Recording and Reporting School Attendance and Absence: International Comparative Views on Attendance Statistics in Sweden, Germany, England, and Japan

: There is general consensus on the negative consequences of school non-attendance, but from an international comparative perspective, it is surprising how few studies have compared school attendance problems (SAPs) in different societies and education systems. In this article, SAPs are analysed through the lens of official statistics in four countries with different education systems: England, Japan, Germany (represented by two federal states), and Sweden. The purpose of this article is to investigate which data on school attendance and absence are available in four different countries and to facilitate a comparison between school attendance statistics and possibly different conceptualisations of SAPs. The article analyses statistics and official data collected by national school authorities and education agencies. Backgrounds within systems are provided and differences between the countries are analysed. England and Japan provide official data to the public on a regular basis, while Sweden and most federal states in Germany do not. A lower threshold for how much absence is considered problematic is found for Japan, England, and Thuringia (one of the investigated German federal states) compared to Sweden and Berlin (the other German federal state under study). Due to differences in recording and reporting school attendance, it is not possible to compare the quantitative extent of the problem or trends regarding SAPs across the four countries based on the available official school statistics.

The right to education is a general feature of modern welfare states, and is declared in the United Nations' Universal Declaration of Human Rights (1948, Art. 26) and Convention on the Rights of the Child (1989).This is closely connected to an obligation to participate in regular education.Non-attendance has cast a shadow on school systems' ability to realise children's rights to education and their rights in education institutions.National school systems have different strategies for addressing school non-attendance, and there are different conceptualisations and definitions of non-attendance.
Internationally, school attendance problems have long been in focus (Reid, 2008;Ricking, 2003).The term school attendance problems (SAPs) has been proposed to describe a wide variety of phenomena related to absenteeism, such as school refusal, truancy, school withdrawal, and exclusion (Heyne et al., 2019).Non-attendance can be seen as a term covering all types of absenteeism from school, but there have been growing concerns that certain types of absenteeism and amounts of non-attendance can be regarded as problematic.These definitions of what is regarded as problematic differ widely.However, there is general consensus on the negative consequences of school non-attendance (e.g., Kearney, 2008;Keppens et al., 2019;Reid, 2012).In the growing body of literature on SAPs (see for example the breadth of references in Gren Landell, 2021), much focus is on individual risk factors for absenteeism while few studies have looked at the macro level and compared SAPs in different societies and education systems (Keppens & Spruyt, 2018).According to Kearney et al. (2019), school absenteeism as well as permanent school dropout have been identified as widespread global phenomena.Still, international comparative studies on SAPs are very rare.
The current knowledge on SAPs has been expanded in part by using the regularly collected data from schools (Gottfried & Hutt, 2019).However, how attendance and absence are registered and reported has received little attention in the literature so far (Bodén, 2016;Nakao & Yamamoto, 2007).In this article, SAPs are analysed through the lens of official statistics.We will study how attendance data is recorded in schools, reported to the education authorities and back to education providers and published for the public and actors concerned.This analysis is part of a wider project that, from an international comparative research perspective, investigates societal, organisational, and individual perspectives on SAPs in different national contexts (Stockholm University, n.d.).The countries for comparison have been chosen to reflect different education systems.Sweden was included as the project is based in Sweden.Germany and England share similarities with Sweden, but also differences when it comes to their education systems, 1 which may influence school attendance statistics (Keppens & Spruyt, 2018).Japan was of interest because SAPs have long been discussed extensively there (Horiguchi, 2018).
Possibilities to compare SAPs in different school systems are limited by insufficient knowledge on the extent of the problem (Fredriksson et al., 2023).Gathering systematic knowledge about systems for attendance control and (possibly different) practices of registering, reporting and publishing school attendance and absence can be seen as an attempt to understand what is regarded as problematic school attendance and a contribution to developing policies for addressing SAPs.
1 Information about the structure of these education systems is for example available via Eurydice at the European Education and Culture Executive Agency, a network by the European Commission, where education systems from countries that belong to the European Union are described and compared (https://eurydice.eacea.ec.europa.eu).

The Aim of the Article and Research Questions
The purpose of this article is to investigate which data on school attendance and absence are available in four different countries and to explore whether and how a comparison between school attendance statistics is possible.The following questions guide our analysis: Which indicators of school attendance and absence are recorded, reported and published in the four countries?
Can any trends on school attendance be seen in the four countries?Do attendance statistics define what is seen as problematic and how can school attendance and absence be compared between these countries?

Methodology, Comparative Approach and Structure of the Article
This article analyses statistics and official data collected by the relevant education authorities in the concerned countries.In line with the study's purpose, to facilitate a comparison between officially available data on school attendance, preconditions for this comparison need to be established.In the following subchapters we describe what kind of data on attendance and absence are available (see sections 3.1 to 3.4).Data can only be compared when what is collected is defined in the same way, or at least similarly.A description of how data are recorded, reported and published can contribute to understanding whether available data can be compared and what limitations need to be considered.After an introduction of the respective country's school system, we introduce how (national) school attendance statistics are recorded and reported and what terms are used.We focus on the situation of reporting data before the COVID-19 pandemic in order to rely on the typical case rather than the extraordinary case of the pandemic.Different strategies for closing schools or keeping them open complicates the comparison of absence dates.We provide examples on published data for the respective school systems and we investigate if any trends can be seen in the different countries.Ways of recording, reporting and publishing data suggests different understandings of what is regarded as the most prone problems of school attendance in the respective country.Possibilities of comparison of attendance statistics are summarised (see sections 4.1−4.2) and limitations of the study are carefully considered.

