Pupils’ Experience of Social Participation in Finnish Primary Schools

Social participation means taking a full and active role in school life, being a valued and integral member of the school community. The purpose of this study was to examine pupils’ experiences of social participation and their active roles in Finnish primary schools. The intention was to find out in what ways primary school pupils’ (n = 3,760) personal experiences of pupil–pupil (PP) and pupil–teacher (PT) interaction vary between genders and across age groups in small and large schools. Pupils’ experiences were examined with questions concerning the active participation of the child and the participatory role of the teacher. The results show that pupils’ experiences vary, and there are differences between schools in terms of pupils’ social participation.


Introduction
This study focuses on pupils' views on social participation in their classrooms. Participation is one of the key values found in international agreements on education, in addition to equality and shared services for all (United Nations [UN], 1993[UN], , 2006. The Children's rights also emphasize the right to participate and have an influential voice in society (UN, 1989). Participation is a crucial element of the current educational reforms, which promote the idea of "school for all" (see Ainscow & Miles, 2009;Booth, 2011;Bossaert, Colpin, Pijl & Petry, 2013). Despite the goodwill behind "school for all" in education policy, good intentions do not automatically lead to social participation, i.e. positive interaction, friendship, and acceptance (see Koster, Nakken, Pijl & van Houten, 2009). One of the goals of the Finnish education policy has been that all learners in comprehensive schools are schooled together (Takala, Pirttimaa & Törmänen, 2009, p. 162), in order to attain the full experience of participation and equality for all children. In Finland this experience of participation has been overshadowed by an intense focus on learning outcomes, according to Naukkarinen (2010). Even though "a fully participatory" school is hardly ever actualized (Florian, Black-Hawkins, & Rouse, 2017), attempts have been made across the world to provide opportunities for participation for all learners in inclusive schools (Black-Hawkins, 2013). The social element is essential to the concept of participation. The Council of Europe defined the social element in this way: "to be heard and to contribute to decision making on matters affecting them" (Committee of Ministers, 2012, p. 3).
Previous studies on social participation have largely focused on pupils with special needs in inclusive schools (see Koster et al., 2009) and therefore, there is a lack of research on a wider definition of inclusion comprising all pupils in shared classrooms (see Black-Hawkins, 2013;Mittler, 200;UNESCO, 2001). Koster et al. (2009) found in a review study that the social dimension of inclusion has been described by using the concepts of social integration, social inclusion, and social participation almost synonymously and interchangeably. They ended up recommending the concept "social participation" when describing the social dimension of inclusion (Koster et al., 2009).
The focus of this article is on Finnish pupils' views of their social participation in primary education. In this study, social participation is seen as a phenomenon encompassing pupils' experience of being heard and being able to affect their own learning.
At the same time, it includes pupils' interaction with both their classmates (pupilpupil, PP) and the teacher (pupil-teacher, PT). Differences between pupils in different age groups and genders and in small and large schools are considered.

Social participation
The concept of participation includes social elements, such as pupils' experience of being valued members of the school community (Booth, 2002;Farrell, 2000). Dyson, Howes, and Roberts (2002) first discussed the idea of pupils' decision-making and pupils' access to the full curriculum. Martin Rouse (2016) and Kristine Black-Hawkins (2013) examined overlapping levels of participation in different ways, and discovered social dimensions, such as the sense of belonging and the feeling of acceptance by other children and teachers. At the level of "participation and collaboration," pupils share learning experiences and spend time together (Black-Hawkins, 2013;Rouse, 2016). Thus, participation means not only shared classrooms, but also genuine interaction, shared experiences, and being seen and heard in one's community. Koster et al. (2009) stated that a key theme of social participation is contacts and interactions that can be divided into aspects such as playing together, working together on tasks, participating in group activities, acknowledged initiations, and expressing pro-sociality.
In daily interactions in schools, teachers and pupils are the two essential actor groups for the successful social participation of pupils. Consequently, social participation stipulates more than merely the presence of a pupil in a shared classroom (Booth & Ainscow, 2011;Florian et al., 2017). This calls for two elements: the participating self, and the participatory activity or action. A crucial factor in this process is equal dialogue between actors (Booth & Ainscow, 2011, 2016. According to Black-Hawkins (2013), "Participation is based on relationships of mutual recognition and acceptance" (p. 396). In a school environment, this consists of a collaborative and balanced relationship between children and status-free dialogue between children and adults, which includes being helped, seen and heard by one's teacher (Backman et al., 2012). The significance of classmates and adults in the school environment changes as pupils get older as the influence of peer relations tends to increase and relationships with teachers become less important to pupils (Bukowski & Adams, 2005). Dyson et al. (2002) stated that pedagogical approaches which encourage pupils to learn together and interact more freely with their peers in learning activities increase pupils' interaction and social participation. Rogoff (2007) emphasized that learning to collaborate is essential in the process of developing participation. Florian and her colleagues (2017) added that students should have an active role in their classroom.
This means cooperating and engaging in open discussions with classmates and taking active responsibility for one's studies (Backman et al., 2012). By doing so, pupils get the experience of contributing and belonging. Together, they form the skills that social participation requires and which can be learned in a school classroom (Black-Hawkins, 2013). Teachers can positively affect the learning of these crucial skills, for example, by supporting the interaction between classmates, by endorsing pupils' cooperation, and by providing mutual help during lessons.
Previous studies show that for pupils the most important place in terms of social participation is the classroom, where negotiations and daily activities take place (Hulme, McKinnely, Hall & Cross, 2011). That is why the daily classroom environment is a relevant place to study pupils' social participation by examining pupils' experience of working together on tasks (Koster et al., 2009).
A teacher's role is to support his or her pupils' participation and involvement in a school environment (Vinterek, 2010). Relationships between pupils and teachers depend on the culture of a school, which rests on the values and beliefs of the staff (Florian & al., 2017). At school, good relationships among pupils and between pupils and teachers support social participation (Black-Hawkins, 2013). According to Mazurkiewicz (2013), learning ought to be organized in such a way that pupils are encouraged to openly interact with their teachers, and pupils should be allowed to ask questions and talk about their ideas freely. By controlling the classroom talk, a teacher can either promote or hinder pupils' ownership of their learning and agency (Kovalainen & Kumpulainen, 2007). By strongly controlling pupils' possibilities to make decisions, a teacher hinders this ownership (Hulme et al., 2011).
During childhood, girls and boys face different kinds of cultural expectations (Gillander Gådin, Weiner & Ahlgren, 2013;Valaitis, 2002), which affect how much space the child is given in a classroom. For example, teachers' expectations for girls' behavior are much stricter than for boys (Valaitis, 2002). Thus, it is possible that the gender of a pupil can have an effect on the experience of social participation in school. Backman et al. (2012) reported that a teacher's appreciation and individual feedback are crucial to pupils. At the same time, being given the opportunity to choose the tasks and methods used during lessons is also significant to pupils. Behind genuine social participation and taking part in the decision making, is pupils' sense of safety (Mazurkiewicz, 2013).
As mentioned before, the active role of a teacher is central when considering barriers and obstacles to pupils' participation. A teacher can either support or prevent pupils' social participation. Barriers to participation are interconnected with the process of participation (Black-Hawkins, 2013). Florian et al. (2017) stated that "increasing participation reduces barriers to participation and vice versa." The processes behind changes in pupils' social participation are not always easy to identify. That is why there should be awareness that teachers' actions can increase opportunities to participate by some pupils while reducing opportunities for others (Florian et al., 2017, p. 50). Teachers' well-meaning actions can have various effects on pupils, and individual experiences can vary widely. Some pupils can be excluded from social participation because of their teacher's vision of learning and teaching (Booth & Ainscow, 2011).

Macartney and Morton (2013) stated:
A central aspect of our role as a teacher must be to develop an awareness of the lived effects of our meanings and practices so that we are in a position to recognize and remove barriers to the learning and participation of every child and family. (p. 787) Studies have yet to address several other aspects of social participation such as, the experience of being heard, of being a genuine subject in terms of one's own learning and of being in a good relationship with classmates. In addition, teachers' support of pupils' social participation has not being studied. In particular, the aspect of gender and the effect of school size are missing from studies of social participation. The voice of children is valuable and can give us a better understanding of this multifaceted matter. That is why in this study the experts on social participation consulted are the learners themselves. The purpose of this article is to examine what kind of views Finnish primary school pupils have on their social participation. In this study, social participation is seen as a phenomenon that includes pupils' interaction with their classmates (PP) and the teacher (PT). Social participation requires pupils' experience of being heard and of being able to affect learning. In addition, differences between genders and pupils in different age groups and school size are considered. In this study we examine the following research questions: • How do pupils experience social participation in primary schools in general?

•
Are there any differences in pupils' experience of social participation between genders and age groups?
• Is the experience of social participation connected to school size?

Procedure
The study was conducted in spring 2011 in primary schools in one of the biggest cities in Finland. Permission to conduct the research was granted by the head of the city's basic education unit. She also sent via e-mail an introduction letter to the principals of all the primary schools to encourage them to participate in the study. Principals were advised to forward the introduction and instruction e-mail to teachers. The data were collected via an Internet-based Webropol questionnaire available through a link on the research project webpage. Because the researchers were not present when the questionnaire was filled out, careful instructions for the pupils and teachers who filled out the forms were crucial. Pupils filled out the questionnaire independently on their devices in a classroom and answered privately and anonymously without others seeing their answers. Teachers were given careful instructions in order to try to ensure the procedure was the same for all respondents. The teachers were told to emphasize confidentiality when informing pupils about how to complete the questionnaire. In order to avoid misunderstandings and the influence of incipient literacy skills, first and second graders were provided with a more illustrative form of the questionnaire with smiley faces to check. The teachers were also instructed to ensure understanding of the questions by reading them aloud one by one and giving the pupils enough time to answer after each question.

Sample and subgroups
The targeted pupils were in grades 1-6 and ranged in age from 7 to 12 years. The questionnaire was addressed to 33 primary schools with 7,535 pupils in total; 4,132 pupils from 25 schools participated in the study, and the response rate was 55%. Among the schools were two centralized service schools that had previously been called "special schools." They are schools with municipally assigned tasks and learners with specified groups of disabilities or pedagogical needs. Certain pupils are referred to these schools instead of attending mainstream schools. These schools also provide consultative practices for other schools.
Only schools with 50 or more completed questionnaires were included in the final sample, with at least two classes with two different teachers and classroom cultures represented in each school included. After schools with too few participants were removed, the data were compiled from 23 school units. The data were cleaned by removing cases where more than half of the questions were unanswered (23 cases), as well as those where respondents had obviously completed the questionnaire outside school (three cases). There were some missing data in the questionnaire statements.
However, the percentage of missing values in these variables was very small (ranging from 0.4% to 2.8%), and these missing values occurred randomly in the data (there was no statistically significant over-representation in any of the subgroups: gender, grade, or school-size). Missing data were imputed using the expectation maximization (EM) algorithm. The final sample consisted of 3,760 completed questionnaires.
In order to compare groups, different subgroups were formed from the sample.
First, 48.8% of participants were girls (n = 1,833) and 51.2% boys (n = 1,927). In order to examine differences between children of different ages, we formed three grade groups, presented in Table 1. The schools represent a wide variety of sizes (50-545 pupils) in different residential areas in a major city in Finland. It seems that many of the smaller schools were located in residential areas where families had low socioeconomic status (SES). To examine differences between schools, the following school-size groups were formed according to the Finnish National Board of Education's (2005) definition: small (and medium) schools (fewer than 300 pupils) and large schools (300 pupils or more). In the sample, seven schools were categorized as large and twelve as small. The proportions of the genders in both groups were as expected for the age groups (Ministry of Education and Culture, 2012).

Measures
The data were collected by using the Finnish translation (Booth & Ainscow, 2005) of the Booth and Ainscow (2002) Index for Inclusion questionnaire 3 called "My Primary School." Due to translation, some statements were slightly different in the Finnish version compared to the English. The English word "friends" was translated to the more general Finnish expression "classmates." The questionnaire includes statements with responses on a 3-point Likert scale (1 = I don't agree, 2 = I agree to some extent, and 3 = I agree). Before the data were collected, the smiley face questionnaire was tested with a small sample group. Minor adjustments in phrasing were made in order to be more precise.
The nineteen 1 statements on the Index for Inclusion questionnaire were subjected to exploratory factor analysis (EFA) to find out what kind of factors describing aspects of participation could be found. Descriptive statistics and correlations are presented in Table 2.
1 Item 20 was left out because it was not suitable for the youngest pupils in grades 1 and 2.

Analyses
The focus of this study was the interactions and collaboration between pupils and the interactions between a pupil and a teacher that took place in a classroom. That is why the PP and PT factors of social participation were chosen for closer examination. Distribution of the PP sum score was slightly skewed to the left, i.e., compared to symmetric normal distribution; higher values (positive answers) were more common.
However, the skewness and kurtosis statistics (skewness = -1.20 and kurtosis = 1.09) were well within reasonable limits for the analyses, assuming normal distribution (see Cain et al., 2017). Moreover, based on the large sample size and the central limit theorem, parametric tests of multivariate analysis of variance (MANOVA) and t-tests were used to compare the subgroups to gain a picture of differences in the experience of participation from the perspective of interaction. Tukey and Dunnet's post hoc tests were used to analyze the differences in detail. The statistical significance and the effect size measures were calculated by using partial eta squared (η p 2 ) and Cohen's d. Differences in proportions were analyzed using cross tabulation and a chi-square-test with effect size measure Cramer's V, where appropriate.

Validity and reliability
According to the questionnaire's authors, the Index for Inclusion is a tool for "developing learning participation in schools." The index was designed by the Centre for Studies on Inclusive Education in Britain and can be used to audit and develop inclusive practices in schools. The indicators and questionnaires used are intended to be used to explore students' knowledge in the process of creating more inclusive school cultures, policies, and practices. This particular questionnaire was chosen because the goal of this study was to gather information from the pupils' point of view and the statements cohered with the theory of social participation.
The reliability of the study was maximized by planning and executing the data collection carefully. The focus was on ensuring that the participants perceived and understood the process and the statements in as concordant a way as possible with the researchers and other participants. Another important point was to try to rule out environmental factors that could affect the answers, and therefore, answer confidentiality was emphasized. How these aspects of reliability were taken into consideration in practice in the researchers' actions has been described.
The response rate was only 55%, but the sample still represents a comprehensive sample of different kinds of schools in the city concerned. The data were cleaned and handled in a purposeful way. In the data analysis, the p values were reported. The effect sizes were added because of the large sample size (n = 3,760).

Pupils' experience of participation in general
The results are presented one research question at a time. The first research question addresses general findings regarding the experience of participation. In general, the means of the sum variables in the whole sample ( In the next section, the results are presented using comparisons between subgroups. Table 3 gives descriptive information about the means for the subgroups. Comparing experiences of participation between grade groups and genders MANOVA indicated differences between the genders in terms of PP and PT interac- p < .001, η p 2 = .03). Girls also experienced this in a more conforming way. The difference between the genders was the biggest in grade group 1 (grades 1-2) and smallest in grade group 2 (grades 3-4), but the difference did not vary statistically significantly between the grade groups F(2, 3748) = 2.37; p = .094).
explained by different patterns in the grade groups. Gender differences in the experience of PT interactions are moderated between the grade groups (F(2, 3748) = 12.44; p < .001; η p 2 = .007). A statistically significant difference between the genders is only found in grade group 1 (Table 4; t(1159.35) = 5.56; p = .002; d = .31) while older boys scored as high as girls or even slightly higher.
Pupils of different ages were compared using the grade groups presented in Table 1. Distribution of the genders was similar in all three grade groups (χ 2 (df = 2, n = 3.760) = .24; p > .05; V = .01). Pupils' experiences of social participation varied depending on the grade (F(4, 7496) = 94.78; p < .001, η p 2 = .048). As shown in Table 5, experience of participation in PP interactions was significantly higher (F(2, 3748) = 27.85; p < .001; = η p 2 = .015) favoring older pupils over younger pupils, whereas PT interactions were experienced less by pupils in upper (M = 2.34, SD = 44) grades than in the early years (M = 2.58, SD = .39). As can be seen in Table 5, all the differences were statistically significant, but the effect sizes varied (p < .001 in all comparisons). To sum up, the experience of social participation between classmates (PP) was more strongly explained by gender, and the interaction between pupil and teacher (PT) was more closely related to the pupil's grade. According to the effect sizes, gender explains more strongly pupils' experience of the PP aspect of social participation and grade group the aspect of PT interactions.
Between grade groups 1 and 2, the gender differences in terms of PP interactions was significantly large, (F(1,2524 = 4.49) p = 0.034; η p 2 = 0.002). As pupils get older, the difference between genders becomes smaller.

Comparing experiences of participation between school size groups
The third research question focuses on a comparison of pupils in small and large schools. How do pupils in small school communities experience aspects of social participation compared to pupils in larger schools? Is there any connection between the experience and the size of the school?
The smallest school in the study had 129 pupils and the largest 545 pupils. The number of responses from pupils varied, as the minimum number was 50 responses per school. The schools were grouped into two size categories for the analyses.
After the schools were grouped according to size, there were 1,878 pupils from 16 schools labeled small (under 300 pupils) ranging from 129 to 268 pupils and 1,882 pupils in seven large schools (300 or more pupils) ranging from 301 to 545 pupils.
Pupils' experiences of social participation varied depending on school size

Combining the effect of school size and gender group
Focusing on gender differences between large and small schools showed that (Table 4) there were significant gender differences in experiences of social participation

Combining the effect of school size and grade groups
Comparing small and large schools in grade groups in PP interactions showed that the experience of pupils in large schools was statistically significantly higher than in small schools across all grades (F(1, 3748) = 11,29; p = .001, η p 2 = .003). The level of PP interactions increased as pupils got older (F(2, 3748) = 27.85; p < .001, η p 2 = .015). The greatest increase was seen between grade groups 2 and 3. There was not a statistically significant difference between small and large schools regarding this increase between grades (no statistically significant grade group × school size effect). The PT interactions of grade group 1 pupils in small schools (M = 2.60, SD = .37) was highest in the study. However, this was reduced by grade group 2. Regarding PT interactions in large schools, the youngest pupils started a little lower in small schools, but by grade 3 the level ended up being slightly higher.

Combining the effect of gender and grade groups with school size
Next, we focus on the combined results of the effect of gender and grade groups between large and small schools. In PT there were borderline statistically significant (F(2, 3748) = 2.82; p = .060, η p 2 = .002) differences between small and large schools in terms of pupil teacher interaction changes in grade groups between boys and girls.
Regarding PP interactions, the result was not statistically significant. In small schools, boys' experience of PT started at a relatively high level and then declined, similar to girls' PT experience. In large schools, the starting level of PT participation was lower and therefore the decline in PT level was not as strong as for the boys in small schools.

Discussion
The aim of this study was to examine pupils' experiences of social participation in primary school by using seven statements from the Booth and Ainscow (2002)  The results showed that the general experience of social participation in both dimensions is at a high level among Finnish primary school pupils. According to Black-Hawkins (2013), complex shared experiences create the experience of participation.
The results indicate that Finnish primary schools as a social and collective environment seem to support the overall positive experience of social participation among pupils.

Pupil-pupil interaction
We chose questions about pupils' mutual helping and working in pairs to express social participation with classmates. As shown in a previous study, liberty of choice regarding discussions during lessons and liberty of choice regarding one's workmate were important for pupils (Backman et al., 2012). We should not undermine the importance of collaboration in terms of participation. Many researchers (Norwich, 2014;Rogoff, 2007;Rouse, 2016) have emphasized the importance of collaboration with other people as an essential factor regarding participation. In our study of PP interactions, the difference between genders in experiencing mutual help and collaboration is clear. The mean for girls was higher than that of boys at the beginning of primary school. An increasing trend was clear for both genders. The results show that by the end (grade group 3) of primary school, boys' experience of PP interactions is the same as for girls. This means that the upturn in PP interactions is clearly stronger for boys.
Previous studies have also found differences between genders. According to Gillander Gådin et al. (2013), taking responsibility for others is a key social discourse of femininity, and the responsibility shown by girls is often confused with maturity (p. 66). Valaitis (2002) found that teachers' behavioral expectations for girls are also much stricter than for boys. Thus, girls are often (culturally) expected to help other pupils in the classroom, and if we assume that girls are usually friends with other girls, they experience help from each other. The expectations for boys are different; boys are not expected to help each other the way girls are. When examining mutual helping as a form of active participation, we must not forget these social expectations and operational models, and their influence on pupils of different genders. This still does not explain the fast change in the boys' results.
Another explanation for increased PP interactions in older pupils is that the influence of peer groups becomes stronger as pupils grow older (see Bukowski & Adams, 2005). From the pupil's perspective, the importance of interactions between classmates becomes more important over the years. It is only natural that the PP aspect of inclusion should be and is stronger as the pupils get older. The pupils in grade group 3 are 12 to 13 years old. They are at the age when adolescence is starting and the importance of friends increases.

