Review on Research Methods for Studying Transition from Early Childhood Education to Primary Education

: The transition from early childhood to primary education is a process of changes that students undergo. Recent studies indicate that it is necessary to involve all actors in research: teachers, families, and students. Nevertheless, some researchers point to an “adult-centred” view, justiﬁed by the lack of linguistic communication among 5–7-year-olds. This study aims to describe the methods used in research on the transition from Early Childhood Education to Primary Education and to evaluate which methods support the participation of all actors involved. We conducted a systematic review of empirical studies between 2016 and 2021. The data show, on the one hand, that students of such a young age are not usually included in these studies. However, there has been an increase in studies that rely on students’ opinions and perceptions. On the other hand, studies that include all actors involved in the transition are a minority. Students of such a young age are not usually included in these studies. When they are included, adapted information collection tools are used. Only by having the students and triangulating the information among all participants is it possible to provide complete information on the process. In addition, there is a lack of action research designs to offer comprehensive and practical improvement actions


Introduction
The transition from early childhood education (ECE) to primary education (PE) is a process that begins before the last day of ECE and ends when the students feel a sense of belonging and well-being in the new primary school group [1,2].In most European Union countries, ECE is not part of compulsory schooling, or at least not until the year before the start of primary school [3].So much so that in many of them, the last year of ECE is known as 'pre-primary'.In the nineteenth century, with the industrialisation of Europe, the first nursery schools, known as 'crèches', were created to care for children while their mothers joined the labour market.With this idea of 'pre-primary', ECE has for years had a social prestige inferior to primary and secondary education and with more welfare than educational character [1,4,5].In the last decade, ECE started to have an identity of its own and abandoned somewhat the idea that it is a preparation for PE.However, it is still too often associated with the idea of reconciling work and personal life, prioritising the 'nursery' concept over the educational one [4].This idea is widespread, especially in the southern part of the EU [3] and much of Latin America [4].
However, the OECD (Organisation for Economic Co-operation and Development) highlights the essential role that ECE plays in students' personal well-being and cognitive and socio-emotional development.Attending this stage contributes to fostering inclusion and mitigating social inequality, especially for children from socio-economically disadvantaged backgrounds.Moreover, reports such as PISA (Programme for International Student Assessment) show that students who have attended ECE perform better overall [6].The expansion of ECE is significant in the second cycle (3-5/6 years).Across the OECD, 87.0%Educ.Sci.2023, 13, 254 2 of 18 of children aged 3-5 years are in school.This percentage rises to 97% in the EU.In 2019, around 3 out of 10 EU regions had a rate above 96%, while 30 regions-mostly in Belgium, Ireland, and France-had 100% participation of children before the start of compulsory PE [7].
In addition to the social difference mentioned above, early childhood and PE are consecutive stages that historically have had a differentiating character that conditions and affects children's transition between these stages [1,[8][9][10]; perhaps the most apparent difference is the pedagogical line of each stage.While PE has a very academic approach, ECE has a much more global and educational character [4,11].Secondly, this pedagogical line conditions the role of the students at each stage, since while in ECE, they are assigned more active and participatory roles, which facilitate their autonomy.In primary school, their actions tend to be more conditioned by the academic component.Just as the roles of children are different, the role of teachers is also different.Therefore, thirdly, early childhood teachers act as motivators or provocateurs, while in primary school they the directors of the activity and transmitters of content.Fourthly and lastly, the use of spaces is another of the major differences between the stages.In ECE, the use of space is globalised, dynamic, and multifunctional, while in PE, the central area is the desks, without making use of the space as a possibility for learning.These situations condition the activities carried out, the resources used, and the grouping of the students at each stage [1].
Some of these changes between stages are clear and very specific, such as the last day of early childhood, the first day of primary school, or the change of physical location, which mark a drastic change.However, other earlier, more subtle and complex processes are also involved.This process of adaptation is experienced by children, families, educators, and other members of the community through physical, social, cognitive, and relational changes [12][13][14].
These inequalities between the stages, together with the fact that ECE is not usually a compulsory stage, make it a transition that is less studied than later ones and condition how it is studied in research.
In research on transition, the vision that adults (parents and teachers) have had of the process, of the experiences, and of the relationships between those involved in the process has always been important.Visualising children as active participants in society and science is not intended to divide or exclude the vision of adults but instead to review a reality that may be conditioned by adultcentrism [15].The adultcentrism belief holds that adults can and should speak on behalf of children.However, relevant research on childhood has shown that children's opinions and experiences are essential to our knowledge of this stage [16][17][18][19][20][21].In this sense, as an opposite concept to this adult-centred tendency, we find the Student Voice, an Anglo-Saxon concept that defines the current that motivates the participation, consultation, and opinion of students in educational centres.Assuming this trend means understanding that students should be participants in the management of the commons and agents of their learning, recognising themselves as authoritative voices [22,23].Research such as Wilder and Lillivist [24] indicate why young children's voices are often excluded from research and knowledge.This situation is due, on the one hand, to the belief that the child will only reproduce the adult's message.However, in childhood, the opinions of peer networks are also transmitted.
On the other hand, this exclusion is also due to the methodological complexity of working with such young children.However, paying attention to and analysing children's discourse allows us to highlight discursive fractions lost when adults speak for children.Despite this, increasingly, researchers in the fields of ECE and childhood studies have a long tradition of including children's voices within research paradigms [20,[25][26][27][28][29][30][31][32].This idea does not seek to contradict or dismiss previous research or understanding but to complement it.Five-year-old students have a vocabulary of between 2000 and 5000 words, speak clearly, and form complex compound sentences.In addition, these students already know how to use the future tense and understand time sequences [33].Taking this into account, we can obtain first-hand information from the students with an appropriate script.In addition, it should be considered that when interviewing a child, attention should be paid not only to the story's content but also to the tone, volume, and non-verbal communication used [33,34].Furthermore, it is essential that the strategies used with these students are varied and have different mechanisms of expression, not only using the spoken word [22].
On this basis, we ask: what research designs are used to investigate the transition from ECE to PE in scientific research?To answer this question, we conducted a systematic review of empirical studies.Our aim is to describe the methods used in research on the transition from ECE to PE and to evaluate which methods support the participation of all actors involved.To describe the method, we analyse four key aspects: (a) study design; (b) sample characteristics; (c) data collection techniques; and (d) data analysis techniques.

Inclusion Criteria and Selection Procedures
This study consists of a systematic review of empirical scientific articles on the transition from early childhood to primary school.The purpose of this study is to analyse the characteristics of empirical studies on the transition from ECE to primary school.This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA Statement [35] (Figure 1).The search was conducted using two databases: Scopus and Web of Science (Web of Science Core Collection; Current Contents Connect; Derwent Innovation Index; KCI-Korean Journal Database; MEDLINE; Russian Science Citation Index; SciELO Citation Index).Search terms included controlled terms (Thesaurus in ERIC) and terms with possible variable endings.
The document search for the systematic review was conducted in four stages.These stages, identified in the PRISMA diagram, were Identification, Screening, Eligibility, and Included [35] (Figure 1).
First, in the Identification phase, a search string was used to select articles with limitations that could be automatically applied in the databases: it had to be a scientific article; it had to be published between 2016 and 2021; it must have been published in English.In this Identification phase, 460 articles were identified (Scopus n = 246 and Web of Science n = 214).In this process, we used Mendeley (version 1.19.8) to manage the documents and eliminate duplicates (n = 118).Once the duplicate articles were removed, 342 articles were screened.
Secondly, in the Screening phase, the titles of the articles were read to eliminate those that dealt with subjects outside the object of study.This phase was not automatic and required the researcher's study.
Third, in the Eligibility phase, the investigators read the abstracts of the 141 articles that reached this phase.To facilitate the analysis, the articles were included in the MAXQDA Analytics Pro 2022 software (22.1.1).Articles that did not have a method of analysing their own data were excluded, as we were interested in studies that worked directly with the review phenomenon.
Finally, in the Included phase, we had 67 articles that were included (Figure 1).The reasons for exclusion in both the Screening and Eligibility phases were as follows: • It was in a language other than English.

Data Extraction and Coding
We used an inductive approach to analyse the method of empirical articles on transition.
Firstly, we analysed the design of this type of study, which was used as a first approach to the study.
Secondly, we analyse the samples' characteristics to ascertain their geographical origin, because it can provide information on the characteristics that motivate these investigations in each region.In addition, we analyse, on the one hand, whether all the agents involved are asked and, on the other hand, the sample size.This is because there may be a relationship between the size of the sample and the type of population it includes, relating large samples to adults, as it is easier to access and collect information with adults than with children.
Thirdly, we analyse the data collection techniques for adults and children because we are interested in the specific methods used in each sample.

Data Extraction and Coding
We used an inductive approach to analyse the method of empirical articles on transition.Firstly, we analysed the design of this type of study, which was used as a first approach to the study.
Secondly, we analyse the samples' characteristics to ascertain their geographical origin, because it can provide information on the characteristics that motivate these investigations in each region.In addition, we analyse, on the one hand, whether all the agents involved are asked and, on the other hand, the sample size.This is because there may be a relationship between the size of the sample and the type of population it includes, relating large samples to adults, as it is easier to access and collect information with adults than with children.
Thirdly, we analyse the data collection techniques for adults and children because we are interested in the specific methods used in each sample.
Finally, we analyse the data analysis techniques used in these articles.This analysis will be conditioned by the nature of the data and the size of the samples.This is a further indicator when analysing the method followed in these studies.
The data extraction, coding, and cleaning were done using the statistical software MAXQDA Analytics Pro 2022 (22.1.1.).

Results
This study aims to determine the research methods used in empirical research on the transition from ECE to PE.

Results of the Research Design
The design classification was based on the research methods manual by Coe et al. [36], allowing for a homogeneous classification of the methods used in these studies (Table A1).
Case study (n = 25) was the most used approach in this type of research, using multiple sources of evidence.This design was used to explore a part of transition that was not very well known, such as the opinion of the students, as well as to learn about the reality being experienced in centres with specific characteristics.
Next were longitudinal study designs (n = 23), with the understanding that the transition did not occur at an exact point in time.These designs allow longitudinal data to be obtained, which are data collected at several points in time.This design enables the research to assess and measure change initially in the early childhood stage and subsequently in primary school.
Survey methods (n = 18) came next, which are standard in studies that aim to obtain information from adults, i.e., teachers and families as agents involved in the transition.
Studies with an intervention design were the next most prevalent (n = 8).These studies attempted to change or modify the transition phenomenon in some way with the intention of evaluating the impact of the intervention.The groups participating in these designs were evaluated before and after the intervention in order to establish relationships with the results obtained at the time of the change of course.
Design through secondary data followed next (n = 7), comprising studies based on data which had already been collected in some way.This provided data mostly on the students or the socio-economic characteristics of the families, which allowed us to analyse these data from a different perspective to that from which they were analysed the first time.
Fewer still were action research designs (n = 4), studies which were action-oriented rather than descriptive.This design was present in some research from 2018 onwards and did not, in all cases, include the vision of all actors.
Lastly, the ethnographic-type design (n = 3) involved a sustained investment of interest and commitment over time for these transition studies.This design seeks to capture the lived reality of transition for all participants by understanding how they behave individually and as a group.

Results of the Characteristics of the Sample
Regarding the sample, different aspects were analysed: the country to which the sample belonged, the population it represented (family, teachers, students, management team, or the most legislative part), and the sample size.
Firstly, of the geographical characteristics of the sample, the empirical studies were mainly distributed in Europe (n = 40), followed by Oceania (n = 20), America (n = 11), and Asia (n = 8).Within these studies, samples including several countries were analysed [37,38].The country with the most published empirical transition studies was Australia (n = 18), where transition awareness was very high and consistently supported by the administration and schools.Many of these Australian studies were driven by the start of compulsory schooling for Aboriginal students.Figure 2 shows the interest in the topic in developed countries.compulsory schooling for Aboriginal students.Figure 2 shows the interest in the topic in developed countries.Secondly, we analysed the population that the sample represented.The subjects sampled in this research belonged to the school sector and the personal industry of the student body.In the case of the school, the most frequently addressed were teachers.
In the samples analysed, there were no notable differences in terms of the level of participation of the different agents: teachers (44.9% of the samples), family (43.5%), and students (42%).However, it is true that when it came to adults, there was a tendency to ask about experience, expectations, and opinions.This was not the case with students, most of whom tended to provide information through adults, through tests, teacher observations, and family perceptions, among others [28,71].Few studies included all stakeholders in their research, as shown in Figure 3. Secondly, we analysed the population that the sample represented.The subjects sampled in this research belonged to the school sector and the personal industry of the student body.In the case of the school, the most frequently addressed were teachers.
In the samples analysed, there were no notable differences in terms of the level of participation of the different agents: teachers (44.9% of the samples), family (43.5%), and students (42%).However, it is true that when it came to adults, there was a tendency to ask about experience, expectations, and opinions.This was not the case with students, most of whom tended to provide information through adults, through tests, teacher observations, and family perceptions, among others [28,71].Few studies included all stakeholders in their research, as shown in Figure 3. Thirdly, we looked at the size of these samples.Research designs involving students mainly involved samples of less than 200 students in different schools.Articles with samples less than 30 subjects focused on a single classroom in a school or on students with specific characteristics, e.g., hearing difficulties or special educational needs (Table A2).It was more common to study students from more than one school (n = 24).In these cases, we found (a) samples below 100 students (n = 7) and (b) samples between 100 and 250 students (n = 8).Although less common, we also found samples of students between (c) 900 and 5000 subjects (n = 8).Less common were student samples like Suntheimer and Wolf [30], with more than 18,000 sampled children.
Samples above 1000 students were obtained through tests, questionnaires, or databases.The sample size increased when direct and active interaction with the students decreased.On the one hand, it may be that information about the children was obtained through questionnaires to parents or teachers.On the other hand, the information may have been obtained through standardised tests which aimed to assess the students' competence in areas such as literacy, numeracy, or behaviour.Information was even obtained from previous macro-surveys, especially in the United States.This was the case of large databases with information from several years where all kinds of data were collected and were useful in this field, analysing factors that had to do with transition, such as attention regulation, language, or emotional competence.This was the case, for example, of the 'Fragile Families and Child Wellbeing' study on families at risk and child well-being or the ECLSK:11 (Early Childhood Longitudinal Study-Kindergarten), with 18,200 children [30] or with 4898 students in 20 large US cities [28].In some of these cases, where data were not drawn specifically from the educational setting, the samples were randomly selected.
In the case of adults, we must differentiate between ECE teachers, primary school teachers, and families.It is common in studies carried out in schools to also obtain information from the students' families, so that in many cases equal samples of students and families were obtained.In the case of teachers, the samples were smaller, mostly less than 50 subjects.The largest samples of families were above 2500 subjects [n = 2662 [51]; n = 3444 [75]; n = 4464 [76]].The highest teacher samples were 748 [77] and 1322 [50].
Empirical studies that sought to give an overall view of all agents, triangulating information from all of them, had a smaller sample size, especially of students: 6 students [21], 8 students [24], 16 students [78], 26 students [20], 49 students [42], and a maximum of 233 students [58].The exception was the research by Cook et al. [45], where student Thirdly, we looked at the size of these samples.Research designs involving students mainly involved samples of less than 200 students in different schools.Articles with samples less than 30 subjects focused on a single classroom in a school or on students with specific characteristics, e.g., hearing difficulties or special educational needs (Table A2).It was more common to study students from more than one school (n = 24).In these cases, we found (a) samples below 100 students (n = 7) and (b) samples between 100 and 250 students (n = 8).Although less common, we also found samples of students between (c) 900 and 5000 subjects (n = 8).Less common were student samples like Suntheimer and Wolf [30], with more than 18,000 sampled children.
Samples above 1000 students were obtained through tests, questionnaires, or databases.The sample size increased when direct and active interaction with the students decreased.On the one hand, it may be that information about the children was obtained through questionnaires to parents or teachers.On the other hand, the information may have been obtained through standardised tests which aimed to assess the students' competence in areas such as literacy, numeracy, or behaviour.Information was even obtained from previous macro-surveys, especially in the United States.This was the case of large databases with information from several years where all kinds of data were collected and were useful in this field, analysing factors that had to do with transition, such as attention regulation, language, or emotional competence.This was the case, for example, of the 'Fragile Families and Child Wellbeing' study on families at risk and child well-being or the ECLSK:11 (Early Childhood Longitudinal Study-Kindergarten), with 18,200 children [30] or with 4898 students in 20 large US cities [28].In some of these cases, where data were not drawn specifically from the educational setting, the samples were randomly selected.
In the case of adults, we must differentiate between ECE teachers, primary school teachers, and families.It is common in studies carried out in schools to also obtain information from the students' families, so that in many cases equal samples of students and families were obtained.In the case of teachers, the samples were smaller, mostly less than 50 subjects.The largest samples of families were above 2500 subjects [n = 2662 [51]; n = 3444 [75]; n = 4464 [76]].The highest teacher samples were 748 [77] and 1322 [50].
Empirical studies that sought to give an overall view of all agents, triangulating information from all of them, had a smaller sample size, especially of students: 6 students [21], 8 students [24], 16 students [78], 26 students [20], 49 students [42], and a maximum of 233 students [58].The exception was the research by Cook et al. [45], where student data were obtained from databases with 1157 subjects in a longitudinal study using data from the Behaviour Outlook Norwegian Developmental Study, with data collection from the age of six months.

Results of Data Collection Techniques
The data collection technique most used by the agents in these types of study was the interview (Figure 4).The use with adults-families and teachers-stands out above all.
data were obtained from databases with 1157 subjects in a longitudinal study using data from the Behaviour Outlook Norwegian Developmental Study, with data collection from the age of six months.

Results of Data Collection Techniques
The data collection technique most used by the agents in these types of study was the interview (Figure 4).The use with adults-families and teachers-stands out above all.The techniques used exclusively with the adult population were the survey and the focus group.The survey was used especially with families.The focus group was mainly used with teachers, as the aim was to get an overview of the situation at their school.The focus group made it possible to deal with new issues that teachers detected at their schools or based on their experience.It was not usual to obtain information from adults from secondary databases, especially from teachers.In this sense, it is worth highlighting, on the one hand, the collection of information from families through social networks such as Facebook [79].On the other hand, the studies mainly analyse the relationships between parenting processes, socio-emotional readiness, and children's reading achievement [75].
In this type of study, observation of adults was not a common technique.However, the use of observation made it possible to analyse important aspects of transition such as team meetings, community activities, or classroom activities [47,48,62].Despite this, the use of observation was more common with children than with adults.With teachers, there was rarely any analysis of documentation or sharing of documents between teachers at either stage [51,60].
In the case of children, test-based data collection was used to a large extent.These tests considered mathematical variables [48,73,80], behavioural variables [30,81], vocabulary variables [73], language variables [30,32], and cognitive skills variables [30], among others.Most of this research dealt with academic or behavioural aspects; only a minority The techniques used exclusively with the adult population were the survey and the focus group.The survey was used especially with families.The focus group was mainly used with teachers, as the aim was to get an overview of the situation at their school.The focus group made it possible to deal with new issues that teachers detected at their schools or based on their experience.It was not usual to obtain information from adults from secondary databases, especially from teachers.In this sense, it is worth highlighting, on the one hand, the collection of information from families through social networks such as Facebook [79].On the other hand, the studies mainly analyse the relationships between parenting processes, socio-emotional readiness, and children's reading achievement [75].
In this type of study, observation of adults was not a common technique.However, the use of observation made it possible to analyse important aspects of transition such as team meetings, community activities, or classroom activities [47,48,62].Despite this, the use of observation was more common with children than with adults.With teachers, there was rarely any analysis of documentation or sharing of documents between teachers at either stage [51,60].
In the case of children, test-based data collection was used to a large extent.These tests considered mathematical variables [48,73,80], behavioural variables [30,81], vocabulary variables [73], language variables [30,32], and cognitive skills variables [30], among others.Most of this research dealt with academic or behavioural aspects; only a minority of these tests dealt with social relationships between teachers and family and between students and teachers [58].
To a lesser extent, children's behaviours were also observed through direct or indirect non-participant observation.These student observation activities involved different activities, games, or moments in the classroom [17,25,28,31,48,62,63,78].These types of techniques made it easier for the researcher to obtain information in a way that was natural for the student.They were familiar strategies for the students, and, from the researcher's point of view, they did not generate biases that could lead to distortions in the student.In the analysis of the samples, there was no significant difference in the participation of the different main agents (teachers, families, and students) (Figure 4).In the case of adults, information about their experiences, opinions, or needs was used.In the case of children, more than 50% of the data came from tests of academic skills or from databases.Interviews with students were a more common tool in recent research; the increase can be seen in Figure 5.
Educ.Sci.2023, 13, x FOR PEER REVIEW 9 of 18 of these tests dealt with social relationships between teachers and family and between students and teachers [58].
To a lesser extent, children's behaviours were also observed through direct or indirect non-participant observation.These student observation activities involved different activities, games, or moments in the classroom [17,25,28,31,48,62,63,78].These types of techniques made it easier for the researcher to obtain information in a way that was natural for the student.They were familiar strategies for the students, and, from the researcher's point of view, they did not generate biases that could lead to distortions in the student.In the analysis of the samples, there was no significant difference in the participation of the different main agents (teachers, families, and students) (Figure 4).In the case of adults, information about their experiences, opinions, or needs was used.In the case of children, more than 50% of the data came from tests of academic skills or from databases.Interviews with students were a more common tool in recent research; the increase can be seen in Figure 5.

Results of Data Analysis Techniques
In terms of data analysis, descriptive analysis was the most common approach in these articles.Forty-four articles' analyses were limited to a description of the data.This was undoubtedly the most used analysis technique in these studies.The next most used analysis techniques were correlations (n = 9) and regressions (n = 8).In a smaller percentage, linear [30,80] and logistic regression [25,38,45,76] was used.Even less common was the use of structural equations (n = 2) [75,86], multilevel analysis (n = 1) [81], modelling techniques [83], and estimation [32].

Summary of Results
In terms of design, most of the articles are case studies and longitudinal design.The sample does not usually include all agents (teachers, families, and children).Moreover, in some cases, the teachers are only from either ECE or PE, which does not provide a

Results of Data Analysis Techniques
In terms of data analysis, descriptive analysis was the most common approach in these articles.Forty-four articles' analyses were limited to a description of the data.This was undoubtedly the most used analysis technique in these studies.The next most used analysis techniques were correlations (n = 9) and regressions (n = 8).In a smaller percentage, linear [30,80] and logistic regression [25,38,45,76] was used.Even less common was the use of structural equations (n = 2) [75,86], multilevel analysis (n = 1) [81], modelling techniques [83], and estimation [32].

Summary of Results
In terms of design, most of the articles are case studies and longitudinal design.The sample does not usually include all agents (teachers, families, and children).Moreover, in some cases, the teachers are only from either ECE or PE, which does not provide a complete analysis of the transition.On the one hand, if only ECE teachers are involved, we only know the opinion of the teachers who initiate the transition.On the other hand, if only PE teachers are asked, this does not provide a complete picture of the transition either, as it started at the previous stage.When families are included, it is usually in a sample proportional to that of the students.The research used with larger samples that obtained information from adults or information from children was through tests of academic qualities or previous databases.However, there is increasing interest in learning about children's experiences, opinions, and needs.The research involving the views of pupils, teachers, and families has a smaller sample size due to the data collection complexity and the context's depth and description.Moreover, samples with children require a more demanding ethical process, considering how the interview is conducted (e.g., conducting pre-research tasks to build trust, conducting interviews in pairs or groups).In addition, alternative communication aids such as images, drawings, or photographs should be included.The nature of the data obtained in this research prioritises a descriptive type of analysis in most studies.

Discussion and Conclusions
Empirical studies have generated new knowledge about the transition from ECE to PE.The systematic review has made it possible to analyse the research methods used in these empirical studies regarding design, sample, data collection techniques, and data analysis.In addition, we evaluate which methods support the collection of information from all actors involved to develop recommendations on how to investigate the transition to school comprehensively.
The most common research designs have been case studies and longitudinal studies, because, on the one hand, the need to know the reality of a school or a region and its particularity made this type of design very common.On the other hand, the very definition of transition implies a time process that justifies the longitudinal nature of the studies.This makes it possible to contrast the ECE and PE years and to obtain data that are better adapted to the transition construct.The designs detected were focused on description and very rarely on action research.
As for the geographical characteristics of the sample, in the EU, Nordic countries are more aware of transition policies than southern countries [3].This situation was reflected in the number of studies on transition.The analysis of the sample by area showed that the study of transition occurred in developed regions or regions with particular social conditions.
More broadly, among the principal actors involved, no differences could be detected between them.Few transition studies have a sample of teachers, families, and students [20,21,24,30,42,45,58,87]. The size of the sample also has a lot to do with this.The complexity of triangulating this information and providing contextual detail becomes difficult in large samples.When the sample consists only of adults (teachers or families) or information previously collected in databases, samples tend to be much larger.
In the case of teachers, the samples consider representatives from both stages [20,24,[51][52][53][54][55][56][57][58][59].If teachers at both stages are not asked, biased information is provided.On the one hand, if only ECE teachers are involved, we only know the opinion of the teachers who initiate the transition.On the other hand, if only primary school teachers are asked, this does not provide a complete picture of the transition either, as it started at the previous stage.Moreover, school management is often not among the sample population from which to obtain information, so the information obtained was isolated to the classroom with each teacher whose students took part in the research.If school principals were also involved in these studies, they could collaborate in the development of school transition protocols and provide an overview of the school outside the classroom.
After analysing the data collection techniques used, a significant presence of adultcentred discourse was observed.The participation of adults was focused on sharing their experiences, opinions, or perceptions, while in the case of children, it was focused on learning about their qualities, competencies, and skills.The data collection techniques used with the students were focused on tests measuring academic or behavioural parameters that could predict the transition of the students at the academic or performance level [88].This type of measurement supports the theory that it is the student who should be prepared for school, not the school that should be prepared for the student.The influence of PE can also be seen here, where more purely academic aspects took precedence.In addition, there was a tendency to obtain information from children through adults, either their teachers or their parents.In recent years, this situation has changed, and there has been an increase in research that actively involves students, especially through interviews [17,21,70,82,85].
Knowing the students' perceptions, concerns, and expectations makes it possible to complete a vision of transition that, for some time, was only represented by adults.This approach is not intended to obviate the role of the adult but to complement it.One of the main conclusions so far in the field of transition to primary school is that the success of these depends to a large extent on the relationships between all those involved.In order to understand the process and establish guidelines that favour these relationships, it is necessary to know the experiences and perceptions of all the agents involved.Good transition implies that the contexts and actors involved engage in the process actively, are connected, and can support each other [8].Leading studies in this field, such as Hacıibrahimo glu and Kargin [46], developed specific tests to identify difficulties in the transition to primary school with the support of teachers.In the same way, transition analysis should always include samples of teachers, families, and students.The systematic review shows that this is not always done.All proposals for action and improvement around transition must involve all stakeholders.In the case of students, their opinions and experiences must be considered as an active part of the transition.In the design of studies on transition, we must establish tools for collecting information that compensates for the lack of linguistic development that pupils of this age may have.In this way, the adults' discourse will complement the students' point of view.
In conclusion, transition research is evolving to include students' views.These studies involve case study designs, with small samples in which they work directly with the students, between 100 and 250 subjects.Due to the very concept of transition, longitudinal studies provide complete information on the process.It is essential to obtain information from the students through different channels that can be triangulated and, in turn, completed with other agents (families and teachers).In this type of research, it would be necessary to increase information on the needs of researchers in dealing with children, the process of trust to initiate the data collection, and the most important ethical aspects of this type of research, which are not being mentioned.Based on the results of our literature review, we make recommendations for designing a methodology to study the transition from ECE to PE.

Implications for Methodological Designs on the Transition from ECE to PE
The systematic review has allowed us to analyse the design of study transition methods.With this in mind, we make recommendations for the design of future transition analysis methods.

Including the Students
First, it is necessary to provide a complete picture by including the students.This vision is not only achieved through tests, but it is necessary to adjust techniques that allow us to obtain their opinions and perceptions.In order to obtain the student's opinion, there are aspects that must be considered, such as the environment in which the data are collected.The environment in which the interview takes place must be familiar to the students and not be full of stimuli that distract them.In these cases, rapport is fundamental when the interviewees are children of such a young age, as the researcher must transmit confidence and tranquillity.It is usual to carry out a previous activity with the children so that they feel close, but it is not usual to mention information like this in the methods, though it is key to the success of the data collection.
Interviews with the students should always be conducted in pairs or trios to avoid them feeling intimidated by the adult.For this, it is important in the design to have different techniques for obtaining effective information with the youngest students, such as drawing conversation, assembly (discussion group), or observation.
Drawing can facilitate the gathering of information from the students and be a complement when interviewing a student.The pre-schematic stage in the artistic development of children arises between 4 and 7 years of age, where scribbling is replaced by figures and visual objects from the children's environment.At this stage, they make more controlled strokes than in the scribbling stage (2-4 years) and these strokes are perfectly identifiable, initiating definitive representative configuration.When students draw, they are creating, rewriting the object to be represented and generating a new reality whose relationship with the original idea is established through reductions, simplifications, and generalisations of its qualities.In many cases, this favours the fluid communication of aspects that the students consider important and that they show in their drawings [89].In addition, the relationships of what is drawn and the representation of this in space are established according to their emotional meaning.Children's drawing has not only been studied in relation to ECE, but has also been widely used in developmental psychology, perception, and motor skills to study different dimensions or variables [90].The true effectiveness of this recording will be marked by the researcher's capacity for questioning and integration.
The assembly that takes place in ECE classrooms is understood as 'a collective space for dialogue where the floor is used, ideas and feelings are expressed, proposals are made, and the ideas and proposals of all are taken into consideration, ( . . .)' [91] (p.10).This situation, which the children have internalised in their classroom routine, could be used to go deeper into the subject and triangulate information obtained from the individual interviews, clearly and directly helping the research with the students [17,92].Like focus groups, understood as a qualitative group technique, this allows for the collection of relevant information on a research problem, but has only been used with adults (teachers and families).The intention of focus groups is to promote communication among participants and generate a group discourse to identify different trends and regularities in opinions [93].In the research analysed, this technique was only used with adults; however, in ECE they use a daily work methodology that is very similar to the focus group, which would be an advantage for research.
Observation of students provides useful information for research with children.Observation work is based on simulated situations with stories, toys, or even observing them playing in everyday classroom life.All these situations allow observation to assess social and academic skills, which research in this field should consider [17,25,28,31,48,62,63,78]. What is essential in studies with such young students is to obtain information in different ways, understanding these as an opportunity for the students to complete their transfer of data.
This type of participation is in line with EU actions to empower children to be active citizens and members of democratic societies.Article 24 of the Charter of Fundamental Rights of the European Union states that: 'Children ( . . . ) may express their views freely.These views shall be considered in matters concerning them in accordance with their age and maturity'.The right of all children to be heard and to have their views taken seriously in accordance with their age and maturity is also set out in Article 12 of the UN Convention on the Rights of the Child.For the 17 Sustainable Development Goals (SDGs) of 2030, the United Nations states in Goal 16.7 to ensure responsive, inclusive, participatory, and representative decision making at all age levels.As is evident, the European Commission promotes and protects children's right to be heard in its legislation and policies.Children must assume a role as agents of change, with a role as global citizens, facilitating and adapting their participation processes.For this, it is necessary to be aware of the myths and beliefs about the social representation of childhood and the vision of children as 'the citizens they will be' rather than as the citizens they are.
As the results show, in recent years, the presence of such young subjects has been significant, which makes it easier to abandon the adult-centred vision.These indicators may show a boost in child participation in the field of research and in the field of school organisation.However, 'this will not be possible without a change in the attitude of the educational community, which shows reluctance anchored in a view of children as passive objects' [94].

Longitudinal Designs
Longitudinal studies are important in this field to obtain data on the whole transition process.In order to address this transition, it is necessary to obtain information from the ECE and PE stages.Moreover, within this design it must be considered that teachers from both stages must be represented.Sometimes, when longitudinal studies are not carried out, ECE teachers are considered more.It is important to bear in mind that the process begins with ECE teachers, but the coordination and perspective of teachers from both stages are fundamental.As far as the students are concerned, it is essential to have a vision at both stages, as the ECE discourse provides a belief of what is going to happen.In ECE, the information will be based on information received through third parties from the transition process.In primary school discourse, the information will be based on what the experience was like for them.This information at two moments, as with the parents, will allow better data triangulation.

Characteristics of Schools
Most transition studies involve schools or teachers who are transition-aware or at least slightly sensitised to accept participation in the study.This bias in the selection of the sample is sometimes taken for granted in the studies, but far from being a disadvantage, it allows the characteristics of these schools to be detailed.Transition to primary school has factors associated with the school context.That is, the sociocultural level of the families, the social environment, and the initiatives implemented in the school, which can be decisive.Therefore, it is essential to provide detailed and complete information on the sample environment to compare schools with different characteristics, populations, and initiatives that may affect the transition.The actions established in the centres for the change of stage, the methodology followed, or the centre's project should be contemplated and detailed in these studies.These aspects, which are specific to each centre, may mean a difference in the results of the transition that are conditioning the differences between centres and should be considered.

Ethical Considerations
Ethical issues become critical in studies involving such young students.The legal guardians of the students must authorise their participation in each study.These legal aspects are not always mentioned in the research in a detailed and explicit way, as was observed in the analysis carried out in this study.However, we consider that, within this type of research method, there is a fundamental ethical part that has to do with the interviewed subjects.Moreover, as adults and from an adult-centred viewpoint, we can come to believe that we only need the relevant legal authorisations to record and interview the child, forgetting to ask whether they want to talk to us or agree to be recorded.This whole ethical process of respect is implicit in having the students respectfully and inclusively within the research and must be included in the research method these studies follow.

Figure 1 .
Figure 1.Diagram to show the process used for review (original figure developed by the authors for this publication).

Figure 1 .
Figure 1.Diagram to show the process used for review (original figure developed by the authors for this publication).

Figure 2 .
Figure 2. Map distribution of empirical studies on the transition from ECE to PE.

Figure 2 .
Figure 2. Map distribution of empirical studies on the transition from ECE to PE.

Figure 4 .
Figure 4. Distribution of data collection techniques used in the articles (in absolute frequency).

Figure 4 .
Figure 4. Distribution of data collection techniques used in the articles (in absolute frequency).

Figure 5 .
Figure 5. Articles where students' perceptions and opinions are considered.

Figure 5 .
Figure 5. Articles where students' perceptions and opinions are considered.