National Statistics on School Attendance and Absence
England and Japan have gathered national statistics for many years, while only a few federal states in Germany gather school attendance data systematically.In Sweden, there have only been occasional investigations.This guides the order in which the findings are presented.

England
The following description focuses specifically on England and not other parts of the UK (Scotland, Wales, and Northern Ireland).Primary school comprises six years (Key Stages 1−2, ages 5−11) and secondary school five years (Key Stages 3−4, age about 11−16).This is followed by two more years of secondary education or education through other education institutions until age 18.The English system can be characterised as a common core curriculum system.The overall responsibility for the education system in England lies with the UK Government's Department for Education (DfE), which since 2010 is also in charge of child protection.About 93% of all school children attend state-funded schools such as academy schools, community schools, free schools, foundation schools, and others.About 7% are enrolled in privately run, fee-charging independent schools.Home schooling is an option to fulfil compulsory education in England.The responsibility to ensure education for children of compulsory age for schooling (5−16 years) lies with the parents.After compulsory schooling until the age of 18, youth must attend either full-time education, apprenticeship, or traineeship, or work/volunteer while receiving a part-time education.The responsibility to support children at risk of missing education lies with the local education authorities (Eurydice, n.d.-b).

The Conceptualisation of SAPs
In England data on school attendance and absence are publicly available for statistical analysis, and can be used freely by researchers, public administration, or other stakeholders.There are weekly updates by the DfE on school attendance and absence that are published online on GOV.UK, and everyone can subscribe to regular information via E-mail.Information on attendance statistics and policy development are regularly communicated to policy makers and education administration, media officers, special advisers and other relevant actors (see for example list of pre-release access of attendance statistics in DfE, 2020b).Raw data is available for analysis.Reporting school attendance statistics has a high priority for addressing SAPs in England.

What Indicators of Absence are Recorded and Reported in England?
Schools are required to take attendance twice a day (DfE, 2020b).Data is collected regularly by education authorities who report to the DfE, and in a systematic way.Schools register attendance and report their data to the DfE, according to a detailed list of categorisations.Registration categories are: (a) present; (b) attending an approved educational activity; (c) absent; and (d) unable to attend due to exceptional circumstances.Altogether there are about 25 codes for different reasons for school non-attendance in the central school census (DfE, 2019, Annex 3), including codes for different forms of authorised and unauthorised absence. 2 Pupils can also be considered absent if they have been expelled from school.There is a great deal of data, available online, that allows analysing attendance for different subgroups, for example according to social and individual categories such as school type, city council, gender, ethnicity, eligibility to free school lunches and more (https:// explore-education-statistics.service.gov.uk/data-catalogue).This means that authorities in England can provide detailed follow-up not only on the number of absences but also on which children have missed school (and for what reasons).
In England, absence has long been registered centrally, and absence statistics are reported back to 1993.Figures with timelines are available back to the 2006/2007 school year when present definitions were established.The DfE provide official statistics on school attendance for state-funded primary and secondary schools and special schools.This information is based on pupil-level absence data collected via the school census (DfE, 2021b).Statistics are published by the DfE on their website three times a year, and data for the whole year is made public annually in March.These data have been used for timelines.
There are two main indicators in the DfE statistics: overall absence rates and persistent absence rates.Absence rates are 'the number of absence sessions expressed as a percentage of the total number of possible sessions' (DfE, 2020a). 3Overall absence is the sum of authorised and unauthorised absence.Attendance is taken twice a day; in other words, one session is equal to half a school day.Authorised absence is recorded illness and absence that was approved by the school in advance or satisfactorily explained.The persistent absence rate is defined as the rate of students who are absent for more than 10% of the half-day sessions they could have attended. 4 The persistent absence rate is based on overall absence.That shows that in England all forms of absence, both authorised and unauthorised are considered as possibly problematic.Definitions of what amount of absence is regarded as problematic has changed over time.In a document from 2011, it is stated that the persistent absence rate in England had earlier been recorded at 15% of missed sessions, and that prior to that the critical percentage had been considered to be 20% of sessions missed (DfE Press release, 19 October 2011).

Examples of Published and Publicly Available Data
The DfE in England provide information on trends over time regularly.Data for timeline comparison are available back to the school year 2006/07.For example, the 2 The list of reasons is recorded in the school attendance guidance, with ten pages explaining the various codes for registration.

3
The formal definition states: "The absence rate is the total number of sessions missed due to overall absence for all pupils as a percentage of the total number of possible sessions for all pupils, where overall absence is the sum of authorised and unauthorised absence and one session is equal to half a day" (DfE, 2020a).4 DfE (2020a).School attendance guidance for maintained schools, academies, independent schools and local authorities, August 2020.This document was withdrawn on 20 September 2022 and has been replaced by Guidance.Working together to improve school attendance.overall absence rate decreased from 6.4% during the 2006/07 school year to 4.7% in 2017/18 (DfE, 2020a).This means that pupils on average missed fewer sessions in recent years than they did a decade earlier.Both the overall absence rate and the authorised absence rate have been declining since then, while the unauthorised absence rate is more or less stable at around 1% (DfE, 2020a).
The persistent absence rate is defined as pupils missing school 10% or more of half-day sessions during the respective term or school year due to authorised or unauthorised absence.This persistent absence rate was 10.9% for all students in state-funded schools (DfE, 2020a).As the school year has 190 school days (380 morning/afternoon sessions), the figure of 10% is the equivalent of 38 morning or afternoon periods, which would be 19 school days (DfE, 2020b).
Overall absence can be categorised for different intervals.Table 1 shows how many pupils in state-funded secondary schools missed a certain amount of schooling according to DfE attendance statistics.These values allow us to see that 64% of students missed less than 5% and 13.6% missed more than 10% of their schooling.Both overall absence and persistent absence rates are published three times a year.Attendance statistics include reasons for absence and can be differentiated for school forms, grades, student backgrounds, and characteristics such as gender, ethnic group, English as first language, free school meal (FSM) eligibility, special education needs, and more.The publicly available statistical data online indicate different reasons for both authorised and unauthorised absences.Illness accounts for slightly more than half of all absences.

Japan
The education system in Japan is a single structure system, with no tracking during compulsory school.Six years of primary school (shôgakkô) are followed by three years of lower secondary school (chûgakkô) and three years of non-compulsory upper secondary school (kôtô gakkô) or alternatively six years of secondary school. 5Japan has nine years of compulsory schooling; however, about 98% of pupils who graduate from lower secondary school are said to continue to upper secondary school (OECD, 2021).Children start school at age six, after daycare (hoikuen) or kindergarten (yôchien).The central responsibility for education lies with the Ministry of Education, Culture, Sports, Science and Technology (MEXT).According to the OECD (2019), most decisions for school education are taken by the local education councils and on the regional level rather than the state or school level.Statistics, however, are available through MEXT for the whole country according to common definitions. 6 The School Basic Survey is published annually since 1948, providing fundamental statistics on all school forms.This census, gathered by MEXT through questionnaires to schools on the 1st of May every year, includes also a Survey on Out-of-School Children at School Age.Since 2003 information is gathered also online and nowadays most data is online available (The National Institute of Education − NIER, n.d.).To our knowledge, raw data are not directly accessible, although many reports have recently been digitally archived in a Web Archiving Project (WARP).There are many surveys conducted regularly including also timelines on education and attendance.The National Institute of Education has gathered and systematised many of these surveys (NIER, n.d.).

The Conceptualisation of SAPs
In Japan there is an extensive body of literature on SAPs focusing on what is called futôkô.Literally, futôkô means not being at school, and is sometimes translated as school absenteeism or truancy.The term has been used by MEXT since 1999.Futôkô has been regularly discussed as one of the most important 'problem behaviours' (mondai kôdô).As developed below, futôkô covers long term absence that is not due to illness or economic reasons.Intensive discussions are ongoing, and MEXT announced in 2016 that they want to change their approach to futôkô fundamentally, in order to counter prejudice against this phenomenon (MEXT, 2016a) in line with the law for ensuring equal educational opportunities (MEXT, 2016b).However, until today the term is still regularly used.

5
In this text, official translations according to the International Standard Classification of Education (ISCED) are used for consistency, although literature on education in Japan as well as the Japanese Government mostly apply American terms such as junior high and high school 6 MEXT was established in 2001 through a merging of the Ministry of Education, Science, Sports and Culture and the Science and Technology Agency.https://www.mext.go.jp/en/about /organization/title03/detail03/1375119.htm

What Indicators of Absence are Recorded and Reported in Japan?
For Japan, annual absence data are available for compulsory schools since 1966, building on attendance records kept by schools (MEXT, 2004).Annual statistics on school attendance are available from the School Basic Survey.Data build on absence recorded in schools and reported by the school leaders in a very detailed school survey to the MEXT.The school statistics include state, public, and private schools.Data can be accessed through the annually published School Basic Survey for different prefectures and cities, timespans, and grades.Absence is recorded according to four categories: illness, economic reasons, futôkô, and other.
Futôkô is the term in focus in all official publications, and is defined as involving 'those who have been absent for a long time because they cannot attend school for some reason other than "illness" or "financial reasons"' (MEXT, 2021).In other words, the term does not stand for overall absence but can neither be seen as equivalent to unauthorised absence.Since 1991, long-term absence is defined as 30 days or more per school year, 7 but data are also available for both shorter and longer intervals (1−3 days; 4−6 days; 7−13 days; 14−20 days; over 21 days; more than 90 days; and completely absent) (MEXT, 2020). 8ata on long-term absence are annually published by MEXT, Children and Students Division, Elementary and Secondary Education Division.Data are available for all years in compulsory school (Grades 1−9) and upper secondary school (Grades 10−12), differentiated by school organisation (state, public, and private) and prefecture.The Ministry has published several reports on school non-attendance that include both attendance statistics (for state, public and private school in different prefectures) and results from detailed school surveys, where reasons for the non-attendance are investigated for categories like for example bullying, family problems, tired of school and many others.There are also statistics from all prefectures on what kind of support the students with attendance problems have received (MEXT, 2019).

Examples of Published and Publicly Available Data
Japan is one of the countries where trends can be reported for lower secondary school students missing 30 days or more per school year, and timelines are often used to show the increase in the problem.There was an increase in futôkô from 1% in 1991 to over 4% in recent years, but with alternating periods of increase, stable values, and moderate decrease.The trends for primary school were similar but at a much lower level (see Table 2).7 Earlier long-term absence was defined as 50 days or more.The development of statistics since the 1950s is well described in Shimizu (2011) and Horiguchi (2018).These articles also include timelines (see more in detail Kreitz-Sandberg & Lesch, 2019).In Japan, information on absenteeism, or specifically futôkô, is published on a regular basis and often makes the news.The focus is typically on futôkô rather than overall absence.Frequently, long reports are published with data from the School Basic Survey and other survey studies.In 2019, 1.88% of compulsory school pupils (grade 1−9) were registered as futôkô long-term absent. 9This prevalence had increased over previous years (MEXT, 2020).

Germany
In Germany, the federal states (in German: Bundesländer, in English texts sometimes: länder) are responsible for organising school education, and the organisation can vary substantially between different federal states.In all federal states there is some type of differentiated lower secondary education.After nine years of fulltime schooling, at the upper secondary level, pupils can choose between compulsory full-time schooling and compulsory part-time schooling in relation to vocational education (Eurydice, n.d.-b).Public schools are free of charge, and the funding of public schools (primary and secondary education) is divided between the federal states and municipalities (Federal Republic of Germany, 2018, 2019).Full-time compulsory education or training ends at the age of 18 or 19 depending on the state (Eurydice, n.d.-a).
In the 16 states we find different albeit parallel types of secondary education from the age of about ten years.For example, in the state of Berlin, primary school lasts six years.After this, pupils continue their education either in integrated secondary school (ISS, Gemeinschaftsschule) or the academic track, grammar school (Gymnasium).There are also various branches of Special Needs Education (SNE) schools.The school system in Thuringia in southeastern Germany is more complicated, with many parallel branches: community school (Gemeinschaftsschule, Grades 1−12), primary school (Grundschule, Grades 1−4), grammar school (Gymnasium, Grades 5−12), comprehensive school (Gesamtschule, Grades 5−9 plus 10−13), basic secondary school (Regelschule, Grades 5−9), SNE school (Förderschule, Grades 1−9), vocational college (Berufbildende Schulen, after Grade 9; up to three years), and college for adults (Kollegschule) (Thüringen, 2019a).9 The values are 3.94% of lower secondary school pupils (Grades 7−9) and 0.83% of primary school pupils.

The Conceptualisation of SAPs
In Germany there are no centrally gathered statistics on school attendance.The Standing Conference of the Ministers of Education and Cultural Affairs of the Länder in the Federal Republic of Germany (Kultusministerkonferenz − KMK) is responsible for coordinating between the education systems in different states (KMK, 2020).However, although KMK gathers a great deal of statistics, these do not include information on school attendance or absence.We wrote to the responsible ministries in all 16 states, most of which provided no applicable information on school attendance.Only four states − Berlin, Hamburg, Hessen and Thuringia − provided some statistics.This shows that school attendance is not transparently documented in most of the federal states.

What Indicators of Absence are Recorded and Reported in Germany?
Digital reporting is not yet the rule in German schools.It is common practice for school attendance to be recorded by the teacher for each lesson in what is called a Klassenbuch, a form book in which the content of teaching, homework, and special occasions is noted and which also includes the class register on a daily and lesson basis.Some schools have changed to digital Klassenbücher.Below we will present the data from the two federal states that provided the most detailed information to our inquiry, Berlin and Thuringia.The Senate Department for Education, Youth and Science in Berlin and the Ministry for Education, Youth and Sports in Thuringia, both are using the term distance from school (Schuldistanz) to investigate attendance problems (Berlin, SenBJF, 2015;Thüringen, 2013).
In Berlin, data on absence are regularly reported twice a year and discussed (Berlin, 2015(Berlin, , 2019a)).At the end of each term, schools report overall absence (Fehltage insgesamt) as well as unauthorised absence (Fehltage unentschuldigt) to the education authority, the Senate Department for Education, Youth and Family, which follows up with information back to the schools.
Overall absence is defined as the proportion of absent days in relation to all teaching days (Fehlquoten).Unauthorised absence is not explicitly defined in the available information.Absence is registered for each lesson, and six missed lessons during a school term are considered one (additional) day of unauthorised absence since the 2018/19 school year.Late arrivals (Verspätungen) are also registered, according to teachers' reports in the class register.The proportion of students with more than 20 days of absence per term is highlighted (slightly over 20% absence, which we might call the local definition of persistent absence in Berlin).Data are also published for the following categories: no school days missed, as well as 1−10, 11−20, 21−40, and over 40 school days per school term missed (Berlin, 2019a(Berlin, , 2019b(Berlin, , 2020a)). 10 10 There are two school term per year.The first term is from August to December and the second term from January to the beginning of the summer holidays.Due to rising numbers, the 1−10 In Thuringia, data on school attendance are recorded in all schools, reported to the education authorities.Data are regularly summarised and distributed by the Ministry of Education, Youth and Sports in Thuringia (Thüringen, 2019b(Thüringen, , 2020)).The data from Thuringia covered only unauthorised absence, and a certain number of days.The tables show the percentage of those who have missed a certain number of school days.Data are presented for different regions, school types, and grade levels, and are recorded for the following ranges per year (and not per term, as is done in Berlin): 1−5, 6−10, 11−20, 21−40, and more than 40 days.The 11−20 day amount is called beständiges Fehlen, which can be translated as persistent absence.There is a timeline for the absence data for the last ten school years (2010/11 to 2019/20; see Thüringen, 2019bThüringen, , 2020)).
For Berlin as well as Thuringia, no data are provided for different student category features such as gender, socioeconomic status (SES), ethnicity, or other categories.