Pupil-teacher interaction
In PT interaction, we examined pupils' experience of teachers listening and helping and teachers' willingness to let pupils choose what to do in a classroom. The results show that the pupils' experience of equal treatment and positive encounters with their teachers is at a good level. Pupils experience positive interactions with their teachers. This result indicates that there is reciprocal dialogue between pupils and teachers (Booth & Ainscow, 2011, 2016. According to the pupils in this study, the Finnish primary school teacher's overall role in PT interaction was the same regardless of pupils' gender or age. In the primary school environment, encouraging the participation and inclusion of all students is largely in the hands of teachers. Teachers can either create opportunities which support their pupil's ability to make their own choices and work together, or turn back in time and use more traditional teacher-centered approaches in their work (Dyson et al., 2002;Kovalainen & Kumpulainen, 2007). Successful social participation requires teachers who support pupils' activity based on mutual acceptance and recognition (Florian et al., 2017). At the same time, we must remember that the results of a teacher's actions are not identical for all pupils in terms of supporting social participation. The same pedagogy may support one pupil's opportunity to participate while reducing opportunities for others. (Florian et al., 2017) Being able to be part of the decision-making process and being heard are important factors of participation (Mazurkiewich, 2013). As Backman et al. (2012) found, positive interaction and the experience of being seen and heard, as well as liberty in making decisions about schoolwork, are important elements for pupils. The present results show that pupils' personal experience of being heard by their teacher and being able to affect classroom activities (PT) is positive. In the early years of primary school, the total mean for the girls was higher than for the boys. Thus, the girls' positive experience of interactions with their teacher declines faster than that of boys. The decreasing trend of PT interactions is clear for both genders. However, the difference between genders diminishes and gradually disappears over time. The results show that by the end of primary school, boys' experience of PT interactions is slightly higher than girls'.  Mazurkiewicz, 2013). In Finland, the number of pupils is often higher in grades 3 to 6.
Furthermore, in the National Core Curriculum, academic requirements and expectations are higher for this age group. There is a considerable jump from being a beginner learner to being an active, self-regulated and competent learner between grades 2 and 3 (see Ministry of Education and Culture, 2014.) These changes in approach shape and are connected to teachers' beliefs, values and expectations regarding pupils.
We know from previous studies that a teacher's appreciation and individual feedback are important for pupils (Backman et al., 2012). Equally important is that teachers use pedagogical approaches that support pupils' free interaction and their learning and working together (Dyson et al., 2002). Is it possible that larger class-sizes and the pedagogy used do not give enough room for similar personal interaction and feelings of safety in the upper classes, which hinders PT interaction?
We can also ponder whether the reason behind the weakening results of PT interaction has something to do with increasing PP interactions. As the importance of classmates grows stronger as pupils become older, the pupils examine their relationship with their teacher more critically. It is quite possible that this trend is also due to the typical development process of children and their growing independence. Another question for future study is whether these trends remain after pupils leave primary school and their class teacher and start their studies in secondary school with subject teachers.

