Statistical Office of the Slovak Republic (SOSR)

Labour and Education Statistics Department

Statistical Office of the Slovak Republic

Lamačská cesta 3/C

840 05 Bratislava 45

Slovakia

The Adult Education Survey (AES) covers adults’ participation in education and training (formal - FED, non-formal - NFE and informal learning - INF). The 2022 AES focuses on people aged 18-69. The reference period for the participation in education and training is the twelve months prior to the interview.

Information available from the AES is grouped around the following topics:

• Participation in formal education, non-formal education and training and informal learning
• Volume of instruction hours
• Characteristics of the learning activities
• Reasons for participating
• Obstacles to participation
• Access to information on learning possibilities and guidance
• Employer financing and costs of learning
• Self-reported language skills

For further information see the 2022 AES legislation (http://ec.europa.eu/eurostat/web/education-and-training/legislation) and the 2022 AES implementation manual (http://ec.europa.eu/eurostat/web/education-and-training/methodology).

- Classification of Learning Activities (CLA, 2016 edition)
- International Standard Classification of Education 2011 (ISCED 2011)
- Classification of Occupations 2008 (ISCO 08)
- Classification of economic activities Rev. 2 (NACE Rev. 2)

AES covers all economic sectors.

Definitions as well as the list of variables covered are available in the 2022 AES implementation manual (http://ec.europa.eu/eurostat/web/education-and-training/methodology).

Individuals, non-formal learning activities.

Individuals aged 18-69 living in private households.

Data refer to all Slovakia.

AES 2022 fieldwork:01/07/2022 - 30/11/2022

AES 2016 fieldwork: 01/07/2016 - 30/11/2016

AES 2011 fieldwork: 01/10/2011 - 15/11/2011

pilot AES 2007: NA

Not applicable.

Number, EUR.

Fieldwork period: 01/07/2022 - 30/11/2022

At European level:

Basic legal act: Regulation (EU) 2019/1700

Implementing act: Commission Implementing Regulation (EU) 2021/861

At national level:

The programme of state statistical surveys

Ensuring the processing of statistical surveys and administrative sources - consolidated (sk pdf)

Not applicable.

• The protection of confidential data is regulated by Act no. 540/2001 Coll on state statistics, as amended.
• The protection of personal data is regulated by Act no. 18/2018 Coll. on the protection of personal data and on the amendment of certain laws.

• Minimum frequency rule with set value equal to 3.
• If the number of the statistical units in any cell of the table formed by aggregating the individual data is less than three the data in the concerned cell is considered confidential. Confidential data are under high protection from direct or indirect identification and cannot be disseminated.
• Most frequent procedures for reducing the risk of disclosure are cell suppression and cell aggregation.

The release calendar for the statistical outputs is publicly accessible in Slovak and English version on the internet website of the SOSR.

First release calendar

From 2022, every 6 years.

No news release is scheduled at the moment.

The AES results will be available only in the online database.

The key AES results are available since 31st October 2023 in the online database DATAcube administrated by the SOSR.

Not applicable.

There will be possibility to obtain the AES microdata in anonymised format for scientific purposes under the strict conditions.

Other data dissemination has not been applied.

Not applicable.

 List of national documentation (see annexes):

Not applicable.

Quality policy of the SOSR

Quality Declaration of SOSR

Quality assurance in general:

Statistical Office of the Slovak Republic holds certificate in the area quality management system and information security management system, which confirms that the SOSR meets requirements of the international standard ISO 9001:2015 in organising, obtaining, processing and providing official statistics according to the current standards. 

The procedures to promote general quality management principles in the organisation and the quality assurance applied for the survey include the electronic guideline on quality management, which thoroughly describes the quality policy of the Statistical Office of the Slovak Republic. The electronic guideline on quality management system is written only in the Slovak language.

Quality assurance applied for 2022 AES:

- Preparation of the detailed organisation framework monitoring all the phases of the project implementation from the beginning to the end. This framework contained:

• Detailed description of all the project phases and their main tasks,
• Identification of departments responsible for fulfilments of each phase and project tasks,
• Deadline for fulfilment of each phase and project tasks.

- Quality assessment of 2016 AES and identification of its weaknesses,

Based on the mentioned directive, quality of the statistical outputs fulfils the standard quality criteria: relevance, accuracy, reliability, timeliness, punctuality, comparability and coherence.

Once a year, a customer satisfaction survey is conducted with the products and services of the Statistical Office of the Slovak Republic.

Currently, there is no information on any lower level of user satisfaction.

All the variables required for transmission have been included in the microdata.

Indicator R1 (data completeness rate) for all variables = 100%.

Not applicable.

The overall accuracy can be considered as reliable. Sample size was calculated in order to fulfil pre-defined criteria according to sample and precision requirements and expected confidence intervals.

In comparison with 2016, the AES 2022 gross sample size was enlarged by 17.9 pp. By using the fresh 2021 Population and Housing Census we managed to reduce coverage error.

There is no evidence of measurement and processing errors in the final statistical outputs. 

In general the sampling errors for 2022 AES can be considered as satisfactory. Compared to 2016, the accuracy of 2022 AES is better for almost all key statistics.

Higher values of coefficient of variation and standard error appear in the groups with small number of respondents (the unemployed, respondents with low educational attainment). However, these groups have low representation even in the total population, so possibility to enlarge them in future is minimal.

Higher values of coefficient of variation and standard error of indicators 'cost of non-formal learning activities' and 'hours spent in formal activities' are affected by higher item non-response rate of variables. On the other hand, they are much better then in 2016 AES.

SAS software, procedure SURVEYMEANS was used for calculation of the coefficients of variation, the standard errors and the confidence intervals.

See table 13.2.1 “Sampling errors - indicators for 2022 AES key statistics” in annex “SK - QR tables 2022 AES (excel)”.

SAS software, procedure SURVEYMEANS was used for calculation of the coefficients of variation, the standard errors and the confidence intervals.

 Non-sampling errors are covered by items 13.3.1 - 13.3.5 below. No additional information.

The total frame population was 5 894 persons, of which 5.1% population was ineligible. The source of the frame population data was 2021 Population and Housing Census.

Compared to AES 2016, we managed to reduce over-coverage rate significantly (by 11.9 p.p.) by using the fresh Census 2021 results. 

Experiences with last AES waves showed the importance of further updates of Census databases for the next AES wave in order to reduce possible coverage errors to minimum.

See table 13.3.1.1 “Over-coverage - rate” in annex “SK - QR tables 2022 AES (excel)”.

Not applicable.

• Household type and age structure of household members,
• Variables HHTYPE and MARSTADEFACTO,
• AGE, HATLEVEL and HATYEAR,
• AGE, HATLEVEL, HATYEAR and FEDