; Dinisman et al., 2012; Gierczyk et al., 2022; Goldan et al., 2022; Gross-Manos & Bradshaw, 2022; Herd, 2022; Huebner & Furlong, 2016; Jiang et al., 2021; Leto et al., 2021; Llosada-Gistau et al., 2015; Montserrat et al., 2015a, b; Moreira et al., 2021, 2022; Rodriguez de la Vega, 2014; Savahl et al., 2017, 2021; Schutz et al., 2022; Tonon & Mikkelsen, 2022; Viñas et al., 2019). Furthermore, there are various studies concerning the opposite of child well-being, such as, according to Leriou et al., (2021, Sect. 2) and Leriou (2022, Sect. 2, p.1,970) child poverty (Landiyanto, 2022; Pinilla-Roncancio et al., 2021).

Studies on the negative impact of the COVID-19 crisis on the well-being of children have been conducted all over the world (Branquinho et al., 2022; Bravo-Sanzana et al., 2022; Chzhen et al., 2022; Gao et al., 2022; Gierczyk et al., 2022). However, the only diachronic study that researched the same in the region of Attica, Greece, have been conducted in the third round of an on-going research by Leriou et al. (2022b) as well as in the fourth round of the same aforesaid research by Leriou (2022). According to Leriou et al. (2022b), “An analysis on the level of simple indicators and dimensions shows a deterioration in the economic well-being of students in the second semester of 2019–2020 as compared to the 1st, mainly due to a sharp increase by 5 percentage points in the rate of unemployed parents/guardians.” In addition, Leriou et al. (2022b) have highlighted that “The impact of the COVID-19 pandemic on child well-being is especially grave in its economic aspect, due to the significant deterioration in D.3, a dimension evidently influencing the other dimensions of economic well-being of children (i.e., D.1 and D.2) negatively.” More specifically, in this study, the most crucial finding about the deterioration of child well-being in the region of Attica, Greece, because of COVID-19, is the extremely and alarmingly sharp increase in the percentage of students living in jobless households. Furthermore, Leriou (2022, pp. 1,980–1,983), encourages the decision-makers of public policies in Greece to focus mainly on Public Finances and, more specifically, on policies based on indirect taxation, merit goods according to Musgrave, etc., in order to mitigate the negative effects of the pandemic on child well-being.

The main aim of this research paper is to present and discuss the results regarding child well-being in Attica, Greece, in the fifth round of the aforesaid, on-going, diachronic research related to the second semester of the school year 2020–2021 and, furthermore, of this specific school year as a whole. Its secondary aim is to investigate the evolving impact of the COVID-19 health crisis on children’s well-being, specifically in Attica, Greece for more periods and propose an action plan with good policies (Michalos, 2017) based on a ChoiCo game designed for the needs of the fifth round (Fig. 1).

Fig. 1

A message from the Game (http://etl.ppp.uoa.gr/choico/?cwsmile)

2 Methodology

2.1 Instruments and Target Population

 The tool used to measure child well-being (Leriou et al., 2021) is structured on indicators (Leriou, 2016, 2019, 2022; Leriou et al., 2021, 2022a, b Tasopoulos & Leriou, 2014) and founded on the same theoretical framework as the first round of the research (Leriou, 2016, 2022; Leriou et al., 2021), which is based on welfare economics (Leriou, 2015, 2016; Leriou & Tasopoulos, 2015–2016; Leriou et al., 2021). It was implemented among children belonging to three school categories—the 6th grade of elementary school (10–11-years-old), the 3rd grade of junior high (13–14-years-old), and the 3rd grade of high school (16–17-years-old)—since children of these levels have a broader perception. The tool includes six dimensions (synthetic indicators): the first three (D.1, D.2, D.3) are related to economic well-being, and the other three (D.4, D.5, D.6) to non-economic well-being (Leriou, 2022, Sect. 2, p. 1,970; Leriou et al., 2021, Sect. 2 and 3). On the basis of welfare economics, the sum of economic and non-economic child well-being constitutes the general (overall / social) child well-being (Leriou, 2022, Sect. 2, p.1,970, Figs. one and Fig. two). Dimension 1 is the home conditions (D.1); Dimension 2 is nutrition (D.2); Dimension 3 is the unemployment of the guardians (employment status of household—D.3); Dimension 4 is free healthcare (D.4); Dimension 5 is moral education (D.5); and Dimension 6 is leisure (D.6) (Leriou, 2022, Appendix, Fig. three). Every dimension includes specific Simple Indicators, each of which corresponds to specific questions in the questionnaires (Leriou et al., 2021) which were used to collect the data.

Simple Indicators and their corresponding questions in the survey questionnaire are identical to those in the first round of the research, regarding the school year 2018–2019 (Leriou et al., 2021). The responses to the questionnaires were rated using a similar approach adopted in the first round of the research (Leriou et al., 2021).

More specifically, the whole methodological framework (indicators, questionnaires, ratings, subgroups, sampling, etc.) and the statistical analysis are the same in all rounds of the research (Leriou et al., 2021). Therefore, in light of the above, all the tables and figures from each published paper for all rounds of the research (Leriou, 2022; Leriou et al., 2021, 2022a, b), actually constitute a long list of datasets. This list includes diachronically comparable data regarding child well-being in the region of Attica (Greece), concerning economic and non-economic aspects of the quality of life of the children.

2.2 Ethics and Funding

The questionnaires, along with the appropriate permission letters (Gross-Manos et al., 2021) addressed to the parents, were examined by specialists and approved by the Ethics and Conduct Committee of Panteion University. They were also approved by the Ministry of Education, which issued two licenses for carrying out the anonymous survey in schools: one for primary institutions and one for secondary institutions. The students were informed that filling in the questionnaire was not compulsory but voluntary and that they could withdraw from the process at any given moment of their choosing—prior to, during, or even after answering. The research received funding from the Hellenic Foundation for Research and Innovation (HFRI) and the General Secretariat for Research and Technology (GSRT) under grant agreement no. 1926.

2.3 Sample

A multi-stage sampling design was adopted to form the students’ sample similar to the first round of the research (Leriou et al., 2021).

In the fifth round, participants were 509 students from 22 public schools and three Support Centers (SCs) of the organization The Smile of the Child (Appendix, Table 1); more specifically, 148 were in the 6th grade of elementary school, 148 in the 3rd grade of junior high school, and 213 were in the 3rd grade of high school. In total, 235 boys and 260 girls answered the questionnaires, while 14 students did not state their gender.

Based on students’ responses to the question regarding the employment status of their guardians (Leriou et al., 2021, Appendix 1, Questionaries in Tables five and eight), it was concluded that 184 children belonged to a two-parent family, 208 children belonged to a single-parent family with the mother as the leader, while 90 children were in a single-parent family with the father as the leader, and 27 children did not provide any information for this question.

To identify groups of municipalities that are relatively socio-economically homogeneous, a hierarchical cluster analysis (HCA) was performed on a set of six variables with municipality-level aggregate data obtained from the 2011 census (Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021).

3 Results

The results of the second semester of the school year 2020–2021 are presented first, followed by the results of the whole school year 2020–2021. Subsequently, the results of the first semester (Leriou, 2022) are compared with those of the second in order to draw conclusions on the evolving impact of the COVID-19 crisis on children’s well-being.

3.1 Geographical Distribution of Children’s Well-being (Second Semester 2020–2021)

On the basis of Synthetic Indicators 1,3,4, and 5 ( ${\mathrm{CWG}}_{i}$, ${\mathrm{CWE}}_{i},{\mathrm{CW{\rm N}}\mathrm{E}}_{i}$, ${\mathrm{CWG}}_{\mathrm{i}}$) (Leriou et al., 2021, Sect. 3.5, pp. 12–13), the percentages of children below and above the threshold of being at risk of lacking general, economic, and non-economic well-being, by cluster, school level, and gender, are estimated for the second semester of the school year 2020–2021. In other words, the geographical distribution of children’s well-being in the region of Attica is monitored. Furthermore, the results of the Pearson’s chi-squared test are included.

According to Table 2¹ (Appendix), the percentages of children below the threshold in economic, non-economic, and general well-being (Appendix, Fig. 2,²3,³ and 4⁴) by municipalities cluster of Attica (Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021), prove statistically significant differences in economic well-being (χ²_{[6, N=509]} = 18.292, p = 0.006), and in general well-being (χ²_{[6, N=509]} = 17.508, p = 0.008). With regards to the second semester of the school year 2020–2021 and all the clusters (Appendix, Table 2), the percentage of the total number of children below the threshold in general well-being amounts to 9.0%, while 14.5% are classified below the threshold for economic well-being. In particular, as far as general well-being is concerned, the highest percentages of the total number of children below the threshold are found in Clusters 4 (14.6%), 3 (13.5%), and 2 (6.3%) while Cluster 7 (0.0%) presents the lowest percentage in increasing order (Appendix, Fig. 2). On the examination of the parameter of the economic well-being (Appendix, Fig. 3), the percentages below the threshold appear to be higher in Clusters 4 (22.6%), 3 (18.3%), and 1 (10.5%), respectively, while Clusters 7 (0.0%), and 6 (5.6%) depict the lowest percentages (Appendix, Table 2).

Furthermore, taking girls into account, their economic well-being (χ²_{[6, N=260]} = 13.348, p = 0.038), is associated with the cluster (Appendix, Table 2). A notable higher representative number of girls is below the economic well-being threshold in Cluster 4 (27.3%), as compared to other municipalities clusters. On the other hand, general well-being of girls falls into the marginally significant level (p > 0.05 & p < 0.15) (Appendix, Table 2). A higher share of girls is below the general well-being threshold in Cluster 4 (18.2%), as compared to other municipalities clusters (χ²_{[6, N=260]} = 12.454, p = 0.053) (Appendix, Table 2).

Moreover, the results of a Pearson’s chi-squared test within school categories portray that economic (χ²_{[2, N=509]} = 17.163, p < 0.001) and general (χ²_{[2, N=509]} = 11.344, p = 0.003) well-being are associated with the school category of the children. More specifically, a comparison of all school categories demonstrates that the highest percentage of children below the threshold of being at risk of lacking general child well-being in Attica is apparent in the elementary school category (14.2%), followed by junior high school (10.8%), while the high school level comes last (4.2%) (Appendix, Table 2). Therefore, children in the elementary school category face the worst situation with regard to the quality of their life in the second semester of the school year 2020–2021. Similarly, elementary school children represent the highest percentage of children (24.3%) below the threshold of being at risk in economic child well-being in Attica (Appendix, Table 2).

In addition, where the family structure is concerned, the highest percentage of children (21.1%) below the threshold of being at risk in economic well-being (χ²_{[2, N=482]} = 14.200, p = 0.001) is found in the single-parent family with father category (Appendix¸ Table 3⁵).

3.2 Children at Risk of Lacking Well-being, per Simple Indicator and Dimension (Second Semester 2020–2021)

As shown in Table 5 ⁶ (Appendix), the dimension of home conditions (D.1), of economic child well-being is associated with Attica’s municipalities clusters (χ²_{[6, N=505]} = 21.508, p = 0.001). With regard to D.1, the highest percentage of children below the threshold is found in Cluster 4 (17.8%). It is notable that each of the Simple Indicators of home conditions, except Simple Indicator 2, is associated with Attica’s municipalities clusters. In particular, the highest percentage of children below the threshold in all clusters (27.4%) is found in Simple Indicator 4, “Humid or cold house” (χ²_{[6, N=442]} = 23.903, p = 0.001). Furthermore, in terms of Simple Indicator 11, “Housing”, the highest percentage (25.7%) of children below the threshold is found in Cluster 3 (χ²_{[6, N=455]} = 17.840, p = 0.007), (Appendix, Table 5). In addition, regarding Simple Indicator 1, “Supply of electricity”, the marginal highest percentage (13.3%) of children below the threshold is found in Cluster 4 (χ²_{[6, N=479]} = 11.286, p = 0.080). Similarly, students in Cluster 4 present a higher percentage (10.0%) below the threshold of Simple Indicator 10, “Existing internet connection”, as compared to other clusters (χ²_{[6, N=481]} = 14.178, p = 0.028) (Appendix, Table 5).

With regards to Simple Indicator 5, “Three meals a day”, of the Dimension of nutrition (D.2), the highest percentage (26.4%) of children below the threshold is found in Cluster 4 (χ²_{[6, N=448]} = 16.313, p = 0.012) (Appendix, Table 5). Similarly, children in Cluster 4 present a marginal higher percentage (16.7%) below the threshold of Simple Indicator 6, “At least one meal weekly with meat or chicken or fish or pulses/vegetables of equal nutritional value”, as compared to other clusters (χ²_{[6, N=444]} = 12.042, p = 0.061) (Appendix, Table 5). In addition, regarding the family structure, the highest percentage of children (14.1%) below the threshold of Simple Indicator 6 (χ²_{[2, N=424]} = 7.247, p = 0.027) is found in the single-parent family with mother category (Appendix, Table 6.⁷)

Furthermore, the dimension of employment status of household (D.3), of economic child well-being, is marginally associated with Attica’s municipalities clusters (χ²_{[6, N=482]} = 11.403, p = 0.077). In particular, as far as D.3 is concerned, the percentages below the threshold appear to be higher in this period in Clusters 4 (23.7%), 3 (19.3%), 2 (18.8%), and 5 (15.1%), while 6 (0.0%) presents the lowest percentage (Appendix, Table 5). In addition, regarding the family structure, the highest percentage of children (32.2%) below the threshold of D.3 (χ²_{[2, N=482]} = 23.750, p < 0.001) is found in the single-parent family with father category (Appendix, Table 7⁸). Taking into account the school category, the highest percentage of children (27.5%), below the threshold of D.3 (χ²_{[2, N=482]} = 14.940, p = 0.001) is found in the elementary school category (Appendix, Table 8⁹).

The dimension of education (D.5), of non-economic child well-being, is associated with Attica’s clusters (χ²_{[6, N=509]} = 13.052, p = 0.042) (Appendix, Table 5). In terms of D.5, the highest percentages of children below the threshold are found in Clusters 6 (44.4%) and 7 (44.4%), while the lowest percentage is presented in Cluster 2 (9.4%) (Appendix, Table 5). In particular, the highest percentage (56.4%) of children below the threshold in all clusters is found in Simple Indicator 16, “Moral values/virtues” (χ²_{[6, N=486]} = 22.936, p = 0.001; Appendix, Table 5). More specifically, with regards to this Simple Indicator, the highest percentages of children below the threshold are found in Clusters 2 (80.6%), 4 (61,1%) and 7 (60.0%). In Simple Indicator 12, “Empathy for all people” (χ²_{[6, N=330]} = 11.390, p = 0.077), the percentage of children below the threshold in all clusters amounts to 47.3%, while the marginal highest percentages of children below the threshold are found in Clusters 7 (85.7%) and 6 (60.0%). Regarding Simple Indicator 14, “Love of animals,” Clusters 1 (59.5%) and 6 (52.9%) present higher percentages below the threshold, as compared to other clusters (χ²_{[6, N=491]} = 19.346, p = 0.004). In addition, as far as Simple Indicator 15, “Love of nature”, is concerned, the highest percentage (47.1%) of children below the threshold is found in Cluster 6 (χ²_{[6, N=491]} = 14.861, p = 0.021). In Simple Indicator 13, “Love of people,” the highest percentage (47.1%) of children below the threshold is found in Cluster 6 (χ²_{[6, N=487]} = 15.123, p = 0.019) (Appendix, Table 5). With regards to school category, the highest percentage (39.2%) of children below the threshold of D.5 is found in the junior high school category, (χ²_{[2, N=509]} = 38.066, p < 0.001) (Appendix, Table 9¹⁰).

