Positive Resources of School Class Communities—Determinants of Student Satisfaction

Abstract

The aim of this study was to determine the factors affecting students’ satisfaction with their class community by measuring the contributions of “positive” resources: either the socio-demographic and individual characteristics of students, or common characteristics of the class community. The research was conducted anonymously using multiple questionnaires on a sample of 267 students attending the higher grades (5th–8th grades) of primary schools from an urban area of the city of Split. It was determined that social cohesion and prosocial behaviour within the class community were the strongest determinants and were also the most important positive resources of satisfaction with the class community. Individual psychological characteristics (especially self-esteem and hope) and the socio-demographic characteristics of students also played a significant role in explaining satisfaction with the class community, but their relative contribution was much weaker than social relations within the class community. The three types of psychological characteristics of the students were positively related to the degree of satisfaction with the class community, but this association was at a low level. More than half of the students were not completely satisfied with their class community. Furthermore, based on the five fundamental dimensions of intrinsic (IM) and extrinsic motivation (EM) within PE motivation, four types of students with different motivational characteristics were determined: the very low type; moderate type; very high IM and low EM type; and extremely high type. It is recommended that experts continue researching the relationship between psychological and social variables in class communities, as well as investigating the effectiveness of possible interventions aimed at the development and improvement of social relationships in the educational and sports talent development environment.

1. Introduction

The class community represents a group of students who attend classes together and among whom there is strong social interaction. This is a mandatory community that every student faces from the first grade of primary school, and it is important to study the dynamics of such groups. One of the main predictors of positive growth and development is a favourable peer environment, especially in the period of adolescence [1]. In today’s age of digitalisation, which affects children from an early age, social interactions are beginning to become more complex [2]. An insufficient approach to this problem can lead to different phenomena, including antisocial interactions that can damage the pleasant atmosphere of a class community [3,4]. As part of this research, the above-mentioned issue was analysed from the aspect of kinesiology education; therefore, the class community was observed in the context of PE.

1.1. Social Dynamics, Prosocial Behaviour and Cohesion

Social behaviours are interactions between one or more individuals. This term can be defined from a biological aspect. In a broad sense, social behaviours can be defined as any modality of communication and/or interaction between two conspecifics of a given species; they are observed in species as simple as single-celled microorganisms and species as complex as humans [5,6]. Social behaviour, as a superordinate term or hypernym, includes prosocial as well as antisocial behaviour. In other words, prosocial behaviour is a type of social behaviour that benefits other people or society as a whole [7]. Furthermore, numerous studies describe the relationship between prosocial and antisocial behaviours as complementary, emphasizing that, for example, empathy, sympathy, or both, which can be defined primarily in affective terms, may inhibit aggressive and antisocial behaviours [8,9,10]. Caprara and Pastorelli (1993), in their research on the connection between aggressiveness and prosocial behaviour, also note an inverse relationship [11]. Therefore, due to the complementarity of this relationship, it is possible to assume that pronounced prosocial behaviour within a social community will result in a low level of antisocial behaviour.

One of the most interesting phenomena within social behaviour as a sub-concept of “negative” resources is aggression. The interest in the above-mentioned phenomenon lies in the fact that aggressiveness is widespread in the social spectrum [12]. One of the most significant problems is the occurrence of peer violence in schools, which can negatively affect an individual’s mental health [13,14,15]. Peer violence has a negative effect on academic success, which also represents a major problem in the growth and development of students [16]. Many forms of aggression and violence are known, from smaller acts such as name-calling or pushing to more serious acts such as hitting, kicking, or punching, to very serious acts such as stabbing, shooting, or killing [17]. There are many different definitions of aggression in the literature due to the complexity of the mentioned term. The authors who dealt with aggression among young athletes used a definition of aggression taken from social psychology [17]. According to the aforementioned definitions, aggression is behaviour that is intended to harm another person, who is then motivated to avoid that harm [17,18,19]. In addition to the above, another definition is used by the World Health Organization, which defines violence as the intentional use of physical force or power, threatened or actual, against oneself or against a group or community, which either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment, or deprivation [20,21].

In contrast, a positive side of the social spectrum is prosocial behaviour, which represents all those interactions that bring benefits to an individual or a community [22]. From the aspect of ontogenetic development, prosocial behaviour is visible from an earlier age [23] and is one of the indicators of optimal growth and development [24,25]. It is highly important to mention the concept of social learning as a basis for the development of prosocial behaviour. People learn prosocial behaviour by observing others, with prosocial behaviour being a positive social interaction that can be learned. In the wake of the aforementioned research, the authors Baumann et al. [26] proposed a three-step developmental sequence, positing that during childhood, prosocial behaviour is a result of material rewards and punishments, while during preadolescence, prosocial behaviour is a product of social and material rewards and punishments; finally, during adolescence and adulthood, prosocial behaviour is a product of internalized self-reward. The learning and manifestation of prosocial behaviour depends on the individual characteristics of the individual, such as a high level of self-esteem, and on situational factors, such as the number of people observing a certain situation [27,28]. It is school, extracurricular activities, and the class community that are the main stages for the manifestation of this kind of behaviour [29,30,31].

Cohesion is defined as “a dynamic process that is reflected in the tendency for a group to stick together and remain united in the pursuit of its instrumental objectives and/or for the satisfaction of member affective needs” [32]. After over fifty years, other authors have confirmed the proposal of Lott and Lott (1965) [33] that cohesion is one of the most important factors within small groups [34,35]. It is necessary to mention the multidimensional conceptualization of cohesion in the domain of sports, within which certain authors propose that this phenomenon should be examined in relation to the problems of the group directed towards tasks and towards society [36,37].

1.2. Self-Esteem, Optimism and Hope

One of the main terms used in the study of positive social behaviours, identified as being one of the most important predictors, is self-esteem. This term was defined by William James back in 1890 as the concept that generally refers to a person’s evaluation of, or attitude toward him- or herself [38]. Self-esteem refers to an individual’s subjective assessment of his or her worth as a person [39,40]. One of the concepts behind this term says that self-esteem does not necessarily reflect a person’s talents and abilities, or even how a person is evaluated by others [41]. Thus, self-esteem is usually expressed as “the feeling that one is ‘good enough’”; that is, individuals with high self-esteem do not necessarily believe that they are superior to others (Rosenberg, 1965, p. 31) [42]. Therefore, unlike the excessive self-esteem and self-aggrandizement that define narcissistic people, self-esteem includes the sentiments of self-acceptance and self-respect [43]. To understand the psychodynamic aspects of PE teaching, the level of satisfaction of PE students is particularly important as a determinant of positive resources. Therefore, in the current study, this parameter will be self-assessed and will be used to detect clusters and related psycho-social parameters.

In addition to the above-mentioned terms, it is extremely important to study optimism among students as one of the most important factors of students’ global satisfaction with their social environment and their class community. Optimism is defined as a set of beliefs that leads people to approach the world in a positive way [44]. Nowadays, the appearance of negative psychological conditions such as anxiety and depression are a daily occurrence on the academic path of young people [45]. Early research in optimism emphasized optimism as a trait or personality disposition that was associated with reduced levels of depression, anxiety, and stress [46]. Therefore, optimism, as a significant factor in preventing the aforementioned conditions that negatively affect the individual and their environment, is extremely important [47].

Next to the concept of optimism is the concept of hope as a positive trait. Hope is defined as an interpersonal characteristic that shows a personal approach to life [48,49]. A high level of this trait enables an individual to have a more favourable social position, improves their level of self-confidence, and reduces negative psychological effects such as anxiety [50,51]. Therefore, a high degree of hope, as a positive trait of an individual, may have a positive effect on the individual’s well-being and, thus, on the entire group in which the subject is located.

1.3. Motivation

One of the main factors in the implementation of the educational process is student motivation [52,53,54]. The degree of motivation also affects cross-curricular outcomes [55] and represents an extremely important component in the education process. In any study of the phenomenon of positive resources regarding the class community, the factor of motivation must not be omitted as a predictor of success within the observed community, as well as success in completing assignments and tasks.

The quality of the teaching process is determined by the intricate relationships that exist between educators, learners, and the subject under study in a broader social context. Kinesiological education increases the complexity of the process by changing students’ overall anthropological status, in addition to teaching new knowledge and abilities. Working within the previously mentioned methodology, students’ morphological, motor, functional, cognitive, and conative qualities show both quantitative and qualitative changes as a result of physical education [56,57,58]. Determining as many variables as possible that may affect the physical education process is crucial for understanding how the future lives of children and adolescents are formed in terms of their growth and development. It is possible to assume the existence of a latent model of the connection between student satisfaction according to the class community with PE and various psycho-sociological characteristics, especially socio-demographic characteristics and individual psychological measures like self-esteem, optimism, hope, and motivation, as well as class community measures like prosocial and social behaviours.

The aims of this study were, first, to determine the determinants of student satisfaction with their class community by measuring the expression of possible “positive” resources (either by the individual psychological characteristics of students or by individual students’ assessments of common characteristics of the class community), which are at the foundation of that satisfaction; second, to determine the types of PE motivation in students, along with a determination of the basic motivational and socio-demographic characteristics of the members of these types.

The goals described in this way can be operationalized through the following partial tasks:

• Analysing the connection between motivation and academic success and the ages of the students;
• Determining the types of the students’ motivational structures;
• Analysing associations between PE motivation types and the socio-demographic characteristics of the students;
• Analysing the distinguishing factors between students with different degrees of satisfaction with the class community;
• Determining the types of students according to their psychological characteristics;
• Analysing associations between the different types of students, grouped according to students’ satisfaction with the class community, as well as by their psychological characteristics.

2. Materials and Methods

2.1. Participants

This research was conducted on a sample of 267 students in higher grades (5th–8th grades) of primary schools from the urban area of the city of Split. There was a total of 134 female and 133 male students. The average age of the students was 13.46 ± 1.01 years. All students were permanent residents of the Republic of Croatia and members of the same ethnic group. The grading system in the Republic of Croatia is based on the numerical grading of all individual school subjects on a five-point scale, as follows: insufficient (1), sufficient (2), good (3), very good (4), and excellent (5). Also, at the end of each school year, students are given a general grade representing the arithmetic mean of the grades of all subjects. Out of the total number of students, 59.2% achieved excellent general academic success at the end of the last school year, 34.5% achieved very good success, and 6.4% achieved good success. A total of 81.6% of students engaged in organised extracurricular sports activities, while 18.4% of students did not. All participants were healthy, with no recorded psychophysical aberrations. Participation in the research was voluntary, with consent obtained from parents or custodians.

2.2. Measures

In this study, three sets of variables were applied: the socio-demographic variables of students, the individual psychological variables of students, and common measures of the class community.

2.2.1. Socio-Demographic Characteristics

Although the questionnaire was applied anonymously and in groups within the class communities, basic socio-demographic characteristics were also collected from the students. This included the following features: gender, age, grade, general academic achievement, and current involvement of the student in extracurricular sports activities.

2.2.2. Individual Psychological Measures

Self-esteem. This measure was measured with the Rosenberg Self-Esteem Scale [42], a 10-item questionnaire used to assess an individual’s self-esteem.

Optimism. This measure was measured by the adapted Life Orientation Test—Revised scale [59], which originally consisted of six items assessing optimism and pessimism. When all six items of the original scale were analysed, the degree of internal reliability of the variable was below an acceptable level (Cronbach’s alpha = 0.54); therefore, the item selection procedure was performed. After conducting the item selection procedure, the adapted scale consisted of three items from the pessimism subscale and one item from the optimism subscale, and the degree of internal reliability of that scale was satisfactory (Cronbach’s alpha = 0.71).

Hope. This measure was measured according to the Children’s Hope Scale [60], which consists of six items for assessing youth skills to identify their goals and the strategies needed to pursue these goals. The internal reliability of that scale was satisfactory (Cronbach’s alpha = 0.82).

Physical education (PE) motivation. PE motivation measures were measured with the adapted Sport Motivation Scale (see Supplementary Materials), which was created for the purpose of measuring PE motivation by Milavic et al. (2015) [61]. The questionnaire was adapted and validated on a sample of Croatian primary school students [61]. The questionnaire consisted of 20 items measuring 5 subscales, as follows: intrinsic motivation to know (IM—Know), intrinsic motivation to accomplish things (IM—Accomplish), intrinsic motivation—stimulation (IM—Stimulation), extrinsic motivation—identified (EM—Identified), and extrinsic motivation—external regulation (EM—Ext. Reg.). Overall PE motivation was determined to be the mean value of all the five subscales’ variables. The variable scores were formed by the condensation of items using the item’s mean determination method. PE motivation measures are of particular interest to the authors of this study because of their specific interest in kinesiological education in primary schools. Furthermore, the authors of this study believe that measures of PE motivation can serve as an adequate substitute for measures of general motivation for learning because they assume that the same mechanisms form the basis of the acquisition of different types of knowledge (whether this is semantic knowledge or the acquisition of motor programs and procedures, as applied in PE classes). Furthermore, the internal reliability of that scale was satisfactory (Cronbach’s alpha = 0.86).

2.2.3. Class Community Measures

Prosocial behaviour. This measure was measured using the Prosocial Behaviour Scale developed by Caprarra and Pastorelli (1993) [11]. This is a questionnaire measuring the child’s behaviour in terms of altruism, trust, and agreeableness, consisting of 10 items. The internal reliability of that scale was satisfactory (Cronbach’s alpha = 0.84).

Social behaviour. This measure was measured using the Social cohesion subscale of the Youth Sports Environment Questionnaire, developed by Eys et al. (2009) [35] for assessing youth cohesion in a group, consisting of 8 items. The internal reliability of that scale was satisfactory (Cronbach’s alpha = 0.89).

Satisfaction with class community. This measure is a single-item measure in which students assess their personal general satisfaction with their class community on a five-point Likert scale (1—not at all; 2—slightly, a little bit; 3—moderate, somewhat; 4—a lot, much; 5—fully, completely).

2.2.4. Experimental Model

The construct of positive youth development (PYD) was taken as the starting point of this research [62]. This perspective emphasizes the importance of and promotes healthy physical and psychosocial development in young people. This approach consists of social–contextual features (e.g., teacher behaviours, classroom structure, and student activities) that help equip young people with attributes, skills, competencies, and values that will contribute to their role as productive, socially conscious, and healthy citizens. The PYD perspective is an approach that views children and adolescents as resources to be developed rather than as problems to be managed [63,64,65,66]. The focus is on promoting skills, competencies, values, and healthy behavioural and psychosocial outcomes as a means of preventing negative health-related behaviours and outcomes. The PYD approach’s goal is to specify how and why educators should be teaching children and adolescents important life skills and core values in physical activity contexts.

Starting from the premise of the complementarity of the relationship between prosocial and antisocial behaviours [7,8,9,10,11], and, based on a reflection on the specificity of the educational process in PE, a theoretical model was constructed that describes the connection between certain students’ psychological characteristics and the characteristics of class community. The model is specified as a simple, interpretable, and practically applicable structure that includes seven explicit variables describing two implicit (latent) concepts. The variables of self-esteem, optimism, hope, and PE motivation form the implicit dimension of individual psychological characteristics, while the variables of prosocial behaviour, social cohesion, and satisfaction with class community form the dimension of class community. Also, a causal connection between the two implicit dimensions was assumed, whereby the individual psychological characteristics dimension affects the class community dimension. The structure of the model is shown in Scheme 1.

2.3. Procedures

Prior to the start of this research, the Ethics Committee of the Faculty of Kinesiology of the University of Split issued its approval for the current study (No.: 2181-205-02-05-20-006, 26 February 2020). The school principals and school boards of the schools where the study was conducted approved the research procedure. The schools obtained the approval of the children’s parents or custodians regarding the children’s participation in the research. The survey was conducted anonymously within the class community, with the support of the surveyor. The survey process lasted from 10 to a maximum of 30 min.

2.4. Data Analysis

As part of the research, the basic metric characteristics of the study measures were determined. The reliability analysis was performed by calculating Cronbach’s alpha coefficient. In order to determine the sensitivity of the measures used, the Kolmogorov–Smirnov max D goodness-of-fit test was performed. The range of the results was also determined, and the skewness and kurtosis coefficients were calculated as indicators of the shape of the distribution.

Descriptive statistics included the calculation of the following parameters: mean, standard deviation, minimal, and maximal results. The means of all scales were computed by the summing of each scale item’s data and were divided by the number of items in that particular scale. This way of calculating the average value provides the possibility for a very easy comparison of the average results of scales with a different number of items. All average values on the scales were interpreted according to the scale recommended by Bavčević, Milavić, and Bavčević (2024) [67]: extremely low (1.00–1.74), very low (1.75–2.24), low (2.25–2.74), moderate (2.75–3.25), high (3.26–3.75), very high (3.76–4.25), and extremely high (4.26–5.00). Univariate differences between the sample subgroups were calculated by using Student’s t-test or by using a one-way ANOVA, and a significance level (p =) for all the difference coefficients was calculated. Pearson’s correlation coefficients were calculated to determine the relationships between the measures used.

The invariance property of the defined experimental model was tested using structural equation modelling (SEM), including the calculation of the reflector matrix and the basic indicators of the constructed model’s adequacy: the discrepancy function, maximum residual cosine, maximum absolute gradient, ICSF criterion, ICS criterion, ML chi-square, DF, p-value, and RMS standardized residual.

Cluster analysis using the k-means clustering method was applied to determine those subgroups of students that represent different types of PE motivation and different types of individual psychological characteristics. The k-means clustering algorithm is categorized as a partitional clustering algorithm. Partitioning existing datasets into clusters involves finding the minimum squared error between the various data points in the dataset and the mean of a cluster, then assigning each data point to the cluster centre that is nearest to it [68]. This method presupposes a pre-definition of the exact number of clusters, which makes it possible to test several initial models before deciding on the final number of clusters. The k-means method also generates non-overlapping clusters while maximizing the distance between clusters. When defining the number of clusters in a specific case, it is important that the model be practically applicable; that is, it should be clearly recognizable to teachers in practice, and it should be applicable to working with students in the future. Descriptive statistics parameters were also calculated for each cluster. Chi-square tests of association were used to determine the association between socio-demographic variables and type variables (variables that represent the student’s membership in a cluster) among students. Data for the categorical variables (groups of students) were arranged in contingency tables. Cramer’s V coefficient was calculated as a test of the effect size of the association between variables and it was used to test any significant chi-square results. Cramer’s V coefficient represents the correlation between two variables and is interpreted accordingly.

Multivariate differences between the sample types (clusters of students) were calculated by using a series of discriminant analyses. The authors of the present study employed the successive discriminant analyses claimed as part of the “hierarchical discriminant analyses” model (Milavić and Bavčević, 2024) [69]. Each new discriminant analysis in the series was used for adding to and involving a new subgroup of independent variables in the analysis. This “hierarchical” model of entry of independent variables for application in discriminant analysis was adapted from the model of hierarchical multiple regression analysis. The hierarchical model is considered to be the most flexible model as it allows the researcher to determine the order of entry of the independent variables into the equation; each independent variable is assessed as its own point in terms of the explanatory power it contributes to the equation [70]. Also, by using the hierarchical model, the researcher may determine which independent variable from a set of independent variables is the strongest predictor of the dependent variable [70]. The authors of the present study have adapted and used this kind of model when applying discriminant analyses to determine the strongest discriminant variable(s) in an equation. In addition to the usually calculated parameters for discriminant analysis (e.g., canonical R, chi-square, Wilks’ λ, and the means of canonical variables), the authors of this study calculated two additional parameters: canonical R² (the canonical coefficient of determination), and the new canonical R² change % (analysis of the canonical coefficient of determination changes in percentages between two successive discriminants). By calculating these parameters, the authors tried to assess the relative importance of the discriminant variable by considering the changes in canonical R-squared values when considering the inclusion of a particular variable into a discriminant equation already containing other discriminant variables.

3. Results

By analysing the results presented in

Table 1. Descriptive and metric parameters of the study measures (n = 267).

Variable	Mean	SD	Min	Max	Skew	Kurt	K-S Max D	Max D p	Cronbach’s Alpha
Self-esteem	3.68	0.80	1.50	5.00	0.12	−1.07	0.10	p < 0.05	0.84
Optimism	3.22	0.82	1.00	5.00	−0.14	−0.83	0.11	p < 0.01	0.71
Hope	3.79	0.82	1.17	5.00	−0.49	−0.26	0.09	p < 0.05	0.82
IM—Know	3.54	1.08	1.00	5.00	−0.67	−0.61	0.13	p < 0.01	0.86
IM—Accomplish	3.81	1.04	1.00	5.00	−0.98	0.17	0.14	p < 0.01	0.85
IM—Stimulation	3.52	1.11	1.00	5.00	−0.73	−0.42	0.13	p < 0.01	0.84
EM—Identified	2.88	1.17	1.00	5.00	0.42	−0.92	0.10	p < 0.05	0.80
EM—Ext. reg.	3.18	1.03	1.00	5.00	−0.04	−0.89	0.08	p < 0.10	0.76
PE Motivation	3.39	0.87	1.00	5.00	−0.42	−0.39	0.07	p < 0.15	0.86
Prosocial behaviour	3.69	0.75	1.30	5.00	−0.65	−0.48	0.10	p < 0.05	0.84
Social cohesion	4.09	0.86	1.00	5.00	−0.89	0.06	0.14	p < 0.01	0.89
Satisfaction with class community	3.90	1.23	1.00	5.00	−0.59	−0.59	0.25	p < 0.01	–

Notes: SD—standard deviation; Min—minimal result; Max—maximal result; Skew—skewness coefficient of the distribution; Kurt—kurtosis coefficient of the distribution; K-S max D—Kolmogorov–Smirnov goodness-of-fit test; max D p—Kolmogorov–Smirnov goodness-of-fit test significance; Cronbach’s alpha—Cronbach’s alpha reliability coefficient.

, it has been determined that all the measures used had satisfactory metric characteristics. The measures were at least satisfactorily reliable, as the values of all the reliability coefficients were above 0.70, with 9 out of a total of 11 measures having a good level of reliability [71]. Although, for 10 measures, it was determined that the distribution of the results deviated significantly from the normal distribution (the significance of the K-S max D test is p < 0.05), it is still possible to conclude that the sensitivity of all measures was satisfactory because the values of other indicators of sensitivity were within acceptable borders. Only three measures (self-esteem, hope, and prosocial behaviour) did not have a maximum possible score range from 1.00 to 5.00. The skewness and kurtosis indices of the results distribution for all measures (except the self-esteem measure) were within the range of ±1.00, which, as stated by George and Mallery (2016) [72], is considered to be an excellent result for most psychometric purposes. The self-esteem measure had a kurtosis coefficient of 1.07, which is within the range of ±2.00, which is acceptable in most cases [72].

Descriptive indicators of the individual measurements of students in the total sample varied from moderate for extrinsic motivation—identified (2.88) to very high for the measure of intrinsic motivation—to accomplish (3.81). The total value for PE motivation, the average of five elements forming the facet of motivation, was high (3.37). The distributions of the results of the individual values for students were mostly negatively skewed, and the value for extrinsic motivation was identified as positively skewed.

Descriptive indicators of the common (class community) measures of students on the total sample varied from high for prosocial behaviour (3.69) to very high for social cohesion (4.09). Student satisfaction with their class was also very high (3.90). The distributions of the results of all common (class community) student measures were negatively skewed.

Almost all the variables used were intercorrelated (

Table 2. Correlations of the study measures.

Variables	Self- Esteem	Optimism	Hope	PE Motivation	Prosocial Behaviour	Social Cohesion
Self-esteem	1.00
Optimism	0.50 ***	1.00
Hope	0.60 ***	0.29 ***	1.00
PE Motivation	0.19 **	0.11	0.41 ***	1.00
Prosocial behaviour	0.19 **	0.15 *	0.32 ***	0.44 ***	1.00
Social cohesion	0.17 **	0.16 **	0.25 ***	0.35 ***	0.54 ***	1.00

Notes: ***— The Pearson correlation coefficient is significant at the level of p < 0.001; **—the Pearson correlation coefficient is significant at the level of p < 0.01; *—the Pearson correlation coefficient is significant at the level of p < 0.05.

). These were mostly the correlations of a low to moderate level of connection. Three “individual” variables (self-esteem, along with the variables of optimism and hope) and two “community” variables (prosocial behaviour with social cohesion) were moderately related. The interrelationships between almost all these variables require further and detailed data processing.

Analysis of the constructed experimental model that describes the connection between the dimensions of individual psychological characteristics and class community was carried out using structural equation modelling (SEM). When doing so, it was necessary to determine whether the model was invariant. The measurement of invariance, itself, assesses the psychometric equivalence of a concept across groups or across time [73].

Table 3. The reflector matrix, showing the value of differences between the empirical and theoretical values of the model.

Variables	Self- Esteem	Optimism	Hope	PE Motivation	Prosocial Behaviour	Social Cohesion	Satisfaction with CC
Self-esteem	−0.000	−0.278	−0.012	0.303	0.212	0.187	−0.021
Optimism	−0.146	0.000	0.121	0.116	0.039	−0.002	0.036
Hope	−0.012	0.217	0.000	−0.244	−0.139	−0.016	0.087
PE Motivation	0.142	0.103	−0.121	−0.000	−0.354	−0.264	−0.130
Prosocial behaviour	0.020	−0.015	−0.182	−0.341	−0.000	0.009	0.018
Social cohesion	0.138	−0.010	0.035	−0.138	0.012	−0.000	−0.025
Satisfaction with CC	−0.083	0.051	0.122	0.125	0.023	−0.023	0.000

shows the reflector matrix that is used for evaluating model invariance properties. The reflector matrix represents the difference between covariance matrices, based on empirical data (I) and theoretical data (R), and it can only be constructed for explicit variables (ε = I−R) [74]. Since the diagonal elements and, consequently, the trace of the reflector matrix were extremely close to zero, it is possible to conclude that the model was invariant under changes of scale, as well as under a constant scaling factor.

The basic indicators of the constructed model’s adequacy are shown in

Table 4. Basic summary statistics.

Discrepancy Function	0.354
Maximum Residual Cosine	0.000
Maximum Absolute Gradient	0.000
ICSF Criterion	0.000
ICS Criterion	0.000
ML Chi-Square	94.083
DF	13.000
p-Value	0.000
RMS Standardized Residual	0.091

. The value of the maximum residual cosine is zero, which confirms that the process of iteration was successful. The ICSF criterion value is zero, which confirms that the structural model is invariant under a constant scaling factor, as well as the ICS criterion, which shows that the structural model is invariant under changes of scale. The chi-square value and the associated probability p-level confirm the good fit of the model. The adequacy of the model was also confirmed by the RMS standardized residual value. Based on all the above, the constructed model can be considered valid and can be used in further analyses.

Student’s t-test for independent samples was conducted on the entire sample of subjects and did not reveal gender differences, except for the value for prosocial behaviour (p < 0.01) (

Table 5. Gender differences in the study measures (n = 267).

Variable	Female Students (n = 134)		Male Students (n = 133)		t-Test	p
	Mean	SD	Mean	SD
Self-esteem	3.59	0.83	3.77	0.75	−1.83	0.07
Optimism	3.24	0.81	3.20	0.83	0.38	0.71
Hope	3.72	0.87	3.85	0.77	−1.34	0.18
IM—Know	3.54	1.05	3.53	1.11	0.11	0.91
IM—Accomplish	3.77	1.11	3.84	0.97	−0.56	0.58
IM—Stimulation	3.41	1.18	3.63	1.03	−1.62	0.11
EM—Identified	2.87	1.20	2.89	1.14	−0.16	0.87
EM—Ext. reg.	3.11	1.04	3.26	1.02	−1.21	0.23
PE Motivation	3.34	0.91	3.43	0.84	−0.85	0.40
Prosocial behaviour	3.83	0.71	3.55	0.77	3.05 **	0.002
Social cohesion	4.07	0.88	4.10	0.83	−0.28	0.78
Satisfaction with class community	3.84	1.19	3.96	1.26	−0.79	0.43

Notes: SD—standard deviation; t-test—Student’s t-test coefficient; p—the significance of the t-test coefficient; **—coefficient significance at the level of p < 0.01.

). Female students (3.83) evaluated their class communities as communities with a higher frequency of prosocial behaviour than male students (3.55). Although one significant gender difference was found in the values of this study, the finding of there being no differences between respondents of different genders allows the entire sample of respondents (including both female and male students) to be used in further statistical analyses of the data. These two subsamples of respondents of different genders, considering the measures used, thus belong to one, common, population of students.

The analysis of variance of PE motivation between groups of students with different general academic achievement levels revealed significant differences in two variables: intrinsic motivation—to know and intrinsic motivation—to accomplish (

Table 6. Analysis of variance of PE motivation by the student’s academic achievement and grade.

Variable	Academic Achievement Groups												F	p
	Good (n = 17)				Very Good (n = 92)				Excellent (n = 158)
	Mean		SD		Mean		SD		Mean		SD
IM—Know	4.22		0.86		3.46		1.26		3.51		0.96		3.79	0.02
IM—Accomplish	3.85		0.99		3.58		1.17		3.93		0.95		3.35	0.04
IM—Stimulation	3.78		1.10		3.49		1.15		3.51		1.09		0.49	0.61
EM—Identified	3.37		1.43		2.73		1.15		2.92		1.14		2.34	0.10
EM—Ext. reg.	3.59		0.88		3.10		1.04		3.19		1.04		1.61	0.20
PE Motivation	3.76		0.82		3.27		0.95		3.41		0.82		2.43	0.09
Variable	Grades												F	p
	5th Grade (n = 26)			6th Grade (n = 101)			7th Grade (n = 75)			8th Grade (n = 65)
	Mean	SD		Mean		SD	Mean	SD		Mean		SD
IM—Know	3.39	0.75		3.90		1.04	3.39	1.07		3.20		1.12	6.89	<0.001
IM—Accomplish	3.48	0.83		3.91		0.98	3.77	1.08		3.83		1.15	1.21	0.31
IM—Stimulation	3.37	0.82		3.78		1.05	3.29	1.15		3.45		1.19	3.28	0.02
EM—Identified	2.54	0.89		2.98		1.24	3.02	1.06		2.71		1.23	1.84	0.14
EM—Ext. reg.	3.07	0.92		3.42		1.00	3.08	1.12		2.98		0.98	3.07	0.03
PE Motivation	3.17	0.53		3.60		0.87	3.31	0.95		3.23		0.86	3.43	0.02

Notes: SD—standard deviation; F—analysis of variance for the F coefficient; p—significance of the F coefficient.

). After the post hoc application of the Scheffe test to these two IM variables, the following was determined: (1) the least successful students had a significantly higher intrinsic motivation—to know for PE than the remaining two groups of more academically successful students; (2) the academically most successful students had a higher intrinsic motivation—to accomplish than the group of students with very good academic success.

By analysing the variance in PE motivation between groups of students from the different grades of primary school, significant differences were found in four variables: intrinsic motivation—to know, intrinsic motivation—stimulation, and extrinsic motivation—external regulation, as well as in the overall PE Motivation variable (Table 6). However, the post hoc application of the Scheffe test on these variables accurately determined a much smaller number of differences between the groups of students. Thus, 6th-grade students were found to have a higher intrinsic motivation—to know than the two groups of students in higher grades (7th and 8th grades). Also, 6th-grade students had higher intrinsic motivation—stimulation than the group of 7th-grade students.

Univariate analysis of variance revealed partial differences between the subgroups of students. In the following analysis, the goal was to determine the existence of PE motivation types and their occurrence in the total number of students, as well as the basic motivational characteristics of each type of motivation, by applying the multivariate k-means clustering procedure to five key variables of PE motivation.

By applying the k-means clustering procedure to the five fundamental variables of PE motivation, a high-quality and easily interpretable model with four types of students with different characteristics of PE motivation was determined (

Table 7. Frequencies and descriptive parameters of the students’ PE motivation types.

Variable	PE Motivation Types								F	p
	Very Low (n = 54)		Moderate (n = 65)		Very high IM, Low EM (n = 69)		Extremely High (n = 79)
	Mean	SD	Mean	SD	Mean	SD	Mean	SD
IM—Know	2.18	0.98	3.20	0.71	4.02	0.73	4.33	0.49	109.44	<0.001
IM—Accomplish	2.20	0.70	3.71	0.63	4.37	0.53	4.49	0.50	193.99	<0.001
IM—Stimulation	2.18	0.90	2.93	0.71	4.08	0.62	4.43	0.49	150.52	<0.001
EM—Identified	1.77	0.56	3.20	0.71	1.99	0.65	4.16	0.60	213.19	<0.001
EM—Ext. reg.	1.91	0.58	2.99	0.64	3.16	0.76	4.23	0.56	145.06	<0.001
PE Motivation	2.05	0.45	3.21	0.30	3.53	0.34	4.33	0.33	-	-

Notes: SD—standard deviation; F—analysis of variance of the F coefficient; p—significance of the F coefficient.

). The first PE motivation type, named the very low PE motivation type, contained 20.2% of the student sample. The second type, called the moderate PE motivation type, contained 24.3% of the student sample. The third type, named the very high IM, low EM motivation type, comprised 25.8% of the student sample. The fourth type, called the extremely high PE motivation type, contained 29.6% of the student sample. The motivational characteristics of these PE motivation types are explained in detail in the Discussion Section and Figure 1 clearly shows the characteristics of the four types of students with their different PE motivations.

Regarding the four types of students with different PE motivations, the chi-square test of association was applied in order to determine possible connections between the PE motivation types and the collected socio-demographic data (

Table 8. Associations of PE motivation types with the students’ socio-demographic characteristics.

Variable	Group	PE Motivation Types				Test of Association
		Very Low (n = 54)	Moderate (n = 65)	Very High IM, Low EM (n = 69)	Extremely High (n = 79)
		%	%	%	%
Age	11–12 years	16.7	20.0	15.9	15.2	Chi-square	7.56
	13 years	33.3	27.7	43.5	40.5	(df)	(9)
	14 years	27.8	36.9	21.7	30.4	Cramer’s V	0.10
	15–16 years	22.3	15.4	18.8	13.9	p	0.58
Grade	5th class	9.3	18.5	8.7	3.8	Chi-square	28.59
	6th class	27.8	23.1	47.8	48.1	(df)	(9)
	7th class	33.3	36.9	13.0	30.4	Cramer’s V	0.19
	8th class	29.6	21.5	30.4	17.7	p	0.001
Academic achievement	Good	5.6	4.6	5.8	8.9	Chi-square (df) Cramer’s V p	8.34 (6) 0.13 0.22
	Very good	44.4	24.6	40.6	30.4
	Excellent	50.0	70.8	53.6	60.8
Gender	Females	50.0	55.4	49.3	46.8	Chi-square	1.08
						(df)	(3)
	Males	50.0	44.6	50.7	53.2	Cramer’s V	0.06
						p	0.78
Sport now	Yes	14.8	12.3	18.8	25.3	Chi-square	4.60
						(df)	(3)
	No	85.2	87.7	81.2	74.7	Cramer’s V	0.13
						p	0.20

Notes: Chi-square—Chi-square test of the association coefficient; df—degrees of freedom; Cramer’s V—Cramer’s V effect size coefficient of the chi-square test; p—significance of the chi-square and Cramer’s V coefficients.

).

The association test showed that the variables of age, academic achievement, gender, and sport now are not related to the types of PE motivation. The only variable related to different PE motivation types is the variable grade. Students in lower grades (5th and 6th grades) more often have the characteristics of higher types of PE motivation, while students in higher grades (7th and 8th grades) more often have the characteristics of lower degrees of PE motivation. Although a significant connection between these two variables was established, the size of the Cramer’s V effect size coefficient confirms that the degree of this connection is nevertheless very low.

By applying a series of three discriminant analyses with the variable satisfaction with class community as a dependent variable, three subsets of variables were gradually introduced: the first set of variables, socio-demographic variables; the second set of variables, individual psychological variables (self-esteem, optimism, hope and 5 “elements” of the PE motivation); and the third set of variables, two common social variables (prosocial behaviour and social cohesion) (

Table 9. Hierarchical discriminant analysis of the satisfaction with class community groups.

Chi-Square Tests, with the Successive Roots Removed
DF	Eigenvalue	Canonical R		Wilks’ λ		Chi-Square	DF		p
Model 1	0.06	0.24		0.92		21.44	10		0.018
Model 2	0.18	0.39		0.81		52.92	26		0.001
Model 3	0.82	0.67		0.52		168.53	30		<0.001
Factor Structure Matrix
Variable	Model 1		Model 2		Model 3			Tolerance
	Canonical root		Canonical root		Canonical root
Age	−0.69		−0.46		−0.21			0.18
Grade	−0.73		−0.45		−0.20			0.18
Academ. ach.	−0.37		−0.12		−0.01			0.85
Gender	0.40		0.19		0.06			0.87
Sport now	−0.25		−0.08		−0.01			0.90
Self-esteem	–		0.65		0.31			0.50
Optimism	–		0.39		0.18			0.71
Hope	–		0.46		0.23			0.51
IM—Know	–		0.63		0.27			0.42
IM—Accom.	–		0.44		0.22			0.36
IM—Stimul.	–		0.50		0.23			0.40
EM—Identif.	–		0.17		0.08			0.60
EM—Ext. reg.	–		0.46		0.21			0.37
Prosocial behaviour	–		–		0.61			0.54
Social cohesion	–		–		0.82			0.63
Means of Canonical Variables
Satisfaction with Class Community	Model 1		Model 2		Model 3
Low	−0.24		−0.73		−1.89
Moderate	−0.21		−0.19		−0.10
High	0.28		0.44		0.77
Canonical R2	0.06		0.15		0.45
Canonical R2 change %	5.9%		9.5%		29.6%

Notes: Eigenvalue—the eigenvalue of the discriminant function; Canonical R—the canonical correlation coefficient; Wilks’ λ—Wilks’ lambda coefficient; Chi-Square—the chi-square coefficient of significance of the discriminant function; df—degrees of freedom; p—level of the coefficient of significance of the discriminant function; Canonical R²—the canonical coefficient of determination; Canonical R² change %—the canonical coefficient of determination’s percentage change.

). All students were divided into one of three groups, according to their assessment of satisfaction with their class community: the low group contained those who are not at all satisfied and those who were barely satisfied with their class community (n = 41); the moderate group contained those who were moderately or quite satisfied with their class community (n = 111); finally, the high group included those who were completely satisfied with their class community (n = 115).

In each of the three successive discriminant analyses, one significant discriminant function (the canonical root) was determined; all three significant discriminant functions are shown in Table 9 as Model 1, Model 2, and Model 3.

In Model 1, which includes the students’ socio-demographic variables, the only significant discriminant function separated the group whose assessment of satisfaction with class community was expressed as high from the other two groups (which were located very close to each other), where the canonical root explains a total of only 5.9% of the variance. When analysing the structure matrix, that is, the correlation coefficients of the variables and the discriminant function, the largest partial contribution to the discrimination model was observed for the variables Age and Grade.

In Model 2, which includes socio-demographic variables and individual psychological variables, the only significant discriminant function separated the group whose assessment of satisfaction with class community was expressed as high compared to the other two groups, where a total of 15.4% of the variance was explained by the canonical root. The relative contribution of individual psychological variables amounted to 9.5% and was much higher than the relative contribution of socio-demographic variables. The analysis of the correlation coefficients of individual variables with the isolated discriminant function showed high projections for the variables self-esteem and IM-know.

In Model 3, which, in addition to socio-demographic variables and individual psychological variables, also includes two variables of common characteristics, the only significant discriminant function divided the group, with an assessment of satisfaction with class community being expressed as high compared with the other two groups, with the canonical root explaining a total of 45% of the dependent variable variance. The relative contribution of these two common variables amounted to 29.6% and was much, much higher than the relative contribution of socio-demographic variables and individual psychological variables. This was confirmed by the high correlation coefficients of the pro-social behaviour and social community variables with the isolated discriminant function. The multicollinearity among independent discriminant variables was not detected because all tolerance coefficients were above 0.10.

Table 10. Analysis of variance of the study measures between the satisfaction with class community groups.

Variable	Satisfaction with Class Community Groups						F	p
	Low (n = 41)		Moderate (n = 111)		High (n = 115)
	Mean	SD	Mean	SD	Mean	SD
Age	13.83	1.05	13.52	1.05	13.27	0.92	5.13	0.006
Grade	6.95	0.92	6.77	0.95	6.48	0.93	4.82	0.009
Academ. ach.	4.46	0.60	4.62	0.59	4.46	0.64	2.22	0.11
Gender	0.49	0.51	0.44	0.50	0.56	0.50	1.51	0.22
Sport now	0.90	0.30	0.84	0.37	0.77	0.43	2.20	0.11
Self-esteem	3.26	0.88	3.63	0.79	3.88	0.71	10.30	<0.001
Optimism	2.97	0.77	3.18	0.80	3.36	0.83	3.74	0.025
Hope	3.43	0.86	3.78	0.82	3.92	0.78	5.66	0.004
IM—Know	3.12	1.18	3.37	0.98	3.85	1.06	9.74	<0.001
IM—Accomplish	3.39	1.29	3.78	0.99	3.98	0.95	5.06	0.007
IM—Stimulation	3.09	1.25	3.44	1.04	3.75	1.08	6.02	0.003
EM—Identified	2.73	1.20	2.84	1.10	2.97	1.21	0.72	0.49
EM—Ext. reg.	2.82	1.19	3.11	0.96	3.38	1.00	5.14	0.006
PE Motivation	3.03	1.02	3.31	0.82	3.59	0.82	7.15	0.001
Prosocial behaviour	2.93	0.90	3.64	0.64	4.00	0.57	40.44	<0.001
Social cohesion	2.97	1.06	4.09	0.65	4.48	0.54	73.62	<0.001

Notes: SD—standard deviation; F—analysis of variance of the F coefficient; p—significance of the F coefficient.

shows the average values of all discriminant variables for three groups of students with different degrees of satisfaction with class community.

It is noticeable that with an increase in the level of satisfaction with their class community, the values of socio-demographic variables “declined”, while the values of students’ individual psychological characteristics “increased”. Of course, the most pronounced differences were in the assessment of common class characteristics (prosocial behaviours and social cohesion).

Since it was not possible to precisely determine the relationship between a whole set of the used variables and the variable of satisfaction with the class community from an overview of Table 10, students were then clustered according to their psychological characteristics (

Table 11. Frequencies and descriptive parameters of the psychological characteristic types.

Variable	Psychological Characteristic Types						F	p
	Low (n = 56)		Moderate (n = 110)		High (n = 101)
	Mean	SD	Mean	SD	Mean	SD
Self-esteem	3.52	0.76	3.46	0.80	4.01	0.71	15.77	<0.001
Optimism	3.17	0.91	3.07	0.69	3.42	0.86	5.19	0.006
Hope	3.33	0.73	3.56	0.88	4.29	0.47	41.99	<0.001
IM—Know	2.10	0.93	3.55	0.72	4.32	0.53	177.73	<0.001
IM—Accomplish	2.29	0.76	3.93	0.68	4.51	0.50	223.62	<0.001
IM—Stimulation	2.08	0.83	3.48	0.80	4.37	0.54	182.12	<0.001
EM—Identified	1.96	0.78	2.51	0.90	3.80	0.95	92.70	<0.001
EM—Ext. reg.	1.96	0.60	2.98	0.70	4.08	0.64	196.20	<0.001
PE Motivation	3.11	0.75	3.60	0.71	4.10	0.53	43.01	<0.001
Prosocial behaviour	3.49	0.98	4.09	0.78	4.41	0.66	24.52	<0.001
Social cohesion	3.52	0.76	3.46	0.80	4.01	0.71	15.77	<0.001

Notes: SD—standard deviation; F—analysis of variance of the F coefficient; p—significance of the F coefficient.

), and the relationship between these types of psychological characteristics and the level of their satisfaction with class community was examined (

Table 12. Association of satisfaction with class community groups and psychological types.

Students		Satisfaction with Class Community Groups				Test of Association
		Low	Moderate	High	All
Psychological Characteristics Types	Low	15	28	13	56	Chi-square	22.98
	Moderate	19	48	43	110	(df)	(4)
	High	7	35	59	101	Cramer’s V	0.21
	All	41	111	115	267	p	<0.001

Notes: Chi-square—Chi-square test of the association coefficient; df—degrees of freedom; Cramer’s V—Cramer’s V effect size coefficient of the chi-square test; p—significance of the chi-square and Cramer’s V coefficients.

). This approach enabled a clear analysis of the connection between different types of psychological characteristics and the defined groups of students expressing different degrees of dissatisfaction with the class community.

By applying the k-means clustering procedure to the variables of the individual and collective psychological characteristics of students, a high-quality and easily interpretable model for the three types of students with different psychological characteristics was established. The first type, called the low type, comprised 21% of the student sample. The second type, called the moderate type, included 41.2% of the student sample. The third type, called the high type, comprised 37.8% of the student sample. The psychological characteristics of these types are explained in detail in the Discussion in Section 4 and Figure 2 clearly shows the characteristics of the three types of students.

The association test determined that the variable psychological characteristics type was significantly related to the variable satisfaction with the class community. Students with a higher degree of satisfaction more often showed the characteristics of higher types of psychological characteristics. Although a significant connection between these two variables was established, the size of the Cramer’s V effect size coefficient confirmed that the degree of this connection was still relatively low.

4. Discussion

Several major findings have resulted from this study: (1) the main positive resources of satisfaction with the class community were social cohesion and prosocial behaviours within the class community; (2) the individual psychological characteristics of students (especially self-esteem and hope) and their socio-demographic characteristics also had a significant impact on their satisfaction with the class community, but their relative contribution was much less pronounced than the aforementioned social relations in the class community; (3) the types of students’ psychological characteristics were related to the level of satisfaction with the class community, but this connection was at a low level; (4) more than half of the students (those with low and medium satisfaction combined) were not satisfied with their class community; (5) the total distribution of the four types of students with different characteristics of PE motivation was determined: these were the very low; moderate; very high IM, low EM, and extremely high PE motivation types. These findings require more precise and detailed interpretation and, thus, will be further analysed and explained.

4.1. Measures Expression and Gender Differences

When reading the interpretations, it is recommended to always take into account that the sample of respondents comes from an urban city environment from the Mediterranean region, a city with a size of approximately 250,000 inhabitants. In the sample of respondents, that is, students from the upper grades of primary school, the expression of self-esteem and hope measures was high. A large number of students rated their self-esteem as very high or high, with much fewer rating it as low or very low. The situation was similar for the variable hope. The expressiveness of the value for optimism was medium. The authors of this study believe that it is possible that the concept of optimism may have been unclear to students, as difficulties were found during the initial validation of this scale. The same issue can be found in the works of other authors, who expressed potential difficulties in studying the observed variables [75,76,77,78].

The expressions of PE motivation measures were measured as medium, high, or very high (IM-Accomplish). All three variables of intrinsic motivation were significantly more pronounced than the two measures of extrinsic motivation. When this finding is added to the fact that in the sample of examinees, approximately four-fifths of the students were engaged in extracurricular sports activities that the city provided, then it is possible to conclude that their PE motivation was good regarding their engagement in classes. However, a smaller number of students showed a generally low level of interest in PE classes. These findings are in line with the findings of other authors, who also found high scores for this variable [11,61,62,63].

The expressions of common measures used to assess class community were high (prosocial behaviour) or very high (satisfaction with class community and social cohesion). In all three of these variables, the number of very high estimates was high. This shows that the majority of respondents evaluated their own classes as highly positive communities. Most respondents in a community felt that it fulfilled and satisfied their social needs and contacts. However, a smaller number of students did not share similar views. From this type of research, it is not possible to determine the reasons for such stated negative attitudes in a small sector of the students, but, in the process of further processing the results, an attempt was made to determine the individual socio-demographic and psychological characteristics of that sector of respondents. This is in line with the previous findings of other authors [79].

Most of the measures used were correlated as low to medium. Three “individual” variables (self-esteem, along with the variables of optimism and hope) and two “common” variables (prosocial behaviour with social cohesion) were moderately related. Common variables were used to assess very closely related concepts (prosocial behaviour and social cohesion), so this connection was to be expected. It is a pity that the prosocial behaviour construct, which, according to the authors of the scale, was made up of three areas of behaviour, did not provide the possibility for precise quantification of these areas, so there was no possibility of determining which group of behaviours had contributed the most strongly to the development of social unity. Task cohesion was not measured by the authors in this study because they assumed that students’ engagement in the class community was co-active and parallel, but not joint, because there were no permanent school activities in which the evaluation of a group of students (and not of each student separately) was carried out. It is possible that in the future, the introduction of such teaching content should be considered, which would facilitate the mutual joint activities of students, and would thus promote the building of social cohesion within the class community.

The relation of three “individual” variables (self-esteem, along with the variables of optimism and hope) showed that it is possible that these different constructs were based on some kind of general and common mechanism or that they developed in parallel. In future research, it is recommended to precisely determine the relationships between these constructs, as well as their specific role in and contribution to the different developmental activities of students.

PE motivation is most closely related to measures of prosocial behaviour, hope, and social cohesion, which shows that PE motivation is higher among those students who belong to class communities with better social relationships. In the sports and exercise environment, it has been known that when in groups with a better social climate, exercise is more pleasant for the group members, but it is also more successful in terms of performance/competition. A quality social climate facilitates the effectiveness and satisfaction of group members in exercise. Finally, one of the principal tasks of PE teachers is to create a task-oriented or mastery-oriented motivational climate, in which all members (regardless of whether they have a low or high level of kinesiology competence) will feel accepted and comfortable. These findings are consistent with those of previous studies that emphasize the role of teachers in generating a positive atmosphere in PE classes, as well as their influence on the positive outcomes of the educational process [56,57,67].

In this study, no gender differences were found, except for the prosocial behaviour measure. This is in line with the previous findings of other authors [80,81,82]. It is to be assumed that female students make more social contacts between members of their gender; it is to be expected that they include their more frequent prosocial experiences in their assessment of the entire class community. Therefore, it is possible and justifiable to conclude that a group of female students, in contrast to a group of male students, uses a different frame of reference for evaluating prosocial behaviours within their class.

4.2. PE Motivation Differences and Types

Significant PE motivation differences were found in relation to the degree of general academic achievement in the two variables of intrinsic motivation, to know and to accomplish. The least successful students had significantly higher intrinsic motivation to know for PE, while the most academically successful students had a higher intrinsic motivation to accomplish compared to the group of students who achieved very good academic success. This finding somewhat negates (rejects) the assumption that PE motivation is similar to the general academic motivation of students because students with the lowest academic success are intrinsically more motivated to learn new content (intrinsic motivation to know) in PE classes. In contrast, what characterizes the most academically successful students is their intrinsic motivation to accomplish the activities they have started, regardless of the possible problems that they may face. To persevere in work and to overcome any adversity in the process of acquiring knowledge is always a characteristic of those who are more successful. Also, significant PE motivation differences were determined by different grades in four variables: intrinsic motivation to know, stimulation, extrinsic motivation—external regulation, and overall PE motivation. In contrast, using the post hoc test, a very small number of differences were determined. Sixth-grade students had the most strongly expressed intrinsic motivation to know and stimulation. It is possible that this is precisely the period that students enjoy the most and when they are most open to acquiring new knowledge in PE activities [63]. These univariate data processing procedures did not allow for making quality conclusions regarding students’ PE motivation. This study’s findings are in line with previous research by other authors [83,84,85].

By applying the multivariate data analysis procedure, four types of students with different characteristics of PE motivation were determined: the very low, moderate, very high IM, low EM, and extremely high PE motivation types. Each of these types was made up of approximately 20% to 30% of students. The very low PE motivation type was characterized by very low values for all measures of PE motivation. The moderate PE motivation type was characterized mostly by moderate values for PE motivation and high IM to accomplish scores. The very high IM, low EM PE motivation type was characterized by very high values for intrinsic motivation and a very low value for EM—identified, as well as a moderately expressed value for EM—external regulation. The extremely high PE motivation type was characterized by three extremely high and two very high values for PE motivation. These types of PE motivation showed completely different motivational characteristics.

In relation to the socio-demographic characteristics of the students, the correlation of PE types was determined only by the variable of grades. However, this correlation was at a low level, where students of the higher grades (7th and 8th grades) of primary school more often belonged to the lower PE motivation types of students. It has been shown that the kinesiological activity of adolescents decreases as they get older. The authors of this study are of the opinion that there are possibly several different reasons for this finding. First, it is possible that there is a reduction in general interest in exercise—kinesiological activity has a negative effect on motivation for PE classes. Second, as they move farther into middle adolescence, many young people intensify their wide and frequent social contacts (whether through live contact or online contacts) and, at the same time, expand their interests, which changes relationships among the motivational “forces” that they invest in some other activities, such as PE classes. Third, early selection often occurs in sports clubs, whereby some young people, especially those with lower general kinesiology or specific sports competencies, are rejected from further organized extracurricular activities, after which they often “reject” everything that is even remotely related to exercise; this may apply to PE classes too. In the end, PE motivation is not related to the variables of age, academic achievement, or gender, nor to the students’ current participation in organized sports.

4.3. Determinants of Satisfaction with Class Community

Approximately 15% of students were not at all or were barely satisfied with their class community, while approximately 43% of them were completely satisfied. The rest were moderately satisfied with their class. Therefore, approximately one-sixth of the students did not consider their class community a social community in which they could achieve and develop in a quality way, and in which they would want to remain. About two-fifths of the students had “serious” objections to their class community. Together, they represented more than half of the students in the study. This finding on the frequency of satisfaction with the class community is extremely important because it indicates the need for all participants in the educational process in primary schools to devote much more attention and personal involvement to addressing this “problem”.

The authors of this study, by applying a series of discriminant analyses, have tried to determine more precisely the possible variables associated with the students’ degree of satisfaction with the class community. Firstly, socio-demographic variables discriminate between these three groups of students, but their discriminatory value (contribution) is low. It is evident that slightly older students and students in higher grades are more often members of a group that estimates their level of satisfaction as low. Secondly, when taken together, the individual psychological and socio-demographic variables represented three groups of students with different levels of satisfaction. The group giving a high assessment for their satisfaction with class community was characterized by highly pronounced self-esteem and IM to know variables, along with slightly more pronounced IM stimulation, EM external regulation, and IM to accomplish variables. In contrast, the low group was characterized by slightly more pronounced age and grade variables. Those students who had high (or higher) values for self-esteem and who showed high intrinsic PE motivation evaluated their class community with a higher degree of personal satisfaction. They were less “bothered” by some kinds of social relations or circumstances than the members of the other two groups who had a moderate or low level of satisfaction. The relative contribution of individual psychological variables was much higher than the relative contribution of socio-demographic variables in differentiating these three groups. Thirdly, when combined together, common variables (prosocial behaviour and social cohesion), individual psychological, and socio-demographic variables best represented the three groups of students with different levels of satisfaction. Those students who evaluated their class community with a high degree of personal satisfaction were characterized by very high evaluations for prosocial behaviour and social cohesion in the class community. They were also characterized by slightly higher levels for some individual psychological variables: self-esteem, PE intrinsic motivation to know, stimulation, and to accomplish, as well as hope. The relative contribution of these two common variables (social cohesion and prosocial behaviour) was extremely high, sometimes even twice as high as the joint relative contribution of socio-demographic variables and individual psychological variables.

With the aim of further and more precisely revealing the relationship between all discriminant variables and the degree of satisfaction, the average results of all measures were compared and three phenomena were observed: (1) the tendency of all socio-demographic measures to decrease with an increase in the degree of satisfaction with the class community; (2) the tendency of all individual psychological measures to increase parallel with increasing the degree of satisfaction with the class community; (3) the tendency of common measures (social cohesion and prosocial behaviour) to increase in parallel with increasing the degree of satisfaction with the class community.

In order to better determine the relationships between individual psychological and common variables, along with the variable of students’ degree of satisfaction with their class community, three types of students with different psychological characteristics were determined. The low psychological characteristics type was characterized by very low values for PE motivation, medium values for psychological characteristics (self-esteem, optimism and hope) and high common measures (social cohesion and prosocial behaviour). The moderate psychological characteristics type was characterized by high values for intrinsic motivation for PE (except for the variable EM—identified) and high values for psychological characteristics (self-esteem and hope), along with a very high value for prosocial behaviour and a high value for social cohesion. The high psychological characteristics type was characterized by extremely high values for intrinsic motivation for PE, with very high values for extrinsic motivation for PE, very high values for psychological characteristics (self-esteem and hope), extremely high values for prosocial behaviour, and very high values for social cohesion. Approximately one-fifth of the students belonged to the low type, and two-fifths of the students each belonged to the moderate and high psychological characteristics types. The only measure that did not completely separate these types of students was the optimism measure, where only the high type differed from the other two types of students.

The types of students’ psychological characteristics were related to the variable of satisfaction with the class community. Students with a higher degree of satisfaction with the class community more often showed higher types of psychological characteristics, but the degree of this connection was still low.

Based on these findings, it is possible to conclude that the main positive resources of satisfaction with one’s class community were social cohesion and prosocial behaviours within the class community. Although the individual psychological characteristics of students and their socio-demographic variables had a definite and also significant role, they were still much less “strong” than social relationships in the class community. Therefore, if someone were to think about how to improve students’ satisfaction with their class community, they should primarily think about improving social relationships within the class community, and, in such a way, try strongly to develop social cohesion. In the sports environment, especially in team sports, and also in the educational environment, there are many ways in which social cohesion can be developed. Also, within class communities, educators should look for ways to additionally develop certain individual psychological characteristics, such as self-esteem and hope. In the field of positive psychology, one can find such effective interventions being aimed at individuals or groups.

4.4. Limitations and Future Research Directions

The main limitations of this study are the relatively small number of respondents in the sample, as well as the small number of socio-demographic and sport-related variables used in the study. Also, the study was conducted as cross-sectional research, which, itself, carries limitations in terms of observing developmental changes in students. Guidelines for future research, in accordance with the limitations of the study, would involve conducting similar future studies with an increase in the number of respondents, and the possible distribution of the sample to secondary schools as well. It is also necessary to consider the possibility of including more socio-demographic, academic, and kinesiology/sports (physical) variables related to the activity, as well as other psychological variables. It is especially important to think about increasing the number of class community variables and including variables such as direct or indirect aggressiveness in the class community and the leadership or teaching style of the teacher. It would be especially useful to conduct a longitudinal study, which would possibly allow us to understand the reasons for the findings of this study. The connection between social cohesion and prosocial behaviour with the talent development process also emphasizes the importance of considering biological age, particularly in the context of motivation and social interactions within school communities. In future research, it is also recommended that researchers develop a structural equation model that describes the connection between students’ satisfaction according to class community and the psycho-social characteristics of students.

5. Conclusions

Satisfaction with one’s class community is possibly one of the most important “variables” regarding the effectiveness of the entire educational process, perhaps for two different reasons: first, it is an extremely important evaluative outcome and the consequence of students’ actions in a mandatory social community; second, since, in the sports environment, it has been established that the quality of social relations (primarily cohesion) within the team as a social community is positively and circularly related to the quality of the group’s performance, i.e., that the quality of social relations facilitates a highly efficient performance, it is possible that social relations could have a similar or the same effect in the class community within the educational environment (facilitating the effectiveness of the teaching process). The correctness and/or the level of accuracy of this position should be determined or rejected in future research.

In this study, it was determined that only approximately two-fifths of students in the upper grades of primary school were completely satisfied with their class community. Approximately 15% of students were not at all or were barely satisfied with their class community. The main determinants and main positive resources of satisfaction with the class community of students were social relations, namely, the variables of social cohesion and prosocial relations. Individual psychological characteristics, such as self-esteem, intrinsic PE motivation, and hope also positively influenced the degree of satisfaction with the class community, but with a much lower relative contribution than the variables of social relations. Socio-demographic characteristics (the variables of age and grade) indicated a negative tendency, and, with an increase in age (which also means an increase in variable grade), the degree of satisfaction with the class community decreased. Since it is not possible to directly influence socio-demographic variables, this information can only be shaped as a determinant of the period in which interventions could or should be undertaken, with the aim of improving social relations in the class community and positively contributing to the development of certain psychological characteristics in students (e.g., self-esteem, intrinsic motivation, or hope). Therefore, the period in which it is recommended to carry out such interventions is in the 5th and 6th grades of primary school. Primary school fulfils the educational process, so the stated goals and possible interventions are precisely the goals of the educational process itself. The obtained findings of this study can be used in immediate kinesiology practice. An insight into the structure of the class community, as well as the psycho-social structure of the class participants, can provide guidance in planning and programming PE teaching content for the purpose of intensifying physical exercise. In addition to the above, teachers can influence the individual segments mentioned in the study through quality communication with students, with the aim of creating a healthy social community and ensuring the well-being of all students within the class community.