School size and social participation
Previous studies concerning pupils' social participation and school size are mostly lacking. We wanted to see whether there is a connection between these two aspects.
In this study, boys and girls regardless of school size experienced both PP and PT interactions similarly. Both genders' experience of PP interactions was stronger in larger schools. It was interesting that the highest means of our study were in large schools, both by girls (2.77) and boys (2.62) in grades 5 and 6.
As discussed earlier, the experience of PT interactions decreases as the pupils age.
One distinct result concerning school size was that the experience of PT interaction declined most in boys in small schools. The difference was close to being statistically significant. For some reason the pupils' experience of social participation was not the same in small schools as in large schools.
The small sample size (number of schools) must be kept in mind when interpreting these results. In the sampled city, families' socioeconomic factors tend to be connected to school size. That is, schools in lower socioeconomic status areas are usually small in the research municipality. Veland, Midthassel and Idsoe (2009)  All teaching and learning takes place within the context of human relationships, shaped by a school's culture and the values and beliefs of its members. Relationships -among students, among staff and between staff and students -are at the heart of understanding and developing policies and practices which support inclusion.
Social participation is one of the core values of inclusion. It requires the feeling of acceptance and involvement and it means that a person is actively able to make decisions about ones' life and engage in equal dialogue with others (Booth & Ainscow, 2016). Mazurkewicz (2013) stated that a sense of safety is required if social participation and active decision-making are to be pursued in a classroom. If we want to understand the complexity of pupils' social participation, the phenomena must be explored from a variety of angles. It is worrying that pupils' sense of PT interaction gets weaker in the upper primary school classes. It makes one wonder whether teachers are able to support the growing independency of older pupils. We must question why pupils' experience of being heard and helped by their teacher and having a say about classroom activities declines. We know that in Finnish primary education the curriculum's academic level changes markedly after the first two years. Is it possible that the pedagogy implemented in the classroom also changes at that point? Could this cause the negative change in pupil-teacher interaction?
There is a lack of studies concerning the whole age group in primary education and particularly a gap in studies concerning the connection between gender or schoolsize and social participation. By increasing understanding and actualization of social participation, it is possible to improve our schools and empower pupils. By increasing pupils' empowerment, it is possible to challenge existing unequal power relations between pupils and teachers in the school community (Baum, 2002). This empowerment should reflect on all pupils regardless of their age or gender.
The significance of this study is its ability to capture the easily overlooked concept of social participation from pupils' point of view. The results show that pupils' experience of social participation in a classroom environment is at a good level. The trends in terms of PP and PT interactions are clear: As pupils get older, interactions between classmates (PP) become more important, and the level of interaction between pupils and teachers (PT) declines. This result is in line with previous studies, which emphasizes the growing importance of peers as children get older, and the simultaneous diminishing role of teachers (Bukowski & Adams, 2005). For girls, the decline in their experience of PT interactions in small and large schools is statistically significant. This decline could be due to normal development and be part of entering adolescence at the age of 12 or 13. However, the trend is clear from the beginning of the school years, and holds even after when we added the effect of school-size to the study. Nevertheless, the reasons for these results require more research. We do not know what factors are causing these trends. The question that needs answering is whether teachers and schools should concentrate more on pupils and thereby change the declining trend of PT interaction.
If we want to gain a better understanding of the formation and manifestation of pupils' social participation, we must study more carefully the everyday environment of education and relationships in classrooms. In order to explain the connection between school size and participation results, we should study more deeply the factors behind the phenomenon. The quality of PP and PT interactions and the differences between schools require more examination. Whether some schools have more effective methods of supporting pupils' social participation is a question left unanswered. In order to develop a more structured perception of social participation, the phenomenon ought to be studied from both pupils' and teachers' perspectives. The use of additional qualitative methods would also be beneficial, since they could afford us a deeper understanding of the multifaceted phenomenon of social participation.

Methodological considerations
Although this study has methodological limitations, it takes scholarship one step further in the process of understanding inclusion and pupils' social participation as its mirror. Firstly, the questionnaire was designed as an instrument for school development, not for scientific purposes. It only has three-level scales, which can cause unreliability and inaccuracy in the results of the analyses and how they are interpreted.
Secondly, in this study only the students' self-reports were used as the data source.
Additional information obtained through observation of peers and teachers would give a deeper understanding of social participation. Thirdly, the size of effects found are minor or small. Furthermore, repeated measures from the same sample group would have increased the level of reliability.