Adult Education Survey (household survey)

The Adult Education Survey (AES) is a pan-European personal survey coordinated by Eurostat, which examines the state of lifelong learning in the context of formal education, training, and self-directed learning. The survey results provide valuable insights at both the national and European levels to understand how many and what types of learning opportunities people utilise, what motivates or hinders them from further education, and whether any societal groups may require additional attention in the context of lifelong learning. Therefore, the survey results are instrumental in shaping national policies on adult education.

Probability survey

Participation in adult education by sex, age group, mother tongue, educational level, region, place of residence, social status and household’s net income; knowledge of foreign languages of population aged 20–64. Adult education is divided into three types: formal studies, in-training, and independent self-development.

Classification of Estonian administrative units and settlements (EHAK);

Estonian Classification of Economic Activities (EMTAK) based on NACE Rev. 2;

Statistical Classification of Regional Units of Estonia;

International Standard Classification of Occupations (ISCO 08);

International Standard Classification of Education (ISCED 2011);

Classification of Ethnicities;

International Standard Codes for the Representation of the Names of Countries (ISO 3166);

Codes for the Representation of Names of Languages (ISO 639-2)

Not applicable

The definitions are entirely based on the AES guidance materials (see the AES2022 Eurostat Manual and the CLA Manual).

Basic level of computer skills – ability to use word processing and spreadsheet programs, to copy files/folders or change their location, etc.

Blue-collar worker – service and sales workers; skilled agricultural, forestry and fishery workers; craft and related trade workers; plant and machine operators and assemblers; elementary occupations; armed forces occupations

Expert level of computer skills – ability to write computer programs, solve software and hardware problems if a computer does not work as it should, etc.

Foreign language – a language that is not a mother tongue

Foreign language knowledge – beginner and advanced level knowledge of a foreign language. Beginner level is defined as understanding and being able to use a few words and phrases, even if only to a small extent.

Formal education – basic, general, vocational and higher education. This is an institutionalised, structured form of learning that conforms to specific standards and typically takes place in a school environment based on level curricula. Formal education is purposeful, and it is guided by teachers with specialised preparation and qualifications. Learning objectives derive from the curriculum and the teacher, and the learning process is monitored and evaluated. Formal education is mandatory up to a certain level or age (in Estonia, until the age of 17).

Household – a group of people who live in a common dwelling (at the same address), share joint financial and/or food resources and whose members consider themselves to belong to the same household. A household may also consist of one member only.

Lifelong learning – participation of adults in education, i.e. participation in formal, non-formal, or informal learning

Minimum wage – the nationally established minimum monthly wage in the year of the survey

Mother tongue – the first language that is spoken in early childhood. A person may have more than one mother tongue.

Net income – a total sum of income from wage labour and self-employment, property income, social transfers, regular inter-household cash transfers received, and receipts for tax adjustments, from which inter-household cash transfers paid, taxes on wealth, and repayments for tax adjustments have been subtracted

Net income per household member – the sum of all household members' net incomes divided by the number of household members

Person with below upper secondary education – a person whose highest level of education is preschool education, primary education, basic education, vocational basic education, or vocational education after basic education

Person with tertiary education – a person whose highest level of education is secondary specialised education after secondary education, bachelor’s, master’s or doctoral level

Person with upper secondary education – a person whose highest level of education is general secondary education, vocational secondary education, or vocational education after secondary education

Primary sector – agriculture, hunting, forestry, fishing

Proficient level of computer skills – ability to format text, create graphs with spreadsheet programs, install simpler devices and programs, etc.

Rural settlement – a small town and village

Secondary sector – mining and quarrying, manufacturing, electricity, gas and water supply, construction

Self-study (intentional self-development) – also known as informal learning. It is self-directed and pre-planned self-improvement. It encompasses any kind of conscious learning through various activities and communication channels, regardless of time and environment. Informal learning is not structured and lacks direct learning objectives, curriculums, and materials. There is no institution organising the teaching. Informal learning does not include participation in trainings or formal education system studies.

Tertiary sector – trade, services, etc.

Training – courses, seminars, private lessons, or guide-on-the-job training. Also known as non-formal education. Non-formal education is associated with both professional self-improvement and personal interests. It is goal-oriented and institutionalised learning that takes place in various environments outside of formal education. Non-formal learning can be conducted by educational institutions, training companies, employers, and private teachers. Trainings are prepared by the instructor, and there is usually also a training plan.

Urban settlement – a city, city without municipal status, and town

White-collar worker – legislators, senior officials and managers; technicians and associate professionals; clerks

Person

Since 2022, the target population consists of Estonian permanent residents aged 18–69 living in private households. Individuals living in institutional households (e.g. nursing homes, care homes, prisons) are excluded from the target population. The target population includes both people who have participated in education and those who have not. Until 2016 (inclusive), the age range of the target population was 20–64.

Estonia as a whole

2007, 2011, 2016, 2022

Not applicable

Participants/non-participants in education – absolute number;

the share of participants/non-participants in education – percentage (%)

The reference period for acquiring education is one year, specifically a 12-month period prior to the time of the interview. As the survey is conducted over a period of 6 months, the responses in the results cover a maximum of an 18-month period. For example, the 2022 survey in Estonia was conducted from 1 July to 31 December 2022. Therefore, the respondents' answers regarding their training experiences cover the period from 1 July 2021 to 31 December 2022.

Data on background characteristics (occupation, workplace, household type, highest education level, health) and language skills are collected as at the time of the interview.

Survey periods:
September 2007 – December 2007;
17 October 2011 – 16 January 2012;
1 July 2016 – 31 December 2016;
1 July 2022 – 31 December 2022.

Official Statistics Act;

Regulation (EC) No 452/2008 of the European Parliament and of the Council of 23 April 2008 concerning the production and development of statistics on education and lifelong learning;

Regulation (EU) 2019/1700 of the European Parliament and of the Council of 10 October 2019 establishing a common framework for European statistics relating to persons and households, based on data at individual level collected from samples, amending Regulations (EC) No 808/2004, (EC) No 452/2008 and (EC) No 1338/2008 of the European Parliament and of the Council, and repealing Regulation (EC) No 1177/2003 of the European Parliament and of the Council and Council Regulation (EC) No 577/98 (Text with EEA relevance);

Commission Delegated Regulation (EU) 2020/256 of 16 December 2019 supplementing Regulation (EU) 2019/1700 of the European Parliament and of the Council by establishing a multiannual rolling planning (Text with EEA relevance);

Commission Delegated Regulation (EU) 2021/859 of 4 February 2021 supplementing Regulation (EU) 2019/1700 of the European Parliament and of the Council by specifying the number and titles of the variables for the data set in the education and training domain (Text with EEA relevance);

Commission Implementing Regulation (EU) 2021/861 of 21 May 2021 specifying the technical items of the data set and establishing the technical formats for transmission of information on the organisation of a sample survey in the education and training domain pursuant to Regulation (EU) 2019/1700 of the European Parliament and of the Council (Text with EEA relevance);

Commission Implementing Regulation (EU) 2019/2181 of 16 December 2019 specifying technical characteristics as regards items common to several datasets pursuant to Regulation (EU) 2019/1700 of the European Parliament and of the Council (Text with EEA relevance);

Commission Implementing Regulation (EU) 2019/2180 of 16 December 2019 specifying the detailed arrangements and content for the quality reports pursuant to Regulation (EU) 2019/1700 of the European Parliament and of the Council (Text with EEA relevance)

None

The dissemination of data collected for the purpose of producing official statistics is guided by the requirements provided for in § 32, § 34, § 35 and § 38 of the Official Statistics Act.

On European level, Regulation (EC) No 223/2009 on European statistics (recital 24 and Article 20(4)) of 11 March 2009 (OJ L 87, p. 164), stipulates the need to establish common principles and guidelines ensuring the confidentiality of data used for the production of European statistics and the access to those confidential data with due account for technical developments and the requirements of users in a democratic society.

The dissemination of data collected for the production of official statistics is based on the requirements laid down in §§ 34 and 35 of the Official Statistics Act.
The principles for handling confidential data can be found on Statistics Estonia's website under Data Protection.

Only the interviewer and the survey manager know the respondent's name and telephone number. The collected data are used only in aggregate form; no one's data are examined at the individual level. Analysts and researchers working with the survey data only have access to anonymised data. Statistics Estonia ensures the protection of all respondents' data based on the Official Statistics Act and the Personal Data Protection Act.

In the databases of Statistics Estonia and Eurostat, only estimates based on at least 20 respondents are published.

Notifications about the dissemination of statistics are published in the release calendar, which is available on the website. Every year on 1 October, the release times of the statistical database, news releases, main indicators by IMF SDDS and publications for the following year are announced in the release calendar (in the case of publications – the release month).

The release calendar is available to consumers on the website Calendar.

All users have been granted equal access to official statistics: dissemination dates of official statistics are announced in advance and no user category (incl. Eurostat, state authorities and mass media) is provided access to official statistics before other users. Official statistics are first published in the statistical database. If there is also a news release, it is published simultaneously with data in the statistical database. Official statistics are available on the website at 8:00 a.m. on the date announced in the release calendar.

Every sixth year

Not published

Not published

Data are published in the statistical database under the subject area Social life / Education / Adult education / Participation in adult education in all tables.

The database tables (HTT31 to HTT51) have been viewed 414 times during 2024, 1,139 times in 2023, 522 times in 2022, 429 times in 2021, 922 times in 2020, 1,364 times in 2019, and 873 times in 2018.

The dissemination of data collected for the purpose of producing official statistics is guided by the requirements provided for in § 33, § 34, § 35, § 36, § 38 of the Official Statistics Act.
Access to microdata and anonymisation of microdata are regulated by Statistics Estonia’s procedure for dissemination of confidential data for scientific purposes.

Anonymised microdata are also transmitted to the European Commission (Eurostat), which publishes aggregated data from all countries in the Eurostat database (Population and social conditions → Education and training → Participation in education and training → Adult learning → tables starting with 'trng_aes'). Until 2016 (inclusive), data were submitted to Eurostat for individuals aged 25–64, although for national needs, we also collected data for those aged 20–24. Since 2022, Eurostat requests data for individuals aged 18–69.

The data are submitted via the EDAMIS Web Portal no later than 6 months after the completion of data collection.

In addition to aggregated data, a quality report is submitted to Eurostat. The quality report is delivered to Eurostat on the ESS Metadata Handler platform within nine months of the end of the data collection period and it must be structured in accordance with the European Statistical System Standard Quality Report Structure.

The metadata related to the statistical activity were viewed 49 times (1 January 2024 – 31 December 2024). This figure does not represent the number of viewers, as the metadata may have been viewed multiple times by a single user.

Eurostat's guidelines for conducting the survey are available in the CIRCABC environment. Materials for the 2022 survey:

Guide;

Annexes to the guide;

Sample questionnaire.

This statistical activity is guided by the European Statistics Code of Practice – revised edition, 2017.
The quality guidelines are included in the survey manual. See "Documentation on methodology".

Pan-European metadata and country-specific reports can be viewed here.

To assure the quality of processes and products, Statistics Estonia applies the European Statistics Code of Practice and the Quality Assurance Framework of the European Statistical System (ESS QAF). Statistics Estonia is also guided by the EFQM Excellence Model and the requirements in section 7 “Principles and quality criteria of producing official statistics” of the Official Statistics Act.

Statistics Estonia performs all statistical activities according to an international model (Generic Statistical Business Process Model – GSBPM). According to the GSBPM, the final phase of statistical activities is overall evaluation using information gathered in each phase or sub-process; this information can take many forms, including feedback from users, process metadata, system metrics and suggestions from employees. This information is used to prepare the evaluation report which outlines all the quality problems related to the specific statistical activity and serves as input for improvement actions.

Results of the 2022 survey assessment:
* IESS set two quality criteria for AES 2022, one of which, 'participation of 18–24-year-olds in formal education', had a larger error than permitted. However, registry data provided an opportunity to validate the assessment obtained from the survey;
* the response rate is 64% (the target of 65% was nearly achieved). The response rate from a sample without selection bias was 65.2%;
* the non-response rate is very low across variables. There are a total of 256 Eurostat variables. For all variables, non-response rate is under 10%. For three variables the rate ranges between 5% and 10%, and for 205 variables it is under 1%;
* the methodology has been fully followed.

The main users of statistical information are policy makers of national level (Ministry of Education and Research, Ministry of Economic Affairs and Communications, Ministry of Social Affairs). The data might also interest the Estonian Qualifications Authority, various research institutions and university researchers.

Internationally, the main users of the data are the institutions of the European Union. In addition, through Eurostat, researchers around the world can apply for access to use the data.

To meet national needs, additional questions have been included in some cases. To identify such needs, a meeting with key users is organised before each new survey. For example, until the 2016 survey (inclusive), there was a separate section in the questionnaire regarding training for the unemployed. In the 2022 survey, there was no national request for adding questions.

All variables as requested by the legislation are covered.

All required indicators have been collected and calculated (completeness 100%).

The overall accuracy of the AES is considered good. The sampling design is chosen in line with EU recommendations. When designing the sample, it is ensured that it meets the precision requirements laid down in the regulations.

Datasets are representative of the population aged 20–64 until 2016 and 18–69 since 2022.

As the results are based on a sample of the population, they are subject to the usual types of errors associated with sampling techniques and interviews (sampling errors, non-sampling errors, measurement errors, processing errors and non-response errors).

The sampling error for the 2022 AES is relatively low. Stratified systematic sampling was used to draw the sample.

Estimates of the sampling error were calculated using the R package survey, utilising the final (calibrated) weights. Standard errors are calculated for the main indicators, which are published in the Eurostat quality report.

For example, the participation rate in non-formal and formal education (age group 25–69) was 41.9% with a standard error of 0.8. The participation rate in informal learning (age group 18–69) was 68.7% with a standard error of 0.72.

Since the sample was not designed for publication at the county level, but this was nevertheless done in table HTT351, standard errors have been calculated there for the estimates. Standard errors have not been published in the rest of the database tables.

Non-sampling errors are covered under items below.

The sample frame for the 2022 survey was a list of permanent residents of Estonia aged 18–69 compiled from the data of the 2011 Population and Housing Census and the Population Register. The age of the persons was determined as of 1 July 2022. The data were extracted on 1 May 2022 and the survey ran until 31 December 2024. To eliminate over-overage, individuals not belonging to the target group were removed from the sample. Over-coverage was mainly due to the sample persons that turned 70 before the interview. They were excluded from the sample.

Measurement errors may arise from the questionnaire (its structure, the wording of the questions, etc.), respondents, interviewers, and the data collection method. Although it is impossible to completely avoid all types of measurement errors in social surveys, Statistics Estonia has made efforts to reduce them as much as possible.

To minimise measurement errors, the AES questionnaire was evaluated by experts before data collection, and an interviewer's manual was prepared to explain specific terms and questions. AES was a standalone survey, and proxy responses were not allowed. Probable measurement errors may occur due to the long reference period – for example, respondents might not remember training events or details of training sessions that took place nearly a year ago. As a result, measurement errors may appear in variables that involve estimates of costs and time spent on learning activities.

See items "Unit non-response – rate" and "Item non-response – rate".

Computer-based data collection is used, which reduces the risk of coding and entry errors. Data validation occurs in two stages: an initial check during the interview with automated validation controls, followed by data cleaning using predefined validation formulas. Weighting and imputations are performed using R scripts.

Not applicable

The preliminary data are published 6 months after the end of the data collection (T + 180 days).

The data are published once. AES 2022 data were published in June 2023 (T + 180 days).

The data are published once. AES 2022 data were published in June 2023 (T + 180 days).

The data have been published at the time announced in the release calendar.

100% of the data has been published on time.

In Estonia, the data are geographically comparable across regional units, counties and the city of Tallinn, as well as by degree of urbanisation.

The data are comparable with the data of other European Union countries because a harmonised methodology is used.

Four waves of the survey have been conducted so far (2007 AES, 2011 AES, 2016 AES, and 2022 AES). The first AES was a pilot project and was carried out between 2005 and 2008 on a voluntary basis in 29 countries in the EU, EFTA (European Free Trade Association), and candidate countries. Since 2011, the AES has been supported by a European legal act and is therefore mandatory in all EU Member States.

The 2011 survey was different from the others in that it was not solely conducted by Statistics Estonia but in collaboration with a subcontractor. Additionally, the response rate that year was lower than usual. Consequently, there may be more deviations than usual in these data (the non-sampling error is greater). In order to get a more consistent overview over time, it may be advisable to exclude the 2011 data from the analysis.

Since 2022, tables HTT45, HTT46, and HTT47 have been amended. Prior to 2017, these tables were compiled in such a way that, if the person surveyed participated in more than one training course during the year, only the information of the training mentioned first was considered in the analysis. However, since 2022, the information all the trainings attended during the year is taken into account.

Starting from the 2022 survey, the training courses for which more detailed data were collected from an individual were randomly selected. Previously, individuals had the opportunity to choose which training they provided information about.

Not applicable

Participation in lifelong learning is also measured in the Labour Force Survey.

Adult training in the workplace is examined in the enterprise survey "Continuing Vocational Training Survey".

In 2011, the results of the adult education survey showed that the percentage of people participating in training in the previous 12 months was the highest ever. However, this conclusion does not align with the results from the Estonian Labour Force Survey, which has shown a steady increase in participation in lifelong learning, without a noticeable change in 2011. The Labour Force Survey data are about training events that occurred during a 4-week period, but the trends can still be compared.

Not applicable as data are not published for periods shorter than one year.

Not applicable as not directly usable in national accounts.

The data are internally coherent. The internal coherence of the data is ensured by the use of a common methodology for data collection and processing. The survey results for a given reference year are based on the same microdata and are calculated using the same estimation methods.

The main cost of the survey is the cost of the staff involved in conducting the survey (project management, sample design, questionnaire development, training the interviewers, calculation of weights, data processing, data analysis, and quality report) ~ 375 full-time equivalent (FTE) working days.
The total fieldwork time spent conducting CATI interviews, which involves direct contact with respondents, was approximately 129 full-time equivalent (FTE) working days. However, the time interviewers spent on training, preparation, attempted calls, and other indirect tasks, was not measured.

To assess the burden on respondents, we can look at the average time spent answering the questionnaire.
1) All interviews:
CAWI: 32 minutes (883 interviews);
CATI: 18 minutes (3,477 interviews),
2) interviews where the person had participated in at least two non-formal learning activities:
CAWI: 41 minutes (353 interviews);
CATI: 27 minutes (716 interviews).

Not applicable

Not applicable

Not applicable

This is a combined activity: the data are based on a personal survey and on administrative data.

In 2007, the target population of the survey consisted of persons aged 20–64. The number of objects in the target population was 810,800, and the sample size was 6,000. The sampling method: systematic stratified sampling by sex and age.

In 2011, the target population of the survey consisted of persons aged 20–64. The number of objects in the target population was 832,000, and the sample size was 5,985. The sampling method: random stratified sampling by sex and age (8 strata).

In 2016, the target population of the survey consisted of persons aged 20–64. The number of objects in the target population was 789,000, and the sample size was 6,000. The sampling method: random stratified sampling by sex and age (10 strata).

In 2022, the target population of the survey consisted of persons aged 18–69. The number of objects in the target population was 873,000, and the sample size was 6,800. The sampling method: random stratified sampling by sex and age (10 strata).

The following background characteristics are obtained from the Population Register: sex, year of birth, age, county of residence, and type of residence. Administrative data may also be used for imputing missing data (e.g. if household income is not provided). There are no substantial prefillings from other data sources.

Until 2016: every five years. Since 2016: every six years.

Data are collected using the questionnaire 1409 "Adult Education". You can also read a brief overview of the survey on the website.

Before the survey begins, all selected participants receive a letter introducing the survey’s purpose and explaining how they will be contacted. The notification letter is sent to the participant's email address as registered in the Population Register. If no email address is available, the letter is sent by mail to the address listed in the Population Register. The survey can be completed in either Estonian or Russian. Most of the questions in this survey are about the sample person; only the number, age, and household income of other household members are asked. Proxy interviews are not allowed – no one else can answer on behalf of the sample person.

The data collection environment used is VVIS (Fieldwork Information System), which includes checks to ensure data quality. If data are illogical or required information is missing, corrections must be made before proceeding with the questionnaire. In some cases, a warning is displayed asking to review the data, but corrections are not mandatory. The questionnaire is prefilled with the sample person's sex and birth date. The questionnaire is divided into several sections, and each new section opens only after the previous one is completed. If the questionnaire is left unfinished, it can be resumed later. The completed form must be confirmed to finalise it, and it cannot be reopened after confirmation.

Since 2022, participants have the option to respond independently online, and customer support is available for any questions. Those who do not complete the online survey are contacted by trained interviewers from Statistics Estonia, and the questionnaire is completed via telephone interview. Face-to-face interviews have not been conducted since 2022.

Breakdown of data collection methods by survey year:
2007 – 100% face-to-face interview (CAPI);
2011 – 50% telephone interview (CATI), 50% face-to-face interview (CAPI);
2016 – 93% telephone interview (CATI), 7% face-to-face interview (CAPI);
2022 – 80% telephone interview (CATI), 20% web interview (CAWI).

After the interview is completed, the data are transferred for processing, where the collected data are verified, and the necessary classification codes are added. Validation includes both arithmetic and qualitative checks. When errors are detected, either manual corrections are made or automatic data correction packages are created.

The data were also validated in Eurostat's data transmission program EDAMIS in order to conduct field level and record level checks. Field level checks monitor whether valid codes and ranges are used and check the consistency between a variable entry and allowed entries, whereas record level checks test the consistency between variables for a single enterprise record.

Imputation
In the case of missing or unreliable data, estimate imputation based on established regulations is used. For household income with high non-response we used the following method: Respondents were asked to either provide their exact household income or indicate the interval to which their household income belongs. In case of both being unknown the interval is first imputed using k-nearest neighbour imputation using the VIM package in R (the number of nearest neighbours was set as 5 and household size and type and the respondent's age, sex, county + Tallinn, and educational attainment were used as distance variables). Once every respondent was given an income interval, the exact household income was imputed using hot-deck imputation.

Weights
After the data processing stage, each respondent is assigned a weight, i.e. it is determined how many people their responses represent. Weights allow the results obtained during analysis to be extrapolated to the general population.
Weights are calculated in three stages:
calculation of design weights;
compensation for loss;
calibration.

Calibration is based on sex and age group (five-year age groups), ethnic nationality (since 2005) (Estonians, non-Estonians), three-tier level of education (since 2022), person's county of residence (+Tallinn), and the degree of urbanisation (rural or urban). The basis is the distribution of the Estonian population by sex, age group, and county as of 1 January of the survey year, as known from demographic data.

After calculating calibration weights, a general expansion factor is found for each responding person, which is the ratio of the total population size to the number of respondents. To get the final weight for the person, this expansion factor is multiplied by the previously calculated weights.

Calculated variables
Based on the variables collected with the questionnaire, the following were calculated: family type, equivalised disposable income quintile, number of foreign languages, training weights.

In the 2022 survey, one variable (household income) was imputed. The imputation rate was 37%.

Not applicable

Not applicable

None