Stéphan Vincent-Lancrin
OECD
Carlos González-Sancho
OECD
OECD Digital Education Outlook 2023
Towards an Effective Digital Education Ecosystem
OECD Digital Education Outlook
Abstract
This chapter highlights the importance of a system-level digital tool: student information systems (also known as education management information systems – EMIS). A new generation of student information systems brings together information about each student and their learning trajectory within an education system. This opens the possibility to use the information in real time for some educational decisions and to personalise learning. The chapter presents the common functionalities and uses of these systems within countries, and shows how a next generation of systems could become a key element of an effective digital education ecosystem.
A student information system is a digital tool that collects and gives access to detailed information about students, including demographic information, school attendance and pathways, and increasingly their learning outcomes. Contrary to a mere student register, student information systems facilitate access to these data through a range of reporting, visualisation and analysis tools so that they have value for stakeholders in real time. Usually, different stakeholders have access to different kind of information about students, schools and possibly teachers, depending on their role. People with no role in the system can typically not access them. While some of these information systems are typically used for administrative purposes and fed by administrative applications, they are also ultimately used to generate education statistics. Their first value in a modern digital education ecosystem is to provide real-time and actionable information to stakeholders based on the information gathered at the entire system level (González-Sancho and Vincent-Lancrin, 2016[1]).
Student information systems are also referred to as Education Management Information Systems (EMIS), State Data System, State Longitudinal Data System (when longitudinal), or Education Information System. Most of the time, these information systems are not limited to student information. They always include information about schools (and have school unique identifiers) and sometimes teachers as well. Calling them “student information systems” showcases that the current generation of systems usually maintains large data about individual students and that the rest of the information can be connected to students, which makes them “student-centred”: most of the time, they also include a lot of data that are not focused on students only, but only linked to them. In this chapter, we will use interchangeably the terms student information system or education information systems (even though the later has a broader connotated scope).
In this report, student information systems always refer to digital systems that are maintained at the system level – as opposed to the institutional/school level. Depending on the country, the boundaries of the education system may be national or correspond to a state/region or a local educational authority. Schools may also use applications called “student information system” or “school information management system” to keep information about their students, including their demographics, their class, their schedule, their parent’s information, etc., but for ease of language and consistence over the report, these school-level student information systems will be referred to as learning management systems (Vincent-Lancrin, 2023[2]).
Longitudinal education or student information systems represent a particular class of information systems. In this report, longitudinal student information systems are defined as a systems which: a) connect student-level data collected at different points in time; b) maintain extensive information about students, including about their learning and attainment outcomes, their schooling pathways, and their socio-demographic backgrounds; and c) facilitate data access and data use through a range of reporting, visualisation and analysis tools (adapted from (National Forum on Education Statistics, 2010[3]).
These three basic features are minimal conditions for education information systems to increase their capacity to support more policy relevant and innovative uses of education data, that is, uses that generate a dynamic rather than static view of students’ experiences and outcomes, that provide a solid basis for establishing claims about the effectiveness of policies and practices, and that result in more timely and actionable feedback for multiple stakeholders in education. This is why longitudinal information systems will be of particular interest in this chapter.
The chapter is organised as follows. The first section presents the prevalence of longitudinal information systems within countries and the main data elements and functionalities of these systems. The second section presents different possible uses of these systems, from a statistical or research use through predictions that help personalise learning. Finally, it reflects on different possible ways of using the information collected by these systems so that it can make the information actionable for teachers and school leaders.
Our comparative analysis covers 29 countries, including two sub-national systems (Belgium) and one OECD accession country (Brazil). All OECD countries were asked to answer the questionnaires. Countries answered a questionnaire, whose answers were validated through interviews of the responding team with the OECD Secretariat. The answers were then double checked and expanded by describing countries’ digital education infrastructure and governance (OECD, 2023[4]). Information about other countries is also included when relevant, even if they did not take part in our comparative study.
Most OECD countries have a longitudinal student information system (Table 2.1). Out of 29 countries/systems, 19 (or 65%) have a national/central student information system and 3 federal countries have all or most of their states or regional authorities maintaining a student information system at that level. (For example, all US states and Canadian provinces have a student information system.) Moreover, even though they do not maintain the information through a student information system, four countries collect granular student information as part of a central student register. For a few countries, the system have either a limited scope (for example, SISTEC covers vocational education and training in Brazil) or lacks the integration that a typical information system provides, so that existing applications allow to bring together basic student information, but not in a straightforward way (for example in the French Community of Belgium)). A small number of countries do not have an information system yet, even though they regularly collect information from schools or regions for statistical purposes (for example Japan or Czechia) (Figure 2.1).
Note: N=29. SIS: Student information system (or EMIS).
Apart from Mexico, federal countries do not have a central information system or a central student register with student-level micro-data. Typically, states or regions report their information to the federal level: in the United States, states are mandated to provide a certain set of data to the National Center for Education Statistics; in Canada, the Council of Ministers of Education of Canada (CMEC) collects national education statistics in aggregate format through a yearly survey sent to provinces; in Brazil, INEP collects statistics from schools directly (but did not maintain a longitudinal information system or student register as of 2023).
Except in a few cases (e.g. Lithuania and Hungary), where a vendor solution is being used in conjunction with an owned government tool, all those education systems are owned by the government or public authority that manage them. In some cases, this is the customisation of existing commercial solutions; in most cases, the systems have been designed with external companies for the ministry or government agency that owns the tool. The main advantage of ownership lies in the sole ownership (and usually custodianship) of the rich data that are collected. A possible disadvantage lies in the maintenance of the system, which can become obsolete or less performant if no regular investment is made to improve the quality of the reporting and visualisation tools (if any).
Student-level longitudinal tracking
Almost all countries maintain longitudinal student information systems or national student registers: 22 out of 29 countries, that is, 76%, have longitudinal data about their students, either at the national or sub-national levels (Figure 2.2 and Table 2.1). Moreover, 26 countries (90%) can use student-level information to inform their policies and practices, either at the national or other governmental level. Collecting longitudinal information is an essential feature for a modern education information system.
Note: N=29.
Unique, permanent identifiers that remain with students, teachers, and schools across data collections carried out at different points in time and, eventually, by different agencies or organisations are essential elements of longitudinal information systems. These longitudinal identifiers make it possible to connect data over time and thus trace sequences of events and individual trajectories in the education system. Unique identifiers are also required to match data entities into nested structures, for instance students within classes or schools.
Longitudinal information systems are thus able to follow individual student trajectories over their time in the education systems and, possibly, beyond. At the other end of the spectrum, information systems with a cross-sectional design maintain data collected at one or several points in time and about a single or multiple student cohorts, but lack the capacity to match individual-level records across time.
Unique, permanent identifiers for each individual student are typically assigned to students as they are first recorded in the system, typically as the result of their first school registration process. It usually take the form of numerical or alphanumerical codes. Permanent identifiers remain with students across subsequent data collections, ideally spanning the entire academic history, from early childhood to post-secondary education.
Individual-level identifiers may be specific to a single information system or agency, or shared for use across multiple data systems and agencies. Examples of shared identifiers include social security and ID-card numbers used across government registries, and education numbers that remain with students across their experiences at levels, jurisdictions and schools within the education system. Shared identifiers help establish students’ identity unambiguously, facilitate collating data from different sources and reduce the burden of new data collections. However, they raise the stakes for privacy and confidentiality protection, as they are a key to revealing more personal information than identifiers specific to a single database. For example, in Estonia or Luxembourg, students have their national identity number as their unique identifier. In France and in New Zealand, those unique identifiers are by law different for all sectors of society because of privacy concerns. In the first case, it is easier to link all data collected by different administrations, but this requires more trust in the government; in the second case, it is more cumbersome (though not impossible) to link information from different sectors, and the privacy risks are lower.
Permanent identifiers for aggregate-level data entities such as schools and districts are a common feature of education information systems supporting monitoring and evaluation efforts at the school and system levels. However, it is the availability of unique longitudinal identifiers at the individual level – for both students and educators - that constitutes one of the building blocks for more innovative uses of education data. All student information systems maintain school identifiers that support the linkage of school-level information over time – a question that was not asked formally as a previous survey showed that all education systems (surveyed) already have such school level unique identifiers (González-Sancho and Vincent-Lancrin, 2016[1]). All systems therefore have the capacity to describe and compare trends at the system and school levels.
Almost all student information systems in the countries for which we have information cover students in primary and secondary education (including vocational education and training (VET) when provided as part of the school system). In a few cases, there are still different student information systems per level of education: for example France and Luxembourg have two different systems for primary and for secondary education. The coverage of those student information systems depend on the country: in some cases, they cover all students regardless of whether they are enrolled in public or private schools, notably when private schools receive funding from public authorities; in other cases, the systems only cover students enrolled in public schools. The main reason is that those systems inform or are the result of administrative tasks, for example informing school funding, teacher allocation, examination planning, learning assessments, school attendance obligation, etc.
When those systems are multiple, they should be interoperable. In the case of not connected or not interoperable systems, having a unique identifier enables linkages between the data collected by the different systems, but the procedure can take some time and be cumbersome.
Student-level data elements and linkages
The second baseline feature of longitudinal student information systems is the availability of extensive student-level information, ranging from information on the schooling and socio-demographic contexts that they experience to indicators of their academic performance and learning outcomes, when possible.
Longitudinal information systems maintain detailed data about students’ participation in the school system over time. This is their minimal function in an education system that allows education administrators to be aware of school enrolments in their country, to ensure students get all the rights provided by law, sometimes to allocate funding to schools – and finally to fuel education statistics, research and feedback to stakeholders.
Student information systems typically include individual-level data on enrolment, school and attendance/absenteeism, study pathway taken, participation in special educational programmes, socio-economic background (or rights to specific benefits such as free or subsidised school meals), all of which reflect critical inputs and factors for student attainment and learning outcomes. In addition, systems typically (or need to include) information about student socio-demographic backgrounds. These data are essential to draw student population profiles in terms of sex, age, family structure, socio-economic status, race and ethnicity, immigration history or other variables that may play a role in their learning or academic performance, as well as to analyse the distribution of student outcomes across these background characteristics. Socio-demographic information may include data elements remaining fixed over time (e.g. birth date, sex) as well as others which may require periodic updates (e.g. family composition, household income) to ensure an accurate depiction of students’ out-of-school environments. Most systems collect some of this socio-demographic information.
A 2015 survey of over 60 education information systems (González-Sancho and Vincent-Lancrin, 2016[1]), showed that almost all of them include basic demographic information such as students’ age and sex, a smaller proportion of systems maintain data on other family background characteristics commonly associated with student outcomes such as socio-economic status (60% of systems), national origin (69%) or area of residence, and a minority of systems on students’ special education status or needs, language spoken at home, family size or other family characteristics. While our comparative data collection did not go in the same depth, the interviews of country representatives showed that most systems (or student registers) have the same basic student data – and likely on average more details about students than in 2015. High-quality information on the secondary dimensions (socio-economic status, race/ethnicity where relevant, immigration status, etc.) is essential to carry out meaningful analyses of gaps in attainment across socio-economic groups or between immigrant and non-immigrant students and or to assess the impact of neighbourhood characteristics on student performance, among others. At the same time, when shared anonymously, these additional characteristics make it easier to possibly re-identify individuals and thus raise privacy issues that need to (and can) be addressed through a variety of techniques.
Information about the organisation and resources of the schools attended by students throughout time is also necessary for gaining a deep understanding of the context in which learning takes place and identifying the factors that can foster or hinder student success, and thus to effectively inform school evaluation and decision making regarding improvement strategies. Countries have differing levels of information about this, and this is currently apparently less often linked to information available within information systems, including where the information is available.
Student outcomes
Student outcomes can be measured and reported through a variety of data. Historically, the focus has largely been placed on end-of-cycle graduation and pathway transitions. These data take the form of exam outcomes (pass or fail), credits gained, grade progression or retention, or completion and graduation at the end of a learning unit, term, school year or educational level. This statutory information is typically available within student information systems and student registers.
While students have received teacher-given grades for ages, in the past few decades educational jurisdictions have gradually developed national evaluations of their education system based on standardised assessments of what they consider key subjects. Digitalisation also allows one to more easily record teacher-given grades.
In this context, there is thus much more scope for education information systems and student registers to maintain student data relating to standardised summative assessments of learning and teacher-given grades at the end of specific courses and grades. Out of 29 countries and jurisdictions for which we have comparative information, 13 (almost half of them) can link student unique identifiers to the results of their jurisdictional standardised assessment and 11 (slightly more than one third) record students’ teacher-given grades at the end of some courses or cycles.
In the United States, standardised learning outcomes within states or jurisdictions have become universal, and their outcomes are typically included as a data element in their longitudinal data systems. In other countries, the information is not directly accessible within the student information system (e.g. Italy, Luxembourg or New Zealand), but standardised learning outcomes are typically connected to students’ unique identifier and this information can be retrieved through data linkages. Where there is a student register only (e.g. Denmark), this information will be included. European countries often keep information about students’ results separate from their student information systems, making them only possible to use for research. In some cases this is justified on privacy grounds. Probably this is related to the institutional administration of the national assessment: in Italy, a specific public agency (INVALSI) is responsible for the national assessment, and it is separate from the ministry of education; in Luxembourg, the University of Luxembourg (LUCET) is responsible for the administration of the national standardised assessment, and it is separate from the ministry of education (or higher education). In France, the national assessment collects information at the school rather than the individual student level, for privacy reasons and as an outcome of collective bargaining.
Most of the national standardised assessments concern literacy (in the national language(s)) and numeracy. In some cases (e.g. Italy, Norway or Sweden), other subjects such as English as a foreign language are also assessed. Teacher-given grades typically cover a wider range of subjects. The need to develop a broad set of knowledge, skills, attitudes and values to prepare students for the future is recognised in the OECD Education 2030 learning compass (OECD, 2018[5]) as well as in the ongoing work on creativity and critical thinking (Vincent-Lancrin et al., 2019[6]), on creative thinking (OECD, 2023[7]), and the work on socio-emotional learning (OECD, 2021[8]), which highlight the need of developing new teaching, learning and assessment methods to better measure and value a broader set of desired student outcomes. While technical knowledge in specific subjects remains a key dimension of students’ education and skills development, the increasing awareness of the importance of this wider range of skills that are useful inside and outside of the classroom are rarely taken into account as part of national assessments or teacher-given assessments, and therefore rarely available within student information systems or central registers (González-Sancho and Vincent-Lancrin, 2016[1]). There is no evidence that this has changed since this previous data collection.
Longitudinal information systems are able to link two of more indicators of student outcomes corresponding to different stages of their trajectories, so that individual student growth trajectories can be analysed. The timeframe of such progression will vary depending on the frequency with which data are collected, but the possibility of linking baseline and follow-up measurements of student outcomes holds the key to move beyond a snapshot description of students’ levels of performance and explore, in addition, their evolution over time. Longitudinal data linkages thereby allow a better evaluation of current individual performance in the light of past outcomes. This type of analysis differs from comparisons of school-level performance indicators over time, which may rely on student-level data but do not serve to identify individual learning trajectories. Making learning outcomes data accessible through an education information system can for example allow providing teachers, school leaders or parents with diagnosis information about specific students based on their past trajectory and on lessons learnt from mining information about the entire education system’s current and past cohorts.
Teacher-student linkages
Teacher data are another critical component of feedback and analyses seeking to understand the factors that promote student learning. For these purposes, comprehensive education information systems could draw on the administrative teacher and staff data routinely maintained by local and school-based systems, which typically maintain information on the qualifications, years of experience, contract status, and roles and duties assigned to teachers and support staff in schools.
By collecting and maintaining detailed information on teachers and, eventually, other school staff, longitudinal information systems can also help policy makers and administrators to better understand important issues about the teaching workforce. High-quality data on teachers is necessary to analyse teacher supply and demand patterns, the distribution of teacher profiles across schools and districts, attrition and mobility patterns, or questions related to the quality of teacher training programmes or in-service professional development.
In addition, teacher surveys can generate a wealth of information about teacher perceptions and behaviours across a range of dimensions, including professional development needs, teaching practices, school climate or job satisfaction. Survey data from school principals can also contribute to a better understanding of school leadership and school policies, including teacher appraisal and feedback. Where there is enough trust to collect such data, probably by third parties, the availability of unique personal identifiers that teachers would maintain throughout their teaching careers enables longitudinal analyses of these and other questions. Where not possible, usual anonymous surveys remain the way forward. Another source of information might eventually come from some of the tools used by teachers, assuming they can be reliably anonymised.
Further, the possibility of matching teacher and student data brings opportunities to explore the relationship between teacher characteristics and practices and student success. Of the 29 countries and jurisdictions for which we have information, 10 (about one third) have the possibility to link teacher and student data (Table 2.1 and Figure 2.2). Gonzalez-Sancho and Vincent-Lancrin (2016[1]) showed that these linkages were already rare in 2016. In most cases, this stems out of a political compromise with the teaching profession and their representative organisations, for example as a way to build buy-in and avoid some possible uses of the information such as the calculation of teacher “value added” indicators. In principle, to the extent that the information and analysis are used in an ethical way, teacher-student linkages could be beneficial, for example as a way to identify teachers that are particularly successful with certain group of students and understand their practice.
Built-in reporting and analysis tools
The third “basic” feature of longitudinal information systems is the integration of built-in solutions for querying data and automating customised feedback to stakeholders eligible to access the system, ideally using real-time or very recent information. These reporting and analysis tools are complementary to functionalities for data collection and storage. In this respect, an important difference between longitudinal information systems and statistical datasets is that information systems incorporate some form of user interfaces to facilitate data visualisation and analysis as distinct from providing access to raw data (then being processed through statistical packages).
Our comparative information shows that 10 out of 29 countries and jurisdictions (about one third) have dashboard or tools providing data analytics to stakeholders as part of their information system and 15 (that is, about half of them), use real-time information (Table 2.1. and Figure 2.1). Additionally, eleven countries (37%) provide public access to the data they collect through a public dashboard, usually a separate website allowing for ready-made as well as custom data queries (e.g. My School in Australia, Find your school in Ontario (Canada), Scuola in Chiaro in Italy, the Education Information Disclosure System (Hakgyo Allimi) in Korea, Skoleporten from 2015 to 2021 and “Point of view” analysis in Norway).1
Examples of user interfaces include data visualisation dashboards, automated reports in print or digital formats, and solutions that enable users to engage in data mining and descriptive statistical analysis of one or more variables (e.g. distributions, cross-tabulations, trends). Analysis and reporting tools can thereby enable comparisons of individual and aggregate-level data relative to benchmarks and allow for adjustments that improve the relevance of these comparisons. More advanced feedback could consist of predictive and diagnostic models that help stakeholders assess current trends and anticipate what might happen in the future (e.g. an early warning system) or the provision of advice based on the data.
Quick and easy access to the data reporting and analysis tools is essential to ensure that the whole range of stakeholders can receive feedback from the information system in a timely manner. The more timely the information, the more valuable it is. Effective reporting tools serve to shorten delays in making information available and speed up the transformation of raw data into contextualised answers to stakeholder questions.
Web-based interfaces offer a practical way to integrate analysis and reporting tools in longitudinal education information systems. Online portals can provide the information most frequently needed by different users in a cost-efficient and flexible manner, for instance by mining data from the system and automatically feeding ready-made reports and featured dashboards in dedicated stakeholder areas of the portal. A single data-querying interface can be designed to provide access to data with varying levels of sensitivity for users with different rights and responsibilities.
Access models
Education data systems may well maintain a goldmine of information and, yet, yield minimal benefits if a large proportion of those who could use the data to inform their decisions cannot do so. Beyond user-friendly reporting and visualisation tools, access policies (who has access to what types of data, and at what level of granularity) critically mediate the value that stakeholders derive from data. Access policies are in principle designed to balance the potential benefits and risks of making available detailed information about students, teachers and schools. Benefits stem, most importantly, from providing enhanced feedback to inform decision-making processes. Risks, on the other hand, may relate to the involuntary disclosure of personal information and the non-intended, controversial uses of the collected data and derived metrics.
A previous study analysed the data access policies of information systems (González-Sancho and Vincent-Lancrin, 2016[1]). Those policies can be characterised across three major dimensions. The first relates to the degree of anonymisation of the individual-level data maintained by the systems. The second concerns the relative scope of coverage across the jurisdiction where the information systems operate. The third pertains to the level of aggregation at which data can be seen. Differentiated access policies can be identified combining these three dimensions.
Table 2.2 summarises the typical conditions for accessing the data maintained for a range of stakeholders (where stakeholders are granted access, which is relatively rare across OECD education systems, given that access is correlated with the availability of dashboard analytics and reporting tools).
The most restrictive access conditions are those applied to students and families. They can generally access their own and other anonymised data only within their establishment or class, and occasionally some aggregated benchmark indicators at the school-, district- or national-/state- levels, for instance average graduation rates in different schools (or information about the school when a public dashboard is also available). The small proportion of systems giving students and parents access to their data and reporting tools apply strong restrictions along these lines. This follows a privacy logic that students and parents have no reason to know about other students’ individual information (even though students may know about the information about their classmates through other channels).
Teachers and school principals have relatively similar rights provided that access to data through visualisation and comparison tools is granted at all, a benefit less often available to teachers than to principals. When access is available, principals and teachers can normally see non-anonymised data required to monitor and support students over whom they have direct responsibilities. However, for both principals and teachers, access tends to be limited to data about students within their own schools or classes. Some systems may allow principals to run comparisons using data from their own schools and like-schools, but not to access individual data files of students attending other schools or campuses. Teachers have generally more limited rights, with access to data from their own classes only (and only rarely the possibility to compare their classes with classes like theirs).
Policy makers and education administrators usually have access to anonymised data across the entire jurisdiction. A small number of designated individuals within local, state or national authorities are generally able to obtain detailed non-anonymised information about all the data subjects followed by the system, but, unless in exceptional circumstances, their access to the information remains subject to strong privacy and confidentiality regulations and thus occurs via anonymised records as for administrators (e.g. superintendents, district leaders, central government officers). This holds true with regard to student data for most student information systems included, while rights on teacher data by certain administrative bodies, especially the inspectorate, more often involve access to personal records.
Lastly, researchers are typically not allowed to access data directly through the information system, but they have access to the anonymised database derived from the information system. In all cases, researchers can only access fully anonymised data, and most often at an individual level. Occasionally researchers have to work with representative samples of the student or teacher population, rather than with datasets covering the entire jurisdiction. A small number of systems make synthetic data files designed to reflect and simulate the true population available for research purposes. In addition, most systems require researchers to meet stringent criteria to gain access to their data. This normally involves the submission of detailed proposals detailing the value and ethical standards of the envisioned research. External researchers can only gain access to cleansed and anonymised data files containing a limited number of variables required for their analyses, rather than to full datasets. Policy makers could make aggregated information available in almost real time to avoid that researchers spend time replicating findings that could be automated.
It is worth noting that no stakeholders are granted extensive access to non-anonymised data. This suggests that there are currently no major privacy or confidentiality risks that may derive from “lax” data access policies. Generally, most systems follow access policies inspired by the “different data for different roles” principle, and which prioritise privacy and confidentiality protection over flexibility of access.
Summary
Most OECD countries have national student information systems, a cornerstone of any digital education infrastructure. Some federal countries have them at the sub-national level, usually with reporting obligations to the central level. However, a few countries lack such systems. While some collect the same data stored in a central student register (which is less actionable), others just collect aggregated statistics.
A key feature of these systems is the use of unique, longitudinal identifiers for schools, students, and sometimes teachers, enabling data linkages over time and over the entire school career of students. They can include extensive student-level data, covering enrolment, attendance, study pathways, socio-economic backgrounds, and other demographics, facilitating the analysis of student outcomes and disparities. These systems also capture student outcomes, such as graduation rates, exam results, credits, and grade progression, with some countries additionally collecting teacher-assigned grades and standardised assessment results.
Data linkages could in principle empower the analysis of individual growth trajectories, enhancing the evaluation of student performance and learning trends, and turn this information collected about all students into actionable information for practitioners in schools.
Only a minority of countries include or link their student information systems with the individual results of their standardised national evaluations (45%), provide dashboards or visualisation tools (31%) to make it easier to use the information in real time, or possess the capability to link teacher and student data (31%), which could possibly provide valuable insights into the impact of teacher characteristics on student success.
Our international survey covered information about countries’ digital infrastructure and governance. A previous survey specifically on longitudinal information systems also collected information about the intended goals and uses of the systems. A great variety exists in terms of information systems’ objectives, the nature of their data, their access policies and the functionalities that they enable.2. This section proposes four ideal-type models to classify the main functionalities and origins of current longitudinal information systems in education: the reporting and research, e-government, school improvement, and expert system approaches.
The reporting and research approach
A first category of longitudinal information system comes from a statistical and evaluation approach – that is, the traditional approach to the function and strengths of information systems. Unique individual-level identifiers and longitudinal linkages have allowed these systems to enrich their reports and performance cards, but the information produced seek mainly to serve policy makers or to inform the public about general trends in the education system. In some cases, the systems are also meant to develop research capacity about educational issues.
In Ontario (Canada), the Ontario School Information System (OnSIS) collects data to inform decision-making related to education policies, programmes and practices within the province. OnSIS was launched in 2006 as part of the broader Managing Information for Student Achievement (MISA) initiative, whose goal was to build capacity for data use at both the local (school board) and provincial levels. OnSIS collects over 100 million new data records across multiple levels and collections every year. Despite the large growth in the volume of data collected with the introduction of OnSIS, the MISA initiative has enabled a significant decrease in the time required to collect data from boards and an increase in the quality of data. A primary function of OnSIS is to support the ministry’s analytical needs and provide key indicators about policy priorities.
The system enables highly granular statistical modelling and trending analysis and a rich contextualisation of student achievement patterns over time. Longitudinal indicators can be constructed at the provincial, school board and school levels, as well as within sub-groups. This enables improved monitoring and a better understanding of the factors relating to student attainment. Longitudinal tracking of individual students is possible through the Ontario Education Number (OEN), a unique student identifier that provides links to data from multiple sources. Uses of linked records include the development of indicators tracking changes in the proportion of students attaining different levels of performance between grades, and examining trends relating to student participation in postsecondary education in Ontario. By collecting timely and quality education data, OnSIS also supports information dissemination through the production of public reports, for instance board progress reports, a school information finder, and trend reports at the provincial level. These products seek to provide timely and consistent evidence to inform strategic planning and decision-making.
In Mexico, the government has established a powerful education information system, SIGE, with student longitudinal information about enrolments coming from schools in all the public schools and states of the country, assessment data (when there was still a national assessment), and a variety of information about schools and teachers. This information is used for some administrative purposes, but mainly to generate reports and allow administrators to monitor the system. The National Institute for Educational Evaluation (INEE) launched in 2016 the Sistema Integral de Resultados de las Evaluaciones (SIRE), a new system using SIGE and assessment data to organise and provide visualisation tools for information on the results of student and teacher assessments as well as on the physical and socio-economic context of primary and secondary schools across the country. The SIRE system is designed to support INEE in its mandate to coordinate the evaluation of the education system at federal and state levels, within the frame of the Sistema Nacional de Educación Educativa (SNEE).
SIRE informs strategic planning and the design of educational evaluation policies by disseminating contextualised evaluation results to public authorities, educators and parents in an accessible manner. An online portal enables users to customise visualisations of hundreds of variables over 600 geographic layers at local and state levels. The platform also supports dynamic queries into the database, provides reports on key aspects of the evaluation framework, and permits researchers to download aggregated data and establish linkages to external datasets from other government agencies. Some parts are reserved to authorised users within the Ministry. While the pause in the national standardised assessment has reduced the richness of the reporting, this is a powerful system that is an interesting example of central collection of student-level data information in a federal context.
In Latvia, the Ministry of Education and Science developed the governance e-tool Schools Map of Latvia in 2016 as part of a broader project seeking to improve information about school networks. The tool includes an interactive geospatial map of all schools in the country that gives access to school-level indicators such as average results in official examinations, teacher-pupil ratios, study programmes offered, languages of instruction, infrastructure and placements of leaving students, among others. Besides providing an overview of school characteristics, availability of transportation and other local services, the platform enables comparisons of schools’ performance as well as analyses of students’ mobility trends. The online public version of the Schools Map is designed to help parents and students orient their school choices. Additional functionalities exist for local and central education authorities to assist in decision making for the planning and resourcing of the school system.
Systems built for research purposes are a variant of the reporting and evaluation model. The longitudinal information system of the state of Washington (United States) was designed to improve research capacity to address critical questions for education policy at the state and local levels. Besides the production of feedback reports for policy makers with information on a wide range of programmes and outcomes, the system serves as a data source for educational researchers. It integrates records from multiple agencies across the state into an operational data store with over one billion records about 6 million individuals, making it possible to follow their trajectories from early childhood education through compulsory schooling, post-secondary and into the workforce. To facilitate data sharing across agencies and with external researchers, the state of Washington has established a multi-agency data governance structure. The system relies on a third party for software and assistance in linking identities in its data warehouse.
Some systems such as the New Zealand National Student Index (NSI) was designed to be able to track students longitudinally and is mainly used for statistical purposes: the system, a web platform, allocates to each student a national student number valid from early childhood to tertiary education. School leaders and students use the platform to search and modify the (student’s) information in the database records, merge duplicate records, and thus assure the quality of the individual data and of the statistics that can be derived from the database (such as enrolments, graduation, etc.) or from linking distinct databases with students’ unique longitudinal identifier. A separate, non-connected system, ENROL, specifically maintains information about enrolments and school attendance at the national level, allowing some interventions to prevent school dropout and enforce mandatory participation in education – as well as statistics on attendance, absenteeism and dropout.
Public websites allowing families and the public to search information about the schools in the country or jurisdiction thanks to the individual data collected and how schools (or districts) compare with other schools give a good (albeit incomplete) public illustration of this approach, and how countries expanded the granularity of their statistical reporting over time.
The e-governance approach
A second type of longitudinal information systems were inspired by an e-government approach, having the main ambition to standardise and increase the efficiency of administrative processes (e.g. school transfer, school choice, university application, etc.). Another important goal behind the design of some of these systems was to optimise financial allocations to schools when funds are based on a student-based formula, which is the case in several OECD countries with such systems. The longitudinal information systems of Korea and Estonia are good examples of the e-government approach, which results in systems with more data and linkage possibilities than systems designed with different approaches, but which tend to have a weaker focus on data use at the school level and on functionalities aimed at supporting instruction.
Korea’s National Education Information System (NEIS) is a web-based “one-stop shop” for offices of education, families and teachers, connecting educational and rich administrative information. Its objectives are the sharing of information to achieve improved cost-efficiency and to reduce teachers’ workload by freeing them from data entry and verification tasks. The system electronically processes administrative data from about 12 000 schools and 17 metropolitan and regional offices of education across the country. The information is used to manage student admission and enrolment, to record students’ standardised test results and teacher-given grades, track student progress and learning trajectory throughout the school year, and transfer student qualifications to other educational institutions (including colleges and universities) in the country. The system also includes teacher information (e.g. their salaries, expenses, and training and development records), the lunch menu of the canteen, etc. Extensive encryption procedures and differential access rights are in place to protect the privacy of sensitive information about students, educators and schools. As an e-government tool, it allows families to create school certificates and sends school permanent records of the third-year students of high school to their applicable university online. It includes interfaces for school leaders, teachers, parents (NEIS for parents) and student themselves (NEIS for students). The system is also interlinked with those of other 15 government agencies through common unique individual identification numbers.
The automation of administrative processes in schools has significantly reduced the time required for processing documents, leading to cost savings estimated at over USD 200 million per year. The system remains strongly focused on improving administrative efficiency and recent initiatives and improvement have kept this focus. Given the great variety of data elements that it maintains and their high reliability, its data was arguably underused to support more personalised teaching and learning as well as educational research. However, the fourth generation NEIS, which was made available in June 2023, provides data on students’ learning history and allows for student course registration and teacher assignment submission. In its original digital governance spirit, its updates contribute to teacher workload’s reduction through administrative task automation.
Established in 2005, the Estonian Education Information System (EHIS) is part of Estonia’s extensive e-government infrastructure. EHIS and the rest of the national information systems use the X-Road data exchange layer to automate data sharing in a secure Internet-based environment and are accessible to citizens through their ID digital cards. By integrating data from different education registries, EHIS allows individuals to access and manage their personal education and training records across the life course. Applying for university studies by transferring personal details to the desired institutions is the most common use of the EHIS database. The data visualisation web environment HaridusSlim (Education Eye) within EHIS provides information about education programmes and institutions across the country in order to support decision making at multiple levels and for different stakeholders, from school choice by students and families to planning and monitoring by local and state policy makers. Plans for development include enriching the information contained in HaridusSlim to enable longitudinal analysis of school effectiveness and graduates’ success in the labour market.
The school improvement approach
A third type of education information systems puts school improvement at the core of their mission. While sharing many features with systems that have a statistical or evaluation inspiration, systems designed with a school improvement approach make more data available to schools, generally through custom templates and visualisation tools, and seek to provide information at the individual level and with a granularity that makes it potentially usable by teachers (for example, item-level reporting of assessments). However, they tend to target a school improvement approach by targeting school principals (or inspectors) as their main stakeholders. Systems in England and Portugal, Hungary or Gujarat (India) belong to this category.
England created in 2004 the RAISEonline web-based information system for analysing and reporting school performance. Within the context of national standardised assessments, its objective was to encourage school principals to respond proactively to achievement gaps in the performance of their school relative to comparable schools. A self-evaluation objective guided its conception, as reflected in the possibility for schools to personalise reports (e.g. by creating reports on particular groups of students), to add information to the system (e.g. additional information from non-national tests), but also in the restriction of access to school principals and administrators. The system also provided value-added measures of pupil progress that control for contextual factors (ethnic group, poverty status, etc.) and statistical confidence intervals are included in the reports. It was used by over 20 000 schools and 3 000 inspectors across the country to support school self-evaluation and the school inspection process. Since 2017, RAISEonline has been replaced by Analyse School Performance (ASP), a new system co-developed with the Office for Standards in Education (Ofsted) that gives access to the detailed pupil level data that was previously available on RAISEonline and reproduces most of its functionalities, such as enabling schools to analyse their results against other schools nationally and develop “improvement plans”.
In the same spirit, Hungary started collecting information from the National Assessment of Basic Competencies in 2001, but it was only after introducing individual student identifiers in 2007 that the system allowed longitudinal tracking across grades. Since 2010, Hungary has aimed to turn these data into actionable information to school practitioners. Individual reports are available for each student (and the school aggregated reports are open to the public). School principals can connect to the database through the FIT analysis software (FIT elemző szoftver) to make customised comparative reports about their school and students and make comparisons with similar schools. Moreover, the Education Authority, an agency of the Ministry of Interior, provides an expert consultant service through its regional pedagogical centres to help schools and teachers understand and utilise the results. In 2022, new subjects were assessed (science, languages) and expanding the scope of the information is in progress (history and digital culture). In 2023, the covered grades were also broadened to students from 4th to 11th grade. The reporting of results is also currently under revision. The fact that the information maintained by the system is directly accessible to multiple stakeholders (school principals, teachers and students) and that institutional support is provided to schools and teachers for improvement, albeit on a voluntary basis, makes the approach different from the reporting and research systems that typically provide school reports.
In Portugal, the Ministry of Education began piloting the new Escola 360° (E-360°) system in 2017, and in 2022/23 about 150 schools and school networks used it. Portugal was compelled by the economic crisis to improve its capacity to evaluate the efficiency and efficacy of its education system and the E-360° system culminates the process of integrating a previously fragmented education data infrastructure into a new centralised information system. The country had previously developed the MISI information system to collect data from independent school management systems and feeding a separate database per school year. Despite containing very rich data at the individual and school levels, these data were hardly used to inform school decisions. The system did also not enable longitudinal linkages and suffered from poorly defined standards. For these reasons, usage was largely limited to producing yearly reports for budgetary and policy planning. The co-existence of multiple electronic platforms for school procedures led to a duplication of work and sometimes poor communication with stakeholders.
The E-360° system was designed to overcome these limitations. It strengthens the ability of the ministry to produce improved indicators and address policy questions by providing a complete view of student educational paths in a more granular and timely manner, building on the individual student and teacher identifiers of the national e-enrolment system introduced in 2010. In addition to the new collections, the central database incorporates old cross-sectional data series, which were matched with a success rate close to 80%. By integrating all administrative information related to students (e.g. personal background, enrolment and transfers, attendance, assessments) on the same platform, E-360° centralises student management operations, from pre-school to upper-secondary education, supporting the entire processes of enrolment, renewals and transfers of students and automating tasks such as certificate issuing. Moreover, the system seeks to improve the exchange of information from administrative bodies and schools to students and families. Despite those “e-government” functionalities, the platform is maintained by the Directorate for education statistics and science of the Ministry of education and tries to support improvement efforts through the provision of indicators. It plans to develop early warning tools for educators and learning analytics to improve instructional practices. While accessible by all school stakeholders, most functionalities and information seem to be mainly relevant for administrators and school principals (Lopes, 2022[9]).
Gujarat (India) developed an information system called Vidya Samiksha Kendra (VSK) to support its system and school improvement efforts. The system brings together a large number of data and applications that were already collected, but brought them together and added strong visualisation tools and human follow up based on the data. The quality of the data in the system is ensured by a central team verifying on a daily basis that all schools and stakeholders have entered the expected data in the system and following up by phone within a certain period of time if this is not the case. In the Indian context, school improvement is related to student enrolments and attendance as well as the presence of teachers and school network (district) coordinators in school. The system allows for such tracking and intervention, for example by geo-localising school network coordinators in real time. All students take a standard set of short assessments every other Saturday, whose results are entered in the system by data phygitisation (digitising data from the students’ physical workbooks by taking a picture of a coded area), which allows to provide regular “report cards” to parents, teachers and school principals about students’ progress. Finally, the system allows for pedagogical support through a random selection of classes that are tele-observed by expert teachers, who then provide the observed teachers with feedback. Making the collected information more visible has allowed to better understand some of the shortcomings of the system and to develop new policies such as the “excellence boarding schools” for students from rural areas. The state is also planning to review its school grading system based on the data collected through the system rather than human observation. This is a way to incentivise schools to improve in specific areas (such as infrastructure or teaching and learning).
The expert system approach
Finally, a fourth type of information systems are inspired by “expert systems” that aim to personalise teaching and learning by providing rapid and granular feedback to teachers, students and principals, as well as support materials to enhance learning. These systems are comparable to the school improvement systems in terms of richness of information, but they place yet a greater emphasis on providing actionable feedback to the instructional process. Notably, they tend to provide actionable information to teachers and not just school principals, and thus allow for “class improvement” in addition to “school improvement”. Some of their features overlap also with the e-government systems, most commonly linkages with post-secondary education data or from other agencies. They indeed tend to draw on a variety of sources of information to provide comparison and reporting tools related to a variety of pre-identified “business cases”. An important feature is that they attempt to provide recommendations to stakeholders beyond mere descriptions. Longitudinal information systems in Colorado (United States), New South Wales (Australia), Charlotte-Mecklenburg (United States) and New York City (United States) can be classified as belonging to this ideal type. The upgrading of a different type of system in New Zealand may also go in the direction of the expert model.
In New South Wales (Australia) a web-based system called SCOUT was launched in 2018, succeeding to an already advanced system (SMART [School Measurement, Assessment, Reporting Toolkit]). The system mainly targets government schools, but provides a more limited set of services to non-government schools. It brings together information from 150 data sources to provide information to school directors, principals and teachers. It notably includes information from national tests (NAPLAN, which all students complete about every two years) as well as exit exams (HSC) and state science diagnosis assessment (VALID). Teachers can customise reports by creating, for instance, specific target groups of pupils whose progress they would like to monitor. The objective is for teachers to gain ownership over the information they are looking at, while providing a tool to map (personalised) teaching into national standards and comparisons. For example, the system offers different comparison tools allowing teachers to compare students’ actual performance with their expected growth (given past achievement) or a group of pairs. The system also enables tracking mobile students from one school to another. School principals can also compare their schools to different averages, including schools of similar characteristics, in terms of learning achievements, but also “engagement” (measured by attendance, absenteeism, sick leaves, etc.). While SMART allowed teachers to use a lesson bank and teaching strategies linked to the data available from the website, this does not seem to be the case with SCOUT. However, the system connects the information to each school improvement plan, and goes beyond the “school improvement” approach by allowing detailed comparative analysis about students that can inform decisions at the class and student levels. Given the wealth and details of information, only eligible people can access the system after a mandatory training.
In New Zealand, the Assessment Tool for Teaching and Learning (e-asTTle) has been in use since 2002. E-asTTle is an educational technology that enables educators to design and generate standardised and curriculum-aligned tests in reading, writing and mathematics in either English or Māori languages, in primary and secondary education. A large bank of calibrated items allows teachers to customise tests to the specific needs of their classrooms. Contrary to other computer-assisted testing systems, e-asTTle was designed with a strong emphasis on formative rather than summative assessment. Tabular reports of students’ scores can be easily integrated with schools’ student management systems, thus combining information from tailored assessment with school-level administrative and operations data.
An expert feature of e-asTTle is the ability to transform assessment results into prompt and interpretative feedback for teachers and school leaders through a range of graphical reports. These show student progress and areas of weakness and strength, as well as comparing performance at the individual, class or school levels to curriculum requirements, national averages, or normed gender, ethnicity, language or socio-economic groups. An example is the Individual Learning Pathway report, which gives information on a student’s strengths and gaps. Another expert application links assessment results to the What Next website, an indexed library of resources to help teachers and learners identify appropriately targeted learning materials.
In 2015, the New Zealand Ministry of Education launched a Student Information Sharing Initiative (SISI) aiming to bring a strategic perspective on data quality and data management practices to the national school system and to address the problems derived from the use of a wide array of non-integrated IT solutions across New Zealand schools. The proposed solution was a central data repository storing and exchanging core student information between the Ministry and school-level student management systems through interoperable data standards and services. Data and resources from e-asTTle and other assessment tools alongside administrative data were meant to be easily linked within the same application, which would have brought the “expert” functionality to the next level. As of 2023, the initiative seemed to have been interrupted and replaced by other projects withing the Integrated Education Data (iEd) programme, a 5-year policy initiative launched by the Ministry of Education in 2017 with the goal to improve the capacity of the New Zealand education sector to access and use data for improvement purposes.
Charlotte-Mecklenburg Schools (North Carolina, United States), the 17th largest school district in the United States with 146 000 students across 189 schools, has a long experience using its longitudinal information system to facilitate data-informed decision making for school improvement. Besides supporting management and accountability at the district level through quarterly and annual school performance reviews, the Charlotte-Mecklenburg information system puts data in the hands of practitioners with a strong focus on promoting changes in practices at the ground level. Three departments of the district (office of accountability, department of data use for school improvement, department of research, evaluation and analytics) analyse the data they collect to provide actionable information at the district school levels, in line with the “school improvement” approach. Moreover, principals and teachers can use the system to examine a wide array of longitudinal student-level data including interim assessment results, attendance and incidence records that are updated daily. These highly granular indicators are the key input for designing and iterating short instructional cycles in a formative and low-stakes approach to data use at the classroom and school levels.
New York City (United States), the largest school district in the United States, serving over 1 million students in about 1 800 schools, developed and used the Achievement Reporting and Innovation System (ARIS) between 2008 and 2014. ARIS aimed at easing administrative tasks and drive innovation through data use. The development of the data system was informed by this strategic goal and intended to provide local actors with key information to meet their needs (and strong incentives to use the information). The web-based system consisted of three components: an information system providing comparison tools and assessment reports to teachers, principals and administrators about performance (and expected performance) (ARIS); a collaboration tool through web 2.0 tools (wikis, blogs, discussion forums, communities) so that educators could share and refine best practices (ARIS connect); and a tool allowing parents to monitor the progress of their children and giving them information and support to help them (ARIS parent link). Since ARIS was discontinued, the department of education has continued to make a wide range of data available to their staff, families and researchers through its InfoHub portal and the NYC Open Data initiative. While it largely follows a reporting and evaluation approach, employees of the New York City Department of Education, including school principals and teachers, have access to a wealth of additional, more detailed information. Teachers can view their students’ report cards, and the city provides a wide range of digital services and tools to its teachers and students.
The state of Colorado (United States) provides public access to information and analysis derived from its longitudinal education information system on the “SchoolView” website. The portal combines four features: a social network for teachers, a learner centre and instructional and assessment resource bank, interactive school performance charts, and access to performance data and reports. While the system does not make recommendations about instructional material and assessment to teachers, bringing all these elements together adds to the school improvement approach. School performance charts are graphical representations of both achievement and growth in achievement. A prime feature is the use of the “Colorado growth model” which expresses growth in performance in percentile form, that is, indicating whether students and schools are improving more or less than the expected improvement for similar students and schools, and therefore presenting growth as a relative rather than absolute concept. Access to such detailed student-level information is available for authorised users only.
More in line with the reporting and research missions, Colorado has also been collecting student-level data from its public higher education institutions since 1988 through the Student Unit Record Data System (SURDS). The Colorado Department of Education has established strong partnerships across agencies to match K-12 data to the wide array of elements in SURDS including application, enrolment, remediation, completion and financial aid student data. These longitudinal matches, going back to 2009 with a success rate of about 94%, allow stakeholders to analyse the factors influencing student post-secondary progress and success, including by disaggregating data by key student demographics and running comparisons between districts and state averages. Another initiative has enabled linkages with earnings data for recent college graduates from the Colorado Department of Labor and Employment, which can be examined by institution or area or study via the College Measures portal. Linkages to data from the National Student Clearinghouse permit to track Colorado high school graduates in out-of-state higher education institutions not included in SURDS.
All these examples give some idea of what an “expert system” approach could look like, even though it may not have fully materialised. The big difference with the school improvement approach is that expert systems use several data sources to try to help teachers and other stakeholders to make decisions about specific students with real-time (the most recent) and past information, usually using diagnosis and predictive digital tools in addition to the human insights and analysis that supports the school improvement approach.
Summary and remarks
This overview shows that strong longitudinal information systems exist in many countries. With no exception, data management tools serve some statistical and evaluation purposes at the system level. Additionally, information systems often facilitate administrative processes and assist families in making their educational choices, most often by providing open information about schools or regions. Less frequently, systems also make data available to schools and educators so that they can devise improvement plans, and in fewer cases, they further provide easy access to instructional material and real-time feedback to personalise teaching and learning processes based on information about current students within a school. The diversity of these systems and of country experiences in developing them is valuable in itself for all countries willing to establish or improve existing systems.
A few countries collect once a year detailed longitudinal information about their students that is recorded in a central student register, that is, an exhaustive database of all students in the system. All countries having a longitudinal student information system also have a student register – but having a database is not the same as having a system that connects to this database and allows some stakeholders to make real-time queries. While they enable strong statistics and post-hoc research on the system and thus align with the research and reporting approach, countries with a mere central student register cannot use them for real-time interventions. (Few countries with a longitudinal information system do so, but in principle they could.)
The ideal types presented above schematise the variety of purposes and functionalities of education information systems. Some countries may have systems using similar data coexist. For example, in England, Analyse School Performance showcased a system geared towards school improvement. However, the system supplements another student information system more aligned with the reporting and research approach though, Get Information About Pupil (GIAP), that includes England’s National Pupil Database, its central student register that is intensively used in education research. The GIAP system can also be accessed by specific people within schools and local authorities to find or verify information about specific students (e.g. information about pupil premium funding or end of key stage results), thus also embarking some e-government features.
While a key feature of a modern student information systems lies in the collection of information that connects to students’ unique longitudinal identifiers (and sometimes teachers’), most of them initially did this with administrative rather than mere statistical purposes. While some systems have as a primary purpose the generation of statistics about the education system, most of the information is collected for other administrative purposes (and merely re-used for statistical purposes). One of the main original purposes is “funding” or “budget allocation” when based on a formula that requires to know with precision how many students a school enrols. This is for example the case in the Netherlands or in Finland. Some e-government purposes have also been mentioned: the transition of students from one school to another (Korea), the enforcement of mandatory schooling as well as other school obligations in relation to health, social support (New Zealand), or the awarding of or verification of student credentials, or the eligibility to benefits (England, Mexico), etc.
The emergence of national standardised assessments to evaluate the learning outcomes within schools has led to the addition of student learning outcomes as a key piece of information for student information systems, notably when those assessments are conducted with a census methodology (that is, all students of a certain grade within a jurisdiction take the test). (In some cases, it has led to separate digital systems that overlap with other information systems.) Whatever their origin, comparable learning data at different levels of granularity across all students within an education system is indeed the key element that can make student information systems particularly relevant to teachers and school principals.
Diversity or convergence
Moving forward, education agencies will likely invest to expand the capabilities of their information systems in multiple directions, from incorporating new and higher quality data, to enhancing linkages and analytic functionalities, and to improving conditions for access for and actual use by different stakeholders.
In fact, the evolution of student information systems over time seems to be about encompassing more and more functionalities. For example, while the very rich Korean National Education Information System was hardly used for research purposes initially, access to its data have eventually been open to researchers, allowing to strengthen its “research and reporting” purpose. It is possible that, in the future, it will also be used for school or improvement, and perhaps even as an expert system. Another example is the awarding of credentials: once they become longitudinal and include trusted information, what was initially statistical can be turned into a credential awarding system, a typical e-government functionality.
In some cases, the evolution can go in the opposite direction, as several systems that included many “expert system” features were also dismantled and replaced. Developing information systems and embedding data use in schools requires building trust among education stakeholders. On the one hand, collaboration between different agencies is essential to establish data sharing agreements that expand possibilities to use data for educational improvement. These partnerships depend on institutional trust and mechanisms that ensure that all parties have a say in the questions that the data should help address. Trust can emerge from regulatory frameworks that promote data integration while protecting privacy and data ownership, but also from a common understanding of the purposes and functions of information systems. On the other hand, turning data use into a routine activity in schools requires that practitioners perceive information systems as tools that serve to support their work rather than instruments to meet reporting and accountability requirements. They also have to be useful tools that provide insights that would be difficult to have without the use of the data.
Will the slow but continuous development of longitudinal student information systems lead to a gradual confluence internationally around specific design features? This remains an open question. However, systems need not converge towards a single approach to how education data could and should be used to be effective. In fact, the variety of goals and motivations revealed by current and previous work (see (González-Sancho and Vincent-Lancrin, 2016[1])), and the country, regional and local experiences presented in this chapter and in the related report (OECD, 2023[4]) suggest that the diversity of approaches will stay to meet a range of varying needs, not the least because these information systems also reflect a range of varying governance arrangements across countries. Rather than focusing on a single model, the next generation of longitudinal information systems in education will likely develop in many different yet complementary ways. The case uses of the systems and how the information can be brought into the hands of practitioners (and researchers) may matter more than the actual architecture or initial purpose of the student information system.
Integration or interoperability
A wealth of data is collected daily in education. However, not all this information is actionable to improve student and teacher learning or simply administrative processes. Data are sometimes maintained in silos that cannot communicate with each other and pieces of information that would gain in value if linked across data subjects and time remain disconnected. Different digital applications have usually taken care of all these different functionalities, and the strength of some information lies in making the data collected and managed by different applications accessible by one specific system to provide information according to a specific, identified need that can add value to stakeholders’ decision making.
Should next-generation longitudinal education or student information systems be a gigantic one-stop shop for all educational data? Not necessarily. Full integration of existing data systems would be difficult to achieve and raise security problems. This is a possible model, but it is in fact not required. The most advanced systems can already “mash up” data from separate, independent sources, and this is likely how next-generation systems will work, bringing data in so-called “data lakes” that are mined for different purposes.
Integrated systems with a multiplicity of functionalities could be seen as a possible ideal as it typically provides access to all stakeholders in the systems. Two examples of such systems can be found in Iceland and Luxembourg (secondary education). In these two countries with a small population, the central student information system is also a school learning management system. Different stakeholders have different authorised accesses and uses, but all the data are held together within the same system and one could easily imagine how they could easily be linked and turned into “expert systems”. As of 2023, Iceland does not have a national standardised assessment that would allow its stakeholders to compare a specific class or student to his or to past student cohorts, and the national assessment in Luxembourg is kept separate from the two national student information systems. The education systems presented above in New South Wales (Australia) and Gujarat (India) follow this idea of “mash up”.
The ability to integrate data from different systems and applications without having an integrated system relies critically on the adoption of interoperability standards (see Vincent-Lancrin and González-Sancho (2023[10])). Standards create a common language to overcome the problems arising from the use of different formats and definitions for education data elements. Greater interoperability also facilitates that personal records can follow individuals when they cross jurisdictional boundaries, for example within federal countries (Box 2.1). Importantly, more interoperable information systems would permit to improve speed of feedback. A substantial share of systems still takes several months to provide information based on student data (assessment and others), thus providing little room for teachers, administrators and families to make a formative use of this information. Data-driven organisational routines often have to rely on information about past rather than current students.
The examples from the United States, India and Mexico
Federal countries and countries with a strong devolution of education to sub-national governments and authorities face specific challenges with education information systems. The national government can typically only incentivise each state, region, province or local education authority to develop or maintain a longitudinal information system, even though it usually mandates the reporting of certain data and statistics.
The United States has been successful in supporting the development of “statewide longitudinal data systems” (SLDS) grant programme of the National Center for Education Statistics: since 2005, it has awarded over USD 800 million to nearly all US states and territories and provided them with technical assistance. The first grant programmes focused on the establishment of a longitudinal data system, then on student-teacher linkages, while the programmes now focus on the improvement of existing infrastructures and the use of the data to improve practice. A number of non-governmental actors have also supported the agenda, which led to all US states having a longitudinal student information system in 2023. As these systems are typically not interoperable, and do often not allow to track students once they cross a US state border, a common data standards initiative was launched in 2009 to work on and publish recommended standards for data elements and document existing state standards. This has allowed for some level of convergence or at least opened more possibilities to compare data across states.
Recent efforts in India to establish information systems in all its states and at the central level as part of the National Digital Education Architecture (NDEAR) launched in 2021 illustrate a different approach. At the central level, the Central Institute of Educational Technology (CIET) at the National Council of Educational Research And Training (NCERT) developed a central education information system called Vidya Samiksha Kendra (VSK) that present data visualisations for the centrally collected information about the country (enrolments, results to the national standardised assessment, national funding provided to different states, etc.). While the first VSK was developed by the state of Gujarat (India), CIET-NCERT has supported the development of state-wide VSKs by providing grants to all states and, perhaps more importantly, an open source VSK platform that states can customise, further develop and share new elements with other states in the spirit of the open software movement. While there is no obligation to adopt any particular approach, a mix of strong incentives, grants and the provision of standards embedded in software may lead to the adoption of such platforms but also perhaps to greater interoperability across states.
Despite being federal, Mexico has developed a national longitudinal student information system by collecting information about students, teachers and schools directly (for all public schools in the country). SIGED contains data from the entire Mexican education system – students, teachers and school staff, and schools – as well as documents and information on credentials, thus serving as an education management information system, a student information system, and a platform for digital credentials. Data entry into SIGED is mandatory. SIGED serves a variety of purposes, from funding schools, generating credentials to providing granular statistics.
A vision for the next generation of information systems
Over the next 5 to 10 years, education information systems could be expected to gradually incorporate a series of advanced features that exemplify innovative and promising solutions for the evolution of longitudinal systems. “Next-generation” longitudinal information systems with these advanced features could set the horizon for education agencies working towards building an enhanced data infrastructure to meet the challenges and seize the opportunities of the digital transformation. In particular, recent technology developments support the help of privacy (see (Vincent-Lancrin and González-Sancho, 2023[11])).3
Higher order capabilities of information systems may involve, among others, the ability to collect data of greater granularity, the provision of real-time feedback through customisable reporting tools, links to banks of digital learning resources, or the integration of automated diagnostics and recommendation solutions to help personalise teaching and learning practices (González-Sancho and Vincent-Lancrin, 2016[1]). Current information systems in education could incorporate these features in a gradual manner.
Next-generation systems could be characterised by the ability to combine administrative education data with data from digital learning and assessment platforms that provides a more granular view of the learning process. Relying on the combination of different types of data, next-generation systems could then be powered by learning analytics and other techniques to make customised recommendations and provide links to specific instructional and learning materials. The application of learning analytics would enable systems to organise the navigation of banks of resources through techniques such as knowledge domain modelling, which uses personal data to compare a learner’s knowledge with the mapping of knowledge in a discipline and provide personalised content and tasks to support progress in learning (Siemens, 2013[12]) (Lang et al., 2017[13]). The same “detect, diagnose, act” process as in adaptive learning systems focusing on task-level learning (Molenaar, 2021[14]) could be replicated at a more macro level, that is, study paths and integrate different sources of information about student learning.
Currently, most of the student assessment data included in education information systems are summative data in a small number of subjects coming from national/jurisdictional standardised assessments. Few systems collect data from teacher-made assessments, from formative assessments, nor data about creative and critical skills or social and behavioural skills, even though some systems maintain data about actual student behaviour. One challenge for next-generation systems will also be to include broader data elements about the learning process than those measured by most standardised tests (reading, maths or science) so as to support a holistic education (OECD, 2019[15]).
Developing systems in this direction would require the adoption of interoperability standards to enable data mining and data linkage across platforms (see (Vidal, 2023[16]))). Data integration does not necessarily involve transferring data from multiple sources into a single system. Instead, integration can more easily be achieved by making systems interoperable, that is, making them able to read, mine and link data from one another. Open technical standards and definitions pave the way for this type of integration, so that data from multiple feeding databases can be combined and analysed through a single system or interface.
Another essential feature of next-generation systems would be a greater integration of analysis and reporting tools to facilitate feedback and knowledge flows. Rapid feedback loops that keep pace with data generation cycles can enable more dynamic learning environments for students and teachers, as well as the faster evaluation of interventions by educational researchers and policymakers. At the school level, enhanced analysis and reporting tools would include automated reports, flexible data visualisation interfaces such as customisable dashboards, and discussion tools. For researchers and administrators, enhanced analytical capacity would stem from reduced delays in the turnaround of data and a greater ability to make relevant comparisons between schools and programmes.
Building on prior waves of systems designed to provide an overview of student and school performance at the system level, the main aim of future systems should be to foster improvement by giving quick feedback and supportive tools to teachers, schools and students. An initial step to enable a closer examination of instructional practices would be to establish student-teacher and student-course data links within longitudinal systems. The enhanced capabilities of next-generation information systems would thus open the door for a forward-looking re-orientation of the use of education data, away from a traditional focus on reporting and accountability purposes and towards informing concrete practices in teaching and learning.
References
[1] González-Sancho, C. and S. Vincent-Lancrin (2016), “Transforming education by using a new generation of information systems”, Policy Futures in Education, Vol. 14/6, pp. 741-758, https://doi.org/10.1177/1478210316649287.
[13] Lang, C; G. Siemens; A. Wise; D. Gašević (2017), Handbook of Learning Analytics, Society for Learning Analytics Research (SOLAR), https://solaresearch.org/wp-content/uploads/2017/05/hla17.pdf (accessed on 9 January 2018).
[9] Lopes, F. (2022), Plataformas Eletrónicas de Gestão de Alunos - A Plataforma e360, https://repositorio.iscte-iul.pt/bitstream/10071/27377/1/master_fernando_mateus_lopes.pdf.
[14] Molenaar, I. (2021), “Personalisation of learning: Towards hybrid human-AI learning technologies”, in OECD Digital Education Outlook 2021, Pushing the Frontiers with Artificial Intelligence, Blockchain and Robots, OECD Publishing, https://doi.org/10.1787/589b283f-en.
[3] National Forum on Education Statistics (2010), Traveling Through Time: The Forum Guide to Longitudinal Data Systems. Book One of Four: What is an LDS?.
[4] OECD (2023), Country Digital Education Ecosystems and Governance. A companion to Digital Education Outlook 2023, OECD Publishing, https://doi.org/10.1787/906134d4-en.
[7] OECD (2023), PISA 2022 Creative Thinking Framework, OECD Publishing, https://doi.org/10.1787/dfe0bf9c-en.
[8] OECD (2021), Beyond Academic Learning: First Results from the Survey of Social and Emotional Skills, OECD Publishing, https://doi.org/10.1787/92a11084-en.
[17] OECD (2019), Measuring Innovation in Education 2019: What has changed in the classroom?, OECD, https://doi.org/10.1787/9789264311671-en.
[15] OECD (2019), OECD Learning Compass 2030, OECD Publishing, https://www.oecd.org/education/2030-project/teaching-and-learning/learning/learning-compass-2030/OECD_Learning_Compass_2030_concept_note.pdf.
[5] OECD (2018), The Future of Education and Skills. Education 2030, OECD, Paris, http://www.oecd.org/education/2030/E2030%20Position%20Paper%20(05.04.2018).pdf (accessed on 28 November 2018).
[12] Siemens, G. (2013), “Learning Analytics: The Emergence of a Discipline”, American Behavioral Scientist, Vol. 57/10, pp. 1380-1400, https://doi.org/10.1177/0002764213498851.
[16] Vidal, Q. (2023), “Digital assessment”, in OECD Digital Education Outlook. Towards an Effective Digital Education Ecosystem, OECD Publishing, https://doi.org/10.1787/c74f03de-en.
[2] Vincent-Lancrin, S. (2023), “Learning management systems and other digital tools for system and institutional management”, in OECD Digital Education Outlook 2023. Towards an Effective Digital Education Ecosystem, OECD Publishing, https://doi.org/10.1787/c74f03de-en.
[11] Vincent-Lancrin, S. and C. González-Sancho (2023), “Data and technology governance: fostering trust in the use of data”, in OECD Digital Education Outlook 2023. Towards an Effective Digital Education Ecosystem, OECD Publishing, https://doi.org/10.1787/c74f03de-en.
[10] Vincent-Lancrin, S. and C. González-Sancho (2023), “Interoperability: unifying and maximising data reuse within digital education ecosystems”, in OECD Digital Education Outlook 2023. Towards an Effective Digital Education Ecosystem, OECD Publishing, https://doi.org/10.1787/c74f03de-en.
[6] Vincent-Lancrin, S., C. González-Sancho, M. Bouckaert, F. de Luca, M. Fernández-Barrerra, G. Jacotin, J. Urgel and Q. Vidal (2019), Fostering Students’ Creativity and Critical Thinking: What it Means in School, Educational Research and Innovation, OECD Publishing, Paris, https://doi.org/10.1787/62212c37-en.
← 1. https://www.myschool.edu.au/; https://www.ontario.ca/page/find-your-school; https://cercalatuascuola.istruzione.it/cercalatuascuola/; https://www.schoolinfo.go.kr; https://www.udir.no/kvalitet-og-kompetanse/stastedsanalyse/om-stastedaanalysen-for-skoler/.
← 2. Complementary to the rich details obtained through the survey about the capabilities and data maintained by the systems, interviews with countries as well as a past survey on longitudinal information systems (González-Sancho and Vincent-Lancrin, 2016[1]), the section builds on a series of international meetings to gain further insights into the current and potential roles of these information systems and to foster peer learning among chief data officers and system managers across countries. Two workshops were held in New York City in June 2010 (in collaboration with the US Social Science Research Council (SSRC) and the Stupski Foundation) and in 2014 (hosted by Barnard College, Columbia University). A third meeting was organised in Washington DC in December 2015 in collaboration with the American Educational Research Association (AERA). Overall, more than 100 participants from 25 countries attended these meetings. The OECD Secretariat continues to be active and participates in the expert groups organised by the United Nations Educational, Scientific and Cultural Organisation (UNESCO).
← 3. See for example the US strategy about these technological advances: https://www.whitehouse.gov/wp-content/uploads/2023/03/National-Strategy-to-Advance-Privacy-Preserving-Data-Sharing-and-Analytics.pdf.