Examples of Published and Publicly Available Data Example 1 − Berlin
The statistics can provide general trends of absence rates in Berlin in recent years.In some districts these rates are rising, while they are stable in other parts of the city (Berlin, 2020a(Berlin, , 2020b)).Table 3 provides rates for overall absence and unauthorised absence for 2019.Information material that is, for example, distributed to schools also provides tables or graphs for the 12 different districts of Berlin and for different school types for three consecutive years (Berlin, 2020b(Berlin, , 2021)).It illustrates that non-attendance is more frequent in certain districts and grades (grouped for Grades 5−6 and 7−10) as well as school types.Raw data are not publicly available.authorities also shows that there are clear differences in absence rates between the different school types, with very low numbers in grammar school and higher numbers in basic secondary school and community school (Thüringen, 2020(Thüringen, , 2021)).Data and some timelines are available for the 2010/11 to 2019/20 school years.11During this decade, the percentage of students with unauthorised absence has risen.The strongest increase is among the group skipping one to five days, but the number of students with longer absence has also risen.The numbers seem to be on a lower level compared to Berlin, however.A great deal of statistical data is available online for Thuringia, and after enquiry to the statistical office attendance statistics (unauthorised absence) could be accessed on the school level.Their policy is to record, report and analyse their data in order to be able to work with prevention and successful interventions (Thüringen, 2013).

Sweden
Sweden has a decentralised education system.Its central government, through the Ministry of Education and Research, has the overall responsibility for the education system, while the municipalities and independent providers are responsible for implementing educational activities, organising and operating school services, allocating resources, and ensuring that the national goals for education are met.Education is compulsory for ten years, starting with preschool at the age of six and followed by compulsory school (Grades 1−9).As its organisational model, Sweden follows a single structure education (Eurydice, n.d.-a).The school system is decentralised and both public and independent actors provide education.All costs are carried by the state and the municipalities.A great majority of students continue to upper secondary school (gymnasieskola), which is differentiated between various academic programmes and vocational/professional-oriented ones.12

The Conceptualisation of SAPs
Discussions about non-attendance have intensified in Sweden in recent years.The National Agency for Education has re-evaluated the categories for reporting absence, and now works more systematically with gathering absence data.Public and independent school providers seem to be gradually becoming more aware of the importance of actively gathering, reporting, and following up on attendance data.
The concepts of giltig frånvaro (authorised absence) and ogiltig frånvaro (unauthorised absence) are used in the Swedish Education Act (SFS 2010:800), but without clear definition (Skolverket, n.d.).This means that teachers or principals themselves must define what should be considered authorised reasons for absence, possibly leading to variation among schools (Skolverket, 2021).Examples of authorised absence include illness and non-attendance that has been authorised by the school in advance.Another term used in the law is upprepad eller längre frånvaro (repeated or long-term absence).
A governmental investigation (SOU 2016:94) into SAPs addressed the public, education actors, and politicians.It introduced the term problematisk frånvaro (problematic absence) and stated that all kinds of non-attendance (both authorised and unauthorised) can be problematic, especially regarding students' participation and performance in school.

What Indicators of Absence are Recorded and Reported in Sweden?
In Sweden, school attendance data are not regularly or centrally reported, although some systematic data on school attendance have been collected on certain occasions (Skolinspektionen, 2011(Skolinspektionen, , 2016;;Skolverket, 2010Skolverket, , 2014Skolverket, , 2021)). 13In 2015, the Swedish School Inspectorate conducted a nationwide survey based on estimated numbers from principals or school administrators.It distinguished between unauthorised continuous absence for at least one month and repeated unauthorised absences exceeding 5% of the teaching time during the previous two months.No data on authorised absence were collected (Skolinspektionen, 2016).
The most recent study by the National Agency for Education (Skolverket, 2021) analyses data reported by municipalities and independent school providers for the autumn term 2019 and the autumn term 2020.These data were based on what had been recorded either manually or by digital register systems.A majority of the municipal schools and about two-thirds of the independent school providers used digital systems for registration.Only 60% of these actors answered the survey, but the reported data nevertheless covered close to 80% of compulsory school pupils.A positive change from the prior national investigations was that data on overall absence, thus also including authorised absence, were collected.
Data are not collected regularly, in a systematic way using the same definitions by the National Agency of Education.Thus, it is not possible to report trends or timelines for absence in compulsory schools based on official school statistics, as data were gathered differently and according to different definitions.An investigation on gathering absence data nationwide on a regular basis was recently presented to the Swedish Government, and there will be a decision in this regard in the near future.

Examples of Published and Publicly Available Data
Data published in Swedish studies usually derive from occasional studies.Data from 2021 on unauthorised absence were presented according to three intervals: students with 5−14%, 15−29%, and 30% or more of unauthorised absence.Information on overall absence was reported according to the categories 15−29%, 30−49%, and more than 50% (Skolverket, 2021).Overall absence increases with age, and is also the most frequent in lower secondary schools.As in other countries, in Sweden unauthorised absence is the most common in lower secondary school; absence rates rise from Grade 7 to Grade 9 (see Table 5).During autumn term 2019, a fourth of eighth and ninth graders and about a fifth of seventh graders had missed more than 15% of their overall schooling (Skolverket, 2021).This report is outstanding for Sweden, as it reports not only unauthorised absence but also overall absence for the first time.Skolverket (2021) presented only the most recent data for autumn 2020 when schooling was shadowed by the COVID-19 pandemic.These data are however not representative and with courtesy of Skolverket we present here also unpublished data for autumn term 2019, before the pandemic.The data show that there is a high percentage of students who miss big parts of their compulsory schooling.Over 20% of students in secondary school missed so much of their overall schooling, that they in some other school systems would be called persistently absent, although Sweden does not use this term.On top of that, there is reason to believe that a substantial proportion of students' absence is not registered.The actual numbers in Sweden are likely even higher than those reported, as principals express doubt that long-term absent students and those with reduced school days are consistently registered when they are absent (Skolverket, 2021).The report also shows that two to four percent of the lower secondary school students had more than 15% of unauthorised absence, a value that seems to be similar to the values for futôkô in Japan, and which is regarded as very alarming there.However, we need to keep in mind that definitions are not identical and recording and reporting of data differs.Possibilities of comparison will be discussed below.