Moreover, D.6, of non-economic child well-being, is associated with Attica’s clusters (χ²_{[6, N=509]} = 23.590, p = 0.001) (Appendix, Table 5). With regards to D.6, the highest percentage (43.0%) of children below the threshold is found in Cluster 1 (Appendix, Table 5). In terms of Simple Indicator 19, “Playing outdoors,” Cluster 1 presents a higher percentage (50.6%) below the threshold (χ²_{[6, N=501]} = 21.503, p = 0.001). Similarly, students in Cluster 1 present a higher percentage (60.2%) below the threshold of Simple Indicator 20, “Vacation”, as compared to other clusters (χ²_{[6, N=497]} = 38.360, p < 0.001). In terms of Simple Indicator 18, “Spare time”, Cluster 4 presents a higher percentage (34.1%) below the threshold (χ²_{[6, N=501]} = 17.149, p = 0.009) (Appendix, Table 5).

3.3 Multiple Correspondence Analysis (second semester 2020–2021)

In order to further explore the relationship between the below/above threshold groups of students per Simple Indicator, a multiple correspondence analysis (MCA) was performed (Appendix, Table 10, Fig. 5¹¹ and 6). The factorial model that was adopted has four main axes that explain 46.6% of the total inertia (the 1^st axis explains 19.6% of the inertia, eigenvalue λ₁ = 0.176; the second, 14.5%, λ₂ = 0.131; the 3^rd, 6.5%, λ₃ = 0.059; and the 4^th, 6.0%, λ₄ = 0.054). All the categories used in the analysis (active and supplementary) are statistically significant according to the calculation of their v. test.

As shown in Table 10 (Appendix), the set of categories most significantly contributing to the definition of Axis 1 includes all simple indicators of D.2 that are pinpointed below the threshold as well as, to a lesser extent, some simple indicators of D.1 (internet, electricity, and cold home) that are located below the threshold, and D.3 that is pinpointed below the threshold. Overall, it appears that Axis 1 predominantly reflects the economic well-being of children. The set of categories that most contributes to the definition of Axis 2 comprises some simple indicators of Dimension 5 (love of people, love of nature, love of animals, empathy for all people). This finding indicates that Axis 2 predominantly represents D.5 of the non-economic aspect of well-being (Appendix, Table 10). Overall, Axis 1 and Axis 2 essentially confirm the structure of the indicators concerning well-being (economic and non-economic), as suggested in this study. Moreover, the aforesaid was confirmed in rounds two (Leriou et al., 2022a) and four (Leriou, 2022) of the research. Hence, the correctness of the theoretical and methodological framework of this study can be confirmed further.

In addition, Axes 3 and 4 reflect underlying patterns of well-being that are more complex (e.g., a combination of non-economic and economic well-being of children) (Appendix, Table 10). The aforesaid was confirmed in round four (Leriou, 2022) of the research. Hence, the correctness of the theoretical and methodological framework of this study can be confirmed again. Specifically, the set of categories contributing the most to the definition of Axis 3 includes one Simple Indicator of D.5 (moral values), the Simple Indicator regarding spare time of D.6 that is pinpointed below the threshold, and one Simple Indicator of D.1 (internet) that is pinpointed below the threshold.

Moreover, the set of categories contributing the most to the definition of Axis 4 includes two simple indicators of D.1 (heating and cold home) that are pinpointed below the threshold, the simple indicator regarding vacation of D.6, and Simple Indicator 16 (moral values) of D.5.

Furthermore, as shown in Figs. 5 and 6 (Appendix), the simple indicators with the largest contribution to child well-being belong to Dimensions 5 (moral education) and 2 (nutrition). The aforesaid has also been confirmed in the previous rounds of the research (Leriou, 2022; Leriou et al., 2022a). Hence, this further confirms the correctness of the theoretical and methodological framework of this study.

3.4 Results of Means (Second Semester 2020–2021)

The processing of data based on Composite Indicators 7– 12 (Leriou, 2022; Leriou et al., 2021, 2022a, b) reveals that, with regard to the dimensions and all clusters, the most serious problem appears to be one belonging to the non-economic aspect of child well-being, which is D.5 (Appendix, Table 11¹²). The dimension that follows is D.3 of economic child well-being. (Appendix, Table 11).

The results of the one-way ANOVA (Appendix, Table 11) indicate statistically significant differences F(6, 474) = 2.399, p < 0.05, in the means of the clusters for Simple Indicator 10, F(6, 435) = 4.145, p < 0.001 for Simple Indicator 4, F(6, 481) = 2.637, p < 0.05 for Simple Indicator 2, F(6, 448) = 3.047, p < 0.05 for Simple Indicator 11, F(6, 441) = 2.778, p < 0.05 for Simple Indicator 5, F(6, 480) = 4.177, p < 0.001 for Simple Indicator 13, F(6, 484) = 4.343, p < 0.001 for Simple Indicator 14, F(6, 484) = 5.069, p < 0.001 for Simple Indicator 15, F(6, 479) = 3.954, p < 0.05 for Simple Indicator 16, F(6, 494) = 3.044, p < 0.05 for Simple Indicator 18, F(6, 494) = 3.220, p < 0.05 for Simple Indicator 19, and F(6, 490) = 6.830, p < 0.001 for Simple Indicator 20.

Furthermore, statistically significant differences were found for Dimension 1 – F (6, 501) = 5.866, p < 0.001, Dimension 3 – F (6, 475) = 3.510, p < 0.05, Dimension 5 – F(6, 502) = 2.610, p < 0.05, and Dimension 6 – F(6, 502) = 6.037, p < 0.001.

Finally, statistically significant differences in the means of the clusters were found for economic well-being, F(6, 502) = 5.131, p < 0.001, for non-economic well-being, F(6, 502) = 3.750, p < 0.05, and for general well-being, F(6, 502) = 5.265, p < 0.001 (Appendix, Table 11).

3.5 Geographical Distribution of Children’s Well-being (School Year 2020–2021)

On the basis of Synthetic Indicators 1,3,4, and 5 (${\mathrm{CWG}}_{i}$, ${\mathrm{CWE}}_{i},{\mathrm{CW{\rm N}}\mathrm{E}}_{i}$, ${\mathrm{CWG}}_{\mathrm{i}}$) (Leriou et al., 2021), the percentages of children below and above the threshold of being at risk of lacking general, economic, and non-economic well-being, by cluster, school level, and gender, are estimated for the school year 2020–2021. Furthermore, the results of the Pearson’s chi-squared test are included.

According to Table 12 ¹³ (Appendix), the percentages of children below the threshold in terms of economic, non-economic, and general well-being by clusters (Leriou, 2022; Leriou et al., 2021, 2022a, b), prove statistically significant differences in economic well-being (χ²_{[6, N = 1,623]} = 59.868, p < 0.001), non-economic well-being (χ²_{[6, N = 1,623]} = 36.503, p < 0.001), and general well-being (χ²_{[6, N = 1,623]} = 44.834, p < 0.001). As far as the school year 2020–2021 and all clusters (Appendix, Table 12) are concerned, the percentage of all children below the threshold is 11.5% in general, 14.6% in economic, and 21.3% in non-economic well-being. In particular, concerning general well-being, the highest percentages of the total number of children below the threshold are found in Clusters 3 (17.9%), 4 (15.0%), 5 (8.7%), 1 (7.5%) and 2 (6.8%), while Clusters 7 (0.0%) and 6 (3.8%) represent the lowest percentages in increasing order. From the examination of the parameter of economic well-being, the percentages below the threshold appear to be higher in Clusters 3 (22.6%), 4 (19.6%), 5 (9.7%), 1 (9.6%), and 2 (9.5%), whereas Clusters 7 (1.8%) and 6 (3.8%) account for the lowest percentages. By focusing on the factor of non-economic well-being, it was concluded that the highest percentages of children below the threshold are observed in Clusters 1 (30.5%), 3 (27.4%), 5 (22.2%), 2 (19.0%), 4 (17.9%), and 6 (12.1%) and the lowest is observed in Cluster 7 (5.5%) (Appendix, Table 12).

Moreover, the results of a Pearson’s chi-squared test within school categories portray that economic (χ²_{[2, N = 1,623]} = 71.705, p < 0.001), non-economic (χ²_{[2, N = 1,623]} = 33.115, p < 0.001) and general (χ²_{[2, N = 1,623]} = 19.414, p < 0.001) well-being are associated with the school category of children. More specifically, a comparison of all school categories demonstrates that the highest percentage of children below the threshold of being at risk in general well-being in Attica is in the elementary school category (17.3%), followed by junior high school (11.4%), with the high school level coming last (8.5%) (Appendix, Table 12). Similarly, elementary school category constitutes the highest percentage of children (27.3%) falling under the threshold of being at risk in economic well-being in Attica (Appendix, Table 12). More specifically, extremely and, alarmingly higher shares of elementary school children are below the economic well-being threshold (χ²_{[6, N = 381]} = 37.703, p < 0.001) in Clusters 3 (43.5%) and 4(34.5%), as compared to other municipalities clusters of Attica (Appendix, Table 12). Therefore, children in Clusters 3 and 4, in the elementary school category for the school year 2020–2021, face the worst situation in terms of economic well-being (Appendix, Table 12).

Furthermore, the general well-being (χ²_{[6, N=381} = 23.790, p = 0.001) of elementary school children is associated with the cluster (Appendix, Table 12). A higher representative number of elementary school children is below the general well-being threshold in Cluster 3 (28.2%), as compared to other municipalities clusters (Appendix, Table 12). Regarding non-economic well-being of elementary school children, it falls into the marginally significant level (p > 0.05 and p < 0.15). Similarly, a higher share (χ²_{[6, N=381} = 10.905, p = 0.091) of elementary school children is below the non-economic well-being threshold in Cluster 3 (20.0%), as compared to other clusters (Appendix, Table 12).

Τhe economic (χ²_{[6, N=499} = 21.983, p = 0.001) and general (χ²_{[6, N=499} = 20.519, p = 0.002) well-being of junior high school children are associated with the municipalities cluster (Appendix, Table 12). More specifically, higher shares of junior high school children are below the economic well-being threshold in Clusters 3 (28.2%), 4(14.4%), 1(14.3%) and 5(10.5%) as compared to other clusters (Appendix, Table 12). A higher representative number of junior high school children is below the general well-being threshold in Cluster 3 (23.5%), as compared to other clusters (Appendix, Table 12) children are below the non-economic well-being threshold (χ²_{[6, N=499} = 11.585, p = 0.072) in Clusters 3 (28.2%), 5 (27.9%) and 4 (21.9%) as compared to other clusters (Appendix, Table 12).

Regarding high school children, their non-economic (χ²_{[6, N = 743]} = 34.210, p < 0.001) well-being is associated with the municipalities cluster. Specifically, higher shares of high school children are below the non-economic well-being threshold in Clusters 1 (41.2%) and 3 (30.3%), as compared to other clusters (Appendix, Table 12).

In addition, regarding male students, their economic (χ²_{[6, N = 730]} = 18.949, p = 0.004), non-economic (χ²_{[6, N = 730]} = 16.798, p = 0.010) and general (χ²_{[6, N = 730]} = 13.946, p = 0.030) well-being are associated with the cluster. Higher shares of male students are below the general well-being threshold in Clusters 3 (16.9%), and 4 (13.2%) as compared to other clusters (Appendix, Table 12). A higher representative number of male students is below the economic well-being threshold in Cluster 3 (21.5%), as compared to other clusters (Appendix, Table 12). Furthermore, a higher share of male students is below the non-economic well-being threshold in Cluster 1 (31.2%), as compared to other clusters (Appendix, Table 12).

Regarding female students, their economic (χ²_{[6, N = 845]} = 42.260, p < 0.001), non-economic (χ²_{[6, N = 845]} = 24.658, p < 0.001) and general (χ²_{[6, N = 845]} = 32.049, p < 0.001) well-being are also associated with the cluster. Specifically, higher shares of female students are below the economic well-being threshold in Clusters 3 (23.3%), and 4 (21.3%) as compared to other municipalities clusters (Appendix, Table 12). A higher representative number of female students is below the non-economic well-being threshold in Cluster 1 (31.1%), as compared to other clusters (Appendix, Table 12). In addition, a higher representative number of female students is below the general well-being threshold in Cluster 3 (18.3%), as compared to other clusters (Appendix, Table 12).

Furthermore, the results of a Pearson’s chi-squared test within family structure categories portray that economic (χ²_{[2, N = 1,524]} = 45.265, p < 0.001), and general (χ²_{[2, N = 1,524]} = 25.888, p < 0.001) well-being are associated with the family structure category of the children. More specifically, a comparison of all family structure categories demonstrates that the highest percentage of children below the threshold of being at risk in general well-being in Attica is in single-parent family with father category (16.6%), followed by single-parent family with mother (14.8%), while the two-parent family category coming last (7.1%) (Appendix, Table 13¹⁴). Similarly, single-parent family with father category constitutes the highest percentage of children (21.5%) falling under the threshold of being at risk in economic well-being in Attica (Appendix, Table 13). More specifically, a higher share of children in single-parent family with father category is below the general well-being threshold (χ²_{[6, N = 181]} = 15.971, p = 0.014) in Cluster 4 (29.6%), as compared to other clusters (Appendix, Table 13). In addition, a higher representative number of students in single-parent family with father category is marginally below the economic well-being threshold (χ²_{[6, N = 181]} = 11.683, p = 0.069) in Cluster 4 (31.5%), as compared to other clusters (Appendix, Table 13).

3.6 Comparison Between the 1st and 2nd Semesters of School Year 2020–2021: The Evolving Impact of the COVID-19 Crisis on Child Well-being

A comparison of the results on the basis of Synthetic Indicators (Leriou et al., 2021) pertaining to the surveys conducted during both the first (round four of the research) and second semester (round five of the research) of the school year 2020–2021 follows (Appendix, Tables 14 and 15). The results of the first semester are presented and discussed by Leriou (2022).

The results of the Pearson’s chi-squared tests show no statistically significant differences in the percentages of students below the economic, non-economic, and general well-being threshold by semester in the school year 2020–2021 (Appendix, Table 14¹⁵).

Further analysis (Appendix, Table 15¹⁶) on the level of simple indicators${(\mathrm{w}}_{{\mathrm{S}}_{\mathrm{j}},{\mathrm{\rm N}}_{\mathrm{i}}}$) (Leriou et al., 2021) and dimensions (${\mathrm{W}}_{{\mathrm{D}}_{\mathrm{\rm N}},\mathrm{i}}$) (Leriou et al., 2021) shows that the school semester is associated statistically with the number of students reporting humid or cold houses (χ²_{[1, N=1,371]} = 7.899, p = 0.005), no sufficient effort of their schools to urge them to love the animals (χ²_{[1, N=1,542)} = 26.951, p < 0.001), and the nature (χ²_{[1, N=1,555)} = 11.070, p = 0.001), as well as the lack of playing outdoors (χ²_{[1, N=1,594]} = 14.389, p < 0.001), and no vacation (χ²_{[1, N=1,596]} = 36.422, p < 0.001). In addition, when the second semester is compared to the first, the percentage of children below the threshold presents a marginal (p > 0.05 and p < 0.15) increase (χ²_{[1, N=1,526]} = 2.073, p = 0.1499) in D.3 (2.9%) (Appendix, Table 15 and Fig. 7), an increase (χ²_{[1, N=1,622} = 5.030, p = 0.025) in D.6 (5.1%) (Appendix, Table 15), and a marginal decrease (χ²_{[1, N=1,619} = 2.813, p = 0.094) in D.5 (4.2%) (Appendix, Table 15).

An analysis on the level of clusters regarding D.3 shows that the school semester is associated statistically with the number of students living in jobless households (χ²_{[1, N=176]} = 5.851, p = 0.016), in Cluster 1 (Appendix, Table 16¹⁷). More specifically, when the second semester is compared to the first, the percentage of children below the threshold in Cluster 1, presents an increase in D.3 (8.8%) (Appendix, Table 16 and Fig. 8).

3.6.1 The Diachronic Impact of COVID-19 Over a Longer Timeframe (2019–2021)

To investigate the diachronic impact of COVID-19 in Attica over a longer timeframe (Leriou et al., 2022b, Sect. 5), further analysis of Dimension 3 (unemployment), regarding Rounds 2, 3, 4 and 5, via two methodologies of thresholds, is carried out (Appendix, Table 17¹⁸). More specifically, for the school year 2019–2020 and the first methodology, the analysis shows a sharp increase of 5 percentage points (χ²_{[1, N=1,618]} = 8.452, p = 0.004), in the second semester, vis-a-vis the first semester, in the number of students who are at risk of lacking well-being because of their parents’ unemployment (Leriou et al., 2022b, Sect. 3.6). As far as the school year 2019–2020 and the second methodology are concerned, an alarmingly sharp increase of almost 13 percentage points (Leriou et al., 2022b, Sect. 4.1) is observed, in the second semester, vis-a-vis the first semester, in the number of the students living in jobless households (χ²_{[1, N=1,618]} = 23.101, p < 0.001) (Appendix, Table 17). Further, for the school year 2020–2021 and the first methodology, the analysis shows a marginal increase of almost 3 percentage points in the second semester, vis-a-vis the first semester, in the number of students at risk in well-being because of their parents’ unemployment (Appendix, Tables 15 and 17). Regarding the school year 2020–2021 and the second methodology, an extremely alarming increase of almost 15 percentage points (χ²_{[1, N=1,526]} = 31.662, p < 0.001) is seen in the second semester, vis-a-vis the first semester in the number of the students living in jobless households (Appendix, Table 17).

3.7 Principal Component Analysis (School Year 2020–2021)

A Principal Component Analysis (PCA) is performed (Appendix) to confirm the aforementioned results and discover which dimensions contribute to child well-being and which do not. A linear combination of two components of well-being is established, which shows that the dimensions that contribute the most to child well-being are Dimensions 5 and 2 (moral education and nutrition, respectively, as shown in Appendix, Table 18¹⁹) and they also arose in all the previous rounds of the research (Leriou, 2022; Leriou et al., 2021, 2022a, b). In addition, the aforementioned is confirmed by the MCA conducted regarding the second semester of the 2020–2021 school year. Hence, the correctness of the theoretical and methodological framework of this study is further confirmed. These are followed by Dimensions 1 (housing conditions), 3 (unemployment), 4 (healthcare), and finally, 6 (leisure).

The scatter plot (Appendix, Fig. 9) indicates the clusters that are recording the best and the worst results for the well-being of children, according to the linear combination of the components generated from the PCA (Appendix). More specifically, Fig. 9 (Appendix) shows that the best results are found in Clusters 6 and 7, while the worst results are observed in Clusters 3 and 4. This has been confirmed by the prior analysis of general well-being (Appendix, Table 12) since the clusters with the best and the worst well-being concord with the results of the PCA.

3.8 Results of Means (School Year 2020–2021)

The processing of data on the grounds of Composite Indicators 7 – 12 (Leriou, 2022; Leriou et al., 2021, 2022a, b) reveals that with regard to the dimensions and all clusters, the most serious problem appears in a dimension belonging to the non-economic aspect of child well-being, which is D.5 (Appendix, Table 19²⁰). The dimension that follows is D.3 of economic child well-being (Appendix, Table 19).

The results of the one-way ANOVA (Appendix, Table 19) indicate statistically significant differences F(6, 1,548) = 5.430, p < 0.001, in the means of the clusters for Simple Indicator 10, F(6, 1,564) = 8.220, p < 0.001 for Simple Indicator 2, F(6, 1,500) = 3.952, p < 0.05 for Simple Indicator 1, F(6, 1,442) = 5.598, p < 0.001 for Simple Indicator 11, F(6, 1,412) = 2.565, p < 0.05 for Simple Indicator 5, F(6, 1,434) = 2.483, p < 0.05 for Simple Indicator 7, F(6, 1,482) = 2.380, p < 0.05 for Simple Indicator 8, F(6, 1,416) = 4.409, p < 0.001 for Simple Indicator 6, F(6, 1,519) = 2.793, p < 0.05 for Simple Indicator 13, F(6, 1,535) = 9.896, p < 0.001 for Simple Indicator 14, F(6, 1,548) = 6.319, p < 0.001 for Simple Indicator 15, F(6, 1,543) = 9.705, p < 0.001 for Simple Indicator 16, F(6, 1,585) = 3.727, p < 0.05 for Simple Indicator 18, F(6, 1,587) = 7.288, p < 0.001 for Simple Indicator 19, and F(6, 1,589) = 16.774, p < 0.001 for Simple Indicator 20.

Furthermore, statistically significant differences were found for Dimension 1 – F (6, 1,615) = 9.349, p < 0.001, Dimension 2 – F (6, 1,590) = 4.955, p < 0.001, Dimension 3 – F(6, 1,519) = 7.618, p < 0.001, Dimension 4 – F(6, 1,185) = 3.683, p < 0.05, Dimension 5 – F(6, 1,612) = 4.698, p < 0.001, and Dimension 6 – F(6, 1,615) = 15.986, p < 0.001.

Finally, statistically significant differences in the means of the clusters were found for economic well-being, F(6, 1,616) = 11.742, p < 0.001, for non-economic well-being, F(6, 1,616) = 7.261, p < 0.001, and for general well-being, F(6, 1,616) = 10.379, p < 0.001 (Appendix, Table 19).

4 Discussion

The processing of data based on Synthetic Indicators 1,3,4 and 5 ( ${\mathrm{CWG}}_{i}$, ${\mathrm{CWE}}_{i},{\mathrm{CW{\rm N}}\mathrm{E}}_{i}$, ${\mathrm{CWG}}_{\mathrm{i}}$) (Appendix, Table 2; Leriou et al., 2021, Sect. 3.5, p. 12) shows that, for the second semester of school year 2020–2021, the worst situation regarding general child well-being is found in Cluster 4 (Athens, Ilioupoli, etc.) (Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021). Furthermore, in terms of school category, elementary school children face the worst situation regarding their economic well-being for the first semester of the school year 2020–2021. In addition, in terms of family structure category, the single-parent families with father children face the worst situation regarding their economic well-being (Appendix, Table 3).

Moreover, regarding the second semester of the school year 2020–2021, the processing of data pertaining to Synthetic Indicator 2 (${\mathrm{W}}_{{\mathrm{D}}_{\mathrm{\rm N}},\mathrm{i}}$) (Appendix, Table 5) reveals that at the dimensional level, in Attica, the gravest problem is observed in Dimension 5 (moral education) (Aristotle, 1926, pp. 47–59, 1932, pp. 635–643, 1934, pp. 25–47, 1952, pp. 257–263; Leriou, 2016, 2022, Appendix, Fig. three; Leriou et al., 2021; Little, 1949, p. 244, 246, 1950, pp. 6, 13–14; Michalos, 2015; Michalos et al., 2012; Pigou, 1920, pp. 12–14, 21; Plato, 1925, pp. 245–257, 1926; Scitovsky, 1941, p. 79, 1951, p. 303, 305, 315; Smith, 1759; Tasopoulos & Leriou, 2014; Zolotas, 1982, pp. 21–22, 47, 55, 59, 162, 171, 191) related to the non-economic aspect of well-being. The dimension that follows (Appendix, Table 5) regarding the economic aspect of well-being is Dimension 2 (nutrition) (Leriou, 2016, 2019, 2022, Appendix, Fig. three; Leriou et al., 2021).

Furthermore, regarding the second semester of the school year 2020–2021, the results on the basis of Composite Indicators 7–12 (Appendix, Table 11; Leriou et al., 2021, Sect. 3.5, pp. 12–13) are confirmed by the findings from previous rounds of the research (Leriou, 2022; Leriou et al., 2021, 2022a, b).

As shown in Fig. 5  (Appendix) regarding the second semester of the school year 2020–2021, classified on the right side of Axis 1 are students mainly from Clusters 4 (Athens, Ilioupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) and 3 (Kallithea, Ilion, Drapetsona, Petroupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021), who tended to report that they were nutritionally deprived (Appendix, Fig. 5), resided in homes with multiple problematic conditions, and lived in jobless households (Appendix, Table 10). Essentially, this children’s profile reflects harsh economic conditions related with their appropriate, qualitative and balanced nutrition and with appropriate, safe and facilitating home conditions, because of the unemployment of their parents. This children’s profile needs immediately social protection from competent bodies. Findings regarding Axis 1 are consistent with the Table 5 (Appendix). Hence, this confirms the correctness of the findings of the fifth round of the research. On the left side of Axis 2 (Appendix, Fig. 5), junior high and high school students (Sch. Levels 2 and 3), mainly girls (Gender 2) from Clusters 6 (Voula, Papagou, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) and 1 (Aspropyrgos, Megara, Perama, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) are located very close to the below-threshold categories of the simple indicators of D.5 – “Empathy for all people,” “Love of animals,” “Love of nature,” and “Love of people,” (Appendix, Fig. 5). These findings are consistent too with the Table 5 (Appendix), regarding the second semester of the school year 2020–2021. Hence, this confirms further the correctness of the findings of the fifth round of the research. In the light of the aforesaid, girls from the superior school categories, mainly in Clusters 6 (Voula, Papagou, etc.; Leriou, 2022, Appendix, Table ixteen; Leriou et al., 2021) and 1 (Aspropyrgos, Megara, Perama, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021), were strongly dissatisfied with their schools’ efforts to cultivate human values in them. On the contrary, regarding the second semester of the school year 2020–2021, boys from the elementary school category, mainly in Cluster 2 (Nea Smyrni, Palaio Faliro, Marousi, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) were very satisfied with their schools’ efforts to cultivate some human values in them. More specifically, on the right side of Axis 2 (Appendix, Fig. 5), elementary school students (Sch. Level 1), mainly boys (Gender 1) from Cluster 2 (Nea Smyrni, Palaio Faliro, Marousi, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) are located very close to the above-threshold categories of the simple indicators of D.5 – “Love of people,” “Love of nature,” and “Love of animals,” (Appendix, Fig. 5). These findings are consistent too with the Table 5 (Appendix). Hence, this confirms furthermore the correctness of the findings of the fifth round of the research.

The processing of data on the grounds of Synthetic Indicators 1, 3, 4, and 5 reveals that for the school year 2020–2021 (Appendix, Table 12), the worst situation with regard to general child well-being is found in clusters 3(Kallithea, Ilion, Drapetsona, Petroupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021), 4(Athens, Ilioupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021), 5 (Aegina island, Spetses island, Hydra island, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) 1(Aspropyrgos, Megara, Perama, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021), and 2 (Nea Smyrni, Palaio Faliro, Marousi, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) and the best in Clusters 7 (Kifissia, Filothei, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) and 6 (Nea Smyrni, Palaio Faliro, Marousi, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021). The PCA regarding the school year 2020–2021, confirms these results. Furthermore, the elementary school children in Clusters 3 (Kallithea, Ilion, Drapetsona, Petroupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) and 4 (Athens, Ilioupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) face the worst situation regarding their economic well-being, for the school year 2020–2021. More specifically, the percentages of elementary school children below the threshold in terms of economic well-being, are very alarmingly high. On the grounds of Composite Indicators 7–12, regarding the school year 2020–2021 (Appendix, Table 19), the data reveal that issues found in D.3 (unemployment; Leriou, 2016; Leriou, 2019, 2022, Appendix, Fig. three; Leriou et al., 2021) and D.5 (moral education; Leriou, 2016, 2022, Appendix, Fig. three; Leriou et al., 2021; Tasopoulos & Leriou, 2014) in the school years 2018–2019 (Leriou et al., 2021) and 2019–2020 (Leriou et al., 2022b) also remain in 2020–2021. In addition, Clusters 3 (Kallithea, Ilion, Drapetsona, Petroupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021), and 4 (Athens, Ilioupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) need more direct interventions in order to achieve the social protection of children in terms of unemployment in their parents (Appendix, Table 19).

With regard to synthetic indicators about economic, non-economic and general child well-being, the non-significant statistical difference between the two semesters of the school year 2020–2021 (Appendix, Table 14), indicates that regarding the second semester, measures may have been taken in order to restrain the phenomenon of deterioration of general population well-being. In the light of the aforesaid, has been stopped the increasing development of this phenomenon. In addition, during the second semester, the marginal increasing of students living in jobless households causes surprise (Appendix, Table 15), as measures have been taken by the government to intercept the unemployment. This deterioration in the second semester as compared to the first (Appendix, Table 16) is mainly due to Cluster 1 (Aspropyrgos, Megara, Perama, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021). More specifically, in this Cluster undeclared work is more widespread than in others, according to the C.W.-SMILE qualitative research via discussions with school principals. In the light of the aforesaid, no government measure to protect employees from being fired can easily succeed in the framework of undeclared work in Cluster 1 (Aspropyrgos, Megara, Perama, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021).

Furthermore, during the second semester, the marginal improvement of child well-being in terms of moral education (Appendix, Table 15) is may due to the replacement of long-distance learning with traditional education towards the end of the semester. More specifically, the long-distance learning implemented with more difficulty for children than the traditional one (Leriou, 2022, Sect. 5, p.1,980). In the light of the aforesaid, during the distance education, teachers are putting more emphasis on the development of students’ skills rather than on the development of their humanistic values, such as loving and protecting the nature and animals (Leriou, 2022, Sect. 5, p.1,980).

Moreover, during the second semester (Appendix, Table 15), the deterioration of child well-being in terms of D.6 (spare time, playing, and relaxing; Leriou, 2016, 2022, Appendix, Fig. three; Leriou et al., 2021; Michalos, 2015) is mainly because of Simple Indicator 20 (vacation; Leriou, 2022, Appendix, Fig. three; Leriou et al., 2021). More specifically, the most appropriate periods for vacation in Greece because of the sunny weather are probably the seasons of the Easter which was included in the second semester of the school year 2020–2021 and the summer. In Greece, during a long period, including Easter, austere measures were in force because of COVID-19, which prohibited the transfer from one area to another and, consequently, the vacation.

Further, the findings regarding the impact of COVID-19 in Attica over a longer timeframe, indicate that the worst deterioration in D.3 (unemployment; Leriou, 2016, 2019, 2022, Appendix, Fig. three; Leriou et al., 2021) vis-a-vis the second methodology of the threshold, was observed in the fifth wave of the research (Appendix, Table 17). More specifically, according to the first methodology, the threshold of the employment status dimension (D.3) is defined as at least one parent being employed, and being below the threshold refers to no parent being employed. According to the second methodology, the threshold of the employment status dimension (D.3) is defined as two parents in employment. Above the threshold of the employment status dimension (D.3) refers to both of the parents being employed, and below the threshold refers to at least one parent not being employed. In the light of the above, it is obvious that if the percentage of children below the threshold in D.3 (unemployment; Leriou, 2016, 2019, 2022, Appendix, Fig. three; Leriou et al., 2021), was considered to be the most severe, the impact of the COVID-19 pandemic on children’s well-being in the regional unit of Attica was found to be at its worst during this round of the research compared with what it was during the previous (Appendix, Table 17). This finding demonstrates that many of the measures of the government to intercept unemployment have not yielded results for reasons such as undeclared work and the closing of the enterprises.

All these estimations reveal the evolving impact of the universal crisis of COVID-19 on child well-being in the regional unit of Attica, Greece (Appendix, Figs. 7 and 8, Table 17). Moreover, they demonstrate the geographical distribution and the multi-dimensional (Burchi et al., 2018; De Muro et al., 2011; Kainulainen & Paananen, 2021) characteristics of child well-being in the region, in the second semester of the school year 2020–2021 as well as the same year as a whole.

4.1 Suggestions for Action Planning on the Basis of a ChoiCo Game and Fuzzy Logic

Drastic measures, as those suggested by Leriou (2022, Sect. 5.1,pp. 1,980–1,983) must be assumed in order to improve children’s well-being, especially in Clusters 3 (Kallithea, Ilion, Drapetsona, Petroupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) and 4 (Athens, Ilioupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) and the elementary school category. With regards to Simple Indicator 6 (at least one meal weekly with meat or chicken or fish or pulses/vegetables of equal nutritional value; Leriou et al., 2021), emphasis must be placed on single parent family with mother children (Appendix, Table 6). In light of unemployment, emphasis must be focused both on single-parent family with father children and the elementary school category (Appendix, Tables 7 and 8). In terms of moral education, emphasis must be placed on the junior high school category (Appendix, Table 9).

Moreover, in the light of the findings of a business meeting of the C.W.-SMILE with the Local Authorities, another suggestion for improvement child well-being in terms of Simple Indicator 6 (at least one meal weekly with meat or chicken or fish or pulses/vegetables of equal nutritional value; Leriou, 2022, Appendix, Fig. three; Leriou et al., 2021), would be as follows: An extra practice, as an alternative of the practice of the school meals, is to implement permanent support in the nutrition of all children through the breadlines or permanent provision of the school brunch from all the municipalities to all schools and all students. More specifically, the school brunch could be strengthened and could evolve to a practice with complete, nutritious, child meals, equivalent to the practice of the school meals and not just including a simple sandwich, as is customary until now. In this regarding, should be taking place a cooperation between KEDE (Central Union of Municipalities of Greece), the Ministry of Labor and Social Affairs and the Ministry of Education of Greece.

Furthermore, regarding the findings of the aforesaid business meeting, one more suggestion is as follows: The competent authorities have to impose a re-examination and time limit on the unemployed who receive chronic social assistance from competent authorities in order to motivate them to search for a job and to not rest, always living on income support and social assistance. This will eliminate the situation where some people choose to live voluntarily on income support and social assistance and have no interest in looking for work. This will lead to a better and fairer distribution of resources intended for social assistance and will help those people who have children and are in poverty because they really cannot find employment even though they are willing to work.

All these suggestions for social protection and social welfare could be based on an auxiliary tool for decision making as a ChoiCo Game using fuzzy logic which is developed in the next sub-section.

4.1.1 A ChoiCo Game using Fuzzy Logic (Artificial Intelligence) as an Auxiliary Tool for Action Planning

The decision-makers (Michalos, 1978) of social welfare who aim to implement measures for the improvement of total general child well-being (e.g., in all of Attica) may consider the results pertaining to Composite Indicator 7 (Leriou et al., 2021) for the second semester of the school year 2020–2021 (Appendix, Table 11), having the following form: $\mathrm{CWT}=\left(0.8631+0.8346+0.5405+0.9360+0.5142+0.7036\right)/6=0.7320$. Given that the score of perfect well-being is 1, they may set goals for increasing the total well-being in relation to 1 by keeping in mind the well-being scores corresponding to the dimensions. Moreover, if they aim to improve the most contributing dimensions to child well-being, according to MCA, they can plan measures based on D.2 and D.5. The rules of a fuzzy inference system (artificial intelligence) are applied as an auxiliary tool for decision-making, similar to that used in the first round of the research (Leriou et al., 2021). So, if, according to these rules, the child well-being for the second semester of the school year 2020–2021 is at Ch.W.9. (Leriou et al., 2021), and the decision-makers aim to reach a well-being score of Ch.W.12., there are 22 rules from a total of 2,985,984 to meet this objective. To further narrow down this number of rules, they can factor in two criteria that should apply at the same time, i.e., a) D.2 and D.5 larger than 0.93; and b) the rules in which each entry or each dimension is larger or equal to the average values of all the data. Thus, they can limit the rules to 7 (Appendix, Table 20). Let’s suppose that they choose the following rule: R. 2,985,984: IF(D.1 = 0.98) AND(D.2 = 0.98) AND(D.3 = 0.98) AND(D.4 = 0.98) AND(D.5 = 0.98) AND(D.6 = 0.98), THEN(CWT = 0.98) (Appendix, Table 20). Based on this rule, to achieve the goal of 0.98 in the following year (i.e., an improvement of total general well-being by 25%), the decision-maker must improve D.1 by 12%, D.2 by 15%, D.3 by 44%, D.4 by 4%, D.5 by 47% and D.6 by 28%. Therefore, the next step is to develop measures that will drive child well-being towards these arithmetic goals.

Based on the aforesaid, a ChoiCo game (Grizioti & Kynigos, 2021; Grizioti et al., 2021; Kynigos & Grizioti, 2020) was designed in order to educate persons regarding the child well-being (http://etl.ppp.uoa.gr/choico/?cwsmile). More specifically, players have to make choices of policies in order to maximize child well-being in region of Attica, Greece. In this framework, they have to set higher arithmetic goals for total child well-being (CWT) and for all dimensions of CWT, in total of Attica and in each separate cluster with municipalities of Attica, than the corresponding arithmetic results (Appendix, Table 11) of the fifth round of the research (Appendix, Fig. 10). All the corresponding arithmetic results of the fifth round (Appendix, Table 11) are depicted on the images in this game. In addition, they have to put emphasis on dimensions that appear to have the gravest problems. In this game, all the policies are rules of the 2,985,984 fuzzy inference rules (artificial intelligence) from a file of 86,655 pages (Leriou et al., 2021) which are depicted in Table twenty one. All messages (Fig. 1 and Appendix, Fig. 11) that appear when a player selects a point of a policy are educational regarding the theoretical and methodological framework of C.W.-SMILE. Similarly, point information and the map with clusters are educational regarding the logic of the C.W.-SMILE. From this point of view, a potential player can learn about child well-being via a fun and easy way. Moreover, this ChoiCo game (http://etl.ppp.uoa.gr/choico/?cwsmile) can be used as an easy and fun auxiliary tool in any given action plan for the improvement of child well-being regarding the results of the fifth round of the research.

5 Conclusions

This paper presents the results of the fifth round of the research related to the second semester of the school year 2020–2021 as well as the findings of this school year. All outcomes are further confirmed by an MCA and a PCA. More specifically, the paper reveals the multi-dimensional characteristics, the extent, and the geographical distribution of child well-being in the region of Attica, Greece, for the fifth round of an ongoing research. Moreover, the evolving impact of the COVID-19 crisis on child well-being is detected and policies based on a ChoiCo game are suggested.

Regarding the main results of the fifth round (second semester of the school year 2020–2021), the value of the Composite Indicator of the Total General Well-being (CWT; Leriou, 2016; Leriou et al., 2021) is 0.73, while the percentage of the total number of children in Attica above or below the threshold of being at the risk in general well-being based on the Synthetic Indicators of Personal General Child Well-being (${\mathrm{CWG}}_{\mathrm{i}}$; Leriou et al., 2021) is 9.0%. Moreover, the results show that elementary school children, mainly in Clusters 3 (Kallithea, Ilion, Drapetsona, Petroupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021) and 4 (Athens, Ilioupoli, etc.; Leriou, 2022, Appendix, Table sixteen; Leriou et al., 2021;), face the worst situation regarding their economic well-being, for the school year 2020–2021.

As far as the improvement of child well-being is concerned, the decision-makers of social welfare should use an easy, educational and fun, auxiliary tool in any given action plan, as the specific ChoiCo game of C.W-SMILE (http://etl.ppp.uoa.gr/choico/?cwsmile), which is based on Fuzzy Logic (artificial intelligence).

Appendix A

Tables

Table 1 Number of students in the schools/SC from which the sample was obtained per school category and per cluster

1,

Table 2 Percentage of students above or below the threshold of being at risk of lacking economic, non-economic, and general child well-being, by cluster, school level, and gender regarding the 2nd semester of the school year 2020–2021

2,

Table 3 Percentage of students above or below the threshold of being at risk of lacking economic child well-being, per family structure category regarding the 2nd semester of the school year 2020–2021

3,

Table 4 At risk of lacking child well-being thresholds per simple indicator and dimension

4,

Table 5 Percentage of students on the threshold and below of being at risk of lacking child well-being per simple indicator, per dimension and per cluster for the 2nd semester of the school year 2020–2021

5,

Table 6 Percentage of students on the threshold and below in S.I.6, per family structure category for the 2nd semester of the school year 2020–2021

6,

Table 7 Percentage of students on the threshold and below in D.3 per family structure category for the 2nd semester of the school year 2020–2021

7,

Table 8 Percentage of students on the threshold and below in D.3, per school category for the 2nd semester of the school year 2020–2021

8,

Table 9 Percentage of students on the threshold and below in D.5, per school category for the 2nd semester of the school year 2020–2021

9,

Table 10 Categories of each simple indicator (+ = above threshold, – = equal or below threshold) with the highest contribution to each factorial axis for the 2nd semester of the school year 2020–2021

10,

Table 11 Mean score per dimension, indicator, and per cluster for the 2nd semester of the school year 2020–2021

11,

Table 12 Percentage of students above or below the threshold of being at risk of lacking economic, non-economic, and general well-being by cluster, school level, and gender for the school year 2020–2021, N = 1,623

12,

Table 13 Percentage of students above or below the threshold of being at risk of lacking economic, non-economic and general well-being by family structure category for the school year 2020–2021

13,

Table 14 Percentage of students above or below the threshold of being at risk of lacking general, economic, and non-economic child well-being, by semester for the school year 2020–2021

14,

Table 15 Percentage of students on the threshold and below of being at risk of lacking child well-being, per simple indicator, and dimension, by school semester for the school year 2020–2021

15,

Table 16 Percentage of students below the threshold of being at risk of lacking child well-being, in terms of D.3, per cluster, by school semester for the school year 2020–2021

16,

Table 17 Percentage of students below the threshold of being at risk of lacking well-being in terms of D.3, per school year, by school semester, via the two methodologies of the threshold

17

Figures 

Fig. 2

Geographical distribution of general child well-being in Attica for the 2nd semester of the school year 2020–2021

2,

Fig. 3

Geographical distribution of economic child well-being in Attica for the 2nd semester of the school year 2020–2021

3,

Fig. 4

Geographical distribution of non-economic child well-being in Attica for the 2nd semester of the school year 2020–2021

4,

Fig. 5

Contribution of Variable Categories (2 = above threshold, 1 = equal or below threshold) for the 2nd semester of the school year 2020–2021

5,

Fig. 6

The most important (or, contributing) variable categories to 1st and 2nd axes – MCA

6,

Fig. 7

The evolving impact of the COVID-19 crisis on child well-being in terms of unemployment

7,

Fig. 8

The evolving impact of the COVID-19 crisis on child well-being in terms of unemployment, in Cluster 1

8

1.1 Principal Component Analysis

A PCA is performed for the school year 2020–2021. The correlation of each of the dimensions and the linear combination of the components are displayed in Table 18. It is observed that the first five dimensions have a strong positive correlation to the well-being indicators and their Pearson correlation coefficient, is statistically significant, different than zero. In addition, the sixth dimension has a positive correlation to the well-being indicators and its Pearson correlation coefficient, is statistically significant, different than zero. Therefore, the PCA demonstrates a close connection to all the six dimensions of well-being adopted in this study.

Tables

Table 18 Pearson’s correlations coefficient between the two components and the six dimensions

18,

Table 19 Mean score per dimension, indicator, and per cluster for the school year 2020–2021

19,

Table 20 The 7 fuzzy inference rules

20,

Table 21 Policies of the Game ( http://etl.ppp.uoa.gr/choico/?cwsmile)

21

Figs. 

Fig. 9

Clusters by the best and worst well-being

9,

Fig. 10

Game Instructions (http://etl.ppp.uoa.gr/choico/?cwsmile)

10,

Fig. 11

Another message from the Game (http://etl.ppp.uoa.gr/choico/?cwsmile)

11