Comparative Analysis of Indicators of Absence Recorded in the Four Countries
In the following, differences and similarities among the four countries investigated here will be discussed in order to analyse and compare how information on recording, reporting and publishing school attendance data can be compared among these countries.This comparison builds on the data that are published and publicly available on school attendance and absence.

Which Indicators of School Attendance and Absence are Recorded, Reported and Published, and Can Any Trends Be Seen on School Attendance in the Four Countries?
All countries have systems where teachers report attendance and absence either for the studied lessons, half-days or days.This is sometimes done on paper, sometimes digitally.The main difference seems to be whether and how the data are reported to the education authorities.There are strong differences, how absence data are summarised, specifically with regard to which form of absence is recorded, reported and published.Another difference is the way in which the collected and analysed data is shared with the schools and the public.We will in the following highlight possibilities and limitations of systematic comparisons of attendance data.This study of attendance statistics in four countries illustrated that two types of variables were useful in understanding school attendance and absence: overall absence rates and long-term absence rates, sometimes called persistent absence.They served different functions, and the available data reflected different national and local practices with regard to which data were gathered and analysed.We will also argue that what is being registered is related to which form of absence is regarded as problematic.Table 6 illustrates which data are provided.The absence rate is useful for education planning on a macro level and for observing the extent of the problem in the overall school system.The absence rate represents the average absence of all students during a defined time span, either a school year or a term.If absence rates are provided, the values can be calculated for both overall absence and un/authorised absence.Berlin provided absence rates in internal documents but only England published systematic data for overall absence.Although it is common that districts and school report average daily attendance it is know that this indicator may mask high levels of persistent absence of students.It is recommended that local, state and federal governments make sure that existing data is being used to monitor and identify chronic absence, starting already in kindergarten (Bruner et al., 2011;Gottfried & Hutt, 2019).It is not known to what extent the studied countries use collected data for the purpose of planning education and preventing school attendance problems.
The persistent absence rate is useful for capturing how many, or which, students are at risk of missing large parts of their education and closely related to what is being described as problematic school non-attendance.England was the country where this term was clearly defined, but some form of long-term absence rate was reported in all the studied cases.In England, persistent absenteeism is defined as more than 10% (approximately 19 school days a year) overall absence, including both authorised and unauthorised absences.This level was set just a few years ago; until 2011 it had been 15%, and earlier 20%.In Japan, 30 days per year is the most common definition of problematic school non-attendance (called futôkô), which is equivalent to 15% of absence due to other reasons than illness or economic explanations.Until 1998 the critical level, then called school refusal (tôkô kyohi), was set at 50 days or 25% (Horiguchi, 2018).For these two countries, there are timelines over the development of persistent absence.
A recurrent question is whether there has been an increase in SAPs.In Sweden, for example, there is an ongoing discussion of whether changes in the national curriculum have contributed to worsened learning conditions, especially for students with special educational needs, thus increasing rates of SAPs.With a wide definition of SAPs and unclear rules according to which intervals school non-attendance should be reported it is obvious that such a question is difficult to answer.Looking only at statistical evidence, regularly collected data, and use of the same definitions over years are needed to answer this question.For Sweden obviously, there is no statistical evidence regarding whether school attendance has risen or declined over time, due to a lack of systematically gathered data.Timeline studies would be needed for this.In Thuringia there are timelines for 2010/11 to 2019/20, and in Berlin at least for the last few years.It is interesting to see that there are timelines with stable definitions for Japan back to 1991 and for England from 2006.In England a decline of persistent absence rates has been reported for the years before the pandemic.However, for Japan, there has been in increase of long-term absence (futôkô), that is very visible in the national attendance statistics of the last years.General trends across school systems cannot be seen.It is not clear from the investigation of the countries whether or how data are used by national authorities to decrease persistent absence.

Do Attendance Statistics Define What Is Seen as Problematic?
Definitions of what is defined as persistent absence or described as critical level of long-term absence vary widely.Sometimes persistent absence is measured in relation to overall absence (England, Berlin, and in certain statistics in Sweden as well), and sometimes in relation to unauthorised absence or parts of absence without illness (Thuringia and Japan; see Table 6).These differences in registration practices illustrates that concepts of what is regarded as problematic absence vary widely between countries.For comparison purposes, it would be necessary to record and report data according to generally agreed criteria and it is most helpful if a percentage (and not only the total number of students) is published.
The analysed data for Germany shows strong variation among different federal states.In both studied states, persistent absenteeism was calculated in relation to unauthorised absence.In Thuringia, however, the critical level is set at ten days per year while in Berlin it is 20 days per term; in other words, the stated level for problematic absenteeism in Berlin is four times higher than in Thuringia.In Sweden there is no official definition of what is considered problematic absence.The term persistent absence is not used in Sweden, and the term längre och upprepad frånvaro (longer and repeated absence) has no clear definition (Regeringskansliet, 2016).In Sweden, prior investigations have only asked for unauthorised absence (Skolinspektionen, 2016).However, the latest investigation reported overall and unauthorised absence.It is stated that all forms of absence are seen as problematic (Skolverket, 2021), but 20% of overall absence is often taken as the point at which to take action.
This article has illustrated how the way of recording, reporting, and publishing data is connected to school attendance in specific ways in the different school systems.How SAPs are conceptualised is related to what kind of data are made available and published.This broad variety in definition leads to statistics for the four studied countries and federal states are generally not comparable with one another.There is no common definition of which amount of absence is seen as problematic.

How Can Data on School Attendance and Absence Be Compared between the Four Countries?
This study has shown that there are many limitations with regard to how absence is recorded, reported and published.Still, it is of interest to try to relate values from the different national or official reports to each other.With the available data, we tried to calculate the percentage of lower secondary school students who missed more than 15% of school days or sessions.
For England and Sweden, there are some data for long-term absent students based on overall absence.In England, 7.1% of secondary school students missed 15% or more during the 2018/19 school year (see Table 1).In Sweden, the values for overall absence of 15% or more were 19% in Grade 7, 23% in Grade 8, and 27% in Grade 9 in the autumn term of 2019 (see Table 5).The numbers indicate higher quota of overall absence from compulsory secondary schools in Sweden compared to England.Still, it is difficult to compare these numbers due to different ways of gathering the data.In England, students in special schools and home-schooled children are for example not included.It is also difficult to compare the values from Sweden (age 13−16), as we presented the average for secondary school students for England (age 11−16).
Respective quotas for unauthorised absence are available for Japan, Sweden, and the two German states.In Japan as of 2019, the rate for missing 30 days or more per year is equivalent to 15% or more of the school year, and was 3.94% for lower secondary school pupils (Grades 7−9, age 12−15, see Table 2).In Sweden the values for unauthorised absence of more than 15% was 2% for students in Grade 7, 3% for students in Grade 8 and 4% for students in Grade 9 in 2019 (see Table 5).In Germany we can calculate the category for 20 days or more per year.In Berlin 3.6% of pupils in Grades 7−10 missed school this much (see Table 3).In Thuringia this would be called 'massive absence', and is reported for 1.6% of pupils in Grade 8 (13−14-year-olds), 0.8% in Grade 9, and 0.2% in Grade 10 (see Table 4).The lower numbers for Thuringia are in line with results by Keppens and Spruyt (2018) who found that the reported absences in differentiated secondary school systems is lower than in comprehensive and single structure systems.Against the background of the available data, it is not possible to give clear answers as to whether the apparent differences in values are related to ways of recording and reporting data and to locally different definitions, or to realities in school and welfare state systems.
School non-attendance is distributed differently among various age groups in the different countries.In all the countries, absence values are the highest in lower secondary school; however, there are differences regarding the peak.In Sweden and Japan, percentages increase up to Grade 9 (15−16-year-olds), while in Germany the highest values are reported in Grade 8 (13−14-year-olds) and are lower for pupils in the final grades of lower secondary school.This triggers an interpretation that it is not a question of individual development but rather conditions in society and in the education system that influence patterns of attendance.Thus, data on absence might be used for analyses and systematic work in order to improve the learning environment and strengthen school attendance.
Understanding attendance statistics is a precondition for knowledge about the state of the problem, and can be an important starting point for policies supporting attendance and the following up of interventions (Gottfried & Hutt, 2019).This article has contributed to understanding how differently the four countries studied have approached these questions.In the presented statistics, we could see higher numbers in secondary school than in primary school; in England and Germany, the statistics showed different values for different school types; and in Berlin, for example, the data displayed varying patterns depending on where the school was located.In England, absence rates can also be connected to group variables, such as students who are eligible for free school meals or those who have different ethnic backgrounds.
A number of challenges have been identified.Firstly, not all absence and attendance is registered.It is of great concern that authorised absence is not reported in most of the countries studied here.Recurrent or prolonged authorised absence constitutes a risk for negative consequences in the same manner as unauthorised absence does (e.g.Bruner et al., 2011).Differences exist between as well as within the countries discussed here.The data from the two federal states in Germany present what seem to be various realities regarding reported attendance statistics, as well as different perspectives.The use of different definitions has been reported elsewhere, for example regarding truancy in the different states in the US (Gentle-Genitty et al., 2015).In Japan the number of persistently absent pupils is rising, while in England it is slowly decreasing, according to timelines presented in the official statistics before the pandemic.In order to be able to compare data between different locations access to raw data would be very helpful.To the best of our knowledge, among the countries discussed in this article, statistical data are only publicly online accessible for England.Timelines could be used to study changes in school systems within countries and between countries as well, if national and respective education authorities would investigate attendance according to the same definitions and with stable ways of recording, reporting, and publishing data.

Limitations and Recommendations for Further Studies
This study has a number of limitations; for instance, we have not systematically gathered information on how the attendance data were collected, recorded, and used in the four countries for a defined equivalent group.The groups of students in lower secondary school included in the national and public statistics were not equivalent in all school systems.The study included information for school systems with a common core curriculum (England), single structure system (Japan), differentiated lower secondary education (Germany, specifically Thuringia), and decentralised single structure system (Sweden).In Sweden, Germany and Japan there is compulsory schooling, in England there is compulsory education, with possibilities to home education.For home schooled children no attendance data is available and home schooling may sometimes be used as an alternative to overcome school distance (Myhill, 2017).
We have tried to understand existing data, and argue that more data need to be collected and made more comparable, through both official statistics and other standardised studies.Besides this, in order to understand the importance of a school system's organisation for strategies regarding school attendance more qualitative studies are needed.Specifics of the school system are related to specific limitations.For example, in England, students who are enrolled in private, non-statefunded schools or are home-schooled are not included in the attendance statistics.Although schools teach according to a common curriculum, school organisation and the gathering of statistics vary.Another example is students who are new to the country.In descriptions of the statistics from Berlin, it is mentioned that in so-called 'welcome classes' for newly arrived children with a mother tongue other than the school language, attendance is not registered.For Sweden, the available statistics were gathered retrospectively through school leadership questionnaires.There are currently developments underway that may in the future provide more comparable information throughout the Swedish school system and hopefully also across different countries' school systems.Systematic information about registration processes is limited, a factor that complicates comparisons between different settings.
Additionally, registration varies within countries.An earlier study from Japan showed variations in the registration of absence between schools and prefectures (Nakao & Yamamoto, 2007): Absence registered at one school as futôkô might be registered at another as being ill.A study in Sweden showed that the introduction of digital registration of attendance had effects on educational settings in school and that it needs to be further studied (Bodén, 2016).Generally, information on how data on different categories of absence had been gathered and were used in the school system − with the exception of England − was not sufficiently described in the available absence statistics.
Another limitation becomes obvious when authorities try to limit authorised absence.In England, for example, authorised absence for holidays decreased from 0.6% to 0.1% between 2006/07 and 2018/19 while the percentage of unauthorised absence for holidays increased from 0.1 to 0.4% (DfE, 2021).This means that absence for holidays was about the same, but with a shift from authorised to unauthorised.Thus, although the value of reliable statistics became evident in this article, it also seems that the processes for gathering these statistics need to be developed with care (Bodén, 2013) and that the pedagogical consequences need to be observed and studied.
Further studies would benefit from an analysis of statistics, working with original data and analysing them according to common criteria.Some countries have such digital data, but to our knowledge only the data in England is freely accessible to the public online.In some other cases, such as Japan or a few German states, data are available and might be accessible for research if one contacts the relevant agencies.While access to raw data would allow for more detailed analysis, as long as the data are gathered differently the possibilities for comparison are limited.

Conclusion
In order to combat the global problem of SAPs, countries could benefit from learning from each other with regard to registering school attendance and absence.Comparisons will only be possible with equivalent problem definitions.Thus, there is a need for more consistency in the use of attendance data between countries.Different countries have different ways of recording and reporting statistical information on absence, and different ways of publishing relevant information.Some countries − in our sample, England and Japan − have developed a system for collecting and disseminating information about school absence on a regular basis, while other countries have no national system (Germany), or collect national data occasionally on a non-regular basis (Sweden).When data are collected, there are attempts to set some level of absence that is considered to be problematic.There seems to be a lower threshold for how much absence is considered problematic in England, and Thuringia compared to Sweden, Japan and Berlin.It is not possible to establish a similar trend across the four countries.If a comparison were to be made between the levels of absenteeism and the trends concerning them in different countries, it would be difficult to rely solely on national statistics; it would instead be necessary to either collect specific data in international surveys designed to do this, or explore whether other international studies contain this information.This study elucidated a need for more descriptive statistics as a precondition for developing strategies for reporting and − hopefully − improving school attendance.The potential offered by using attendance data from schools together with data from research and surveys involving, for example, children's well-being and bullying, school performance, dropout, and graduation should be further explored.Finally, we believe that qualitative studies are necessary in order to understand how statistics can be used.How do actors in the school react when SAPs are detected?Developing meaningful systems for recording, reporting and publishing attendance and absence statistics is important in connection to the central question of how the data can contribute to combating and preventing SAPs.

Table 1
Absence rates for overall absence in secondary schools in England during 2018/

Table 2
Long-term absence futôkô in Japanese lower secondary schools (% of students)

1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2019
Information for futôkô, percentage of students with absence for 30 days or more per school year due to reasons other than illness or economic reasons.Source: MEXT 2021, p. 15.

Table 3
Students with overall and unauthorized absence (in %) in Grades 7−10 in Berlin, autumn term 2019, all school types.

Table 4
Students with unauthorised absence (in %) by school grade in Thuringia (school year 2019/20) Note.The table presents how many percent of secondary school students had a certain number of unauthorised absent school days during a full school year.Source: Data provided by the statistical office of Thüringer Ministerium für Bildung, Jugend und Sport, school year 2019/20.

Table 5
Overall absence and unauthorized absence in Swedish lower secondary schools (Grades 7−9), autumn terms 2019 and 2020

Students (in %) − autumn term 2019 (autumn term 2020) Grade Overall absence 15−29% 30−49% 50% and more Sum (15% and more)
Note.The data for autumn term 2020 covers the period of the Covid-19 pandemic, that is why unpublished, rounded data for the autumn term 2019 was added.Data for compulsory education (including lower secondary schools, schools for indigenous/Sami, and special schools) based on a survey to school leaders; data cover about 80% of students in this group.Source:Skolverket  (2021, p. 25−26)for unauthorized and overall absence 2020; unpublished statistical data provided by Skolverket for 2019.

Table 6
Availability of school attendance statistics in four countries Note.This table summarizes the information presented in this article by above cited sources.For Germany, Berlin and Thuringia are used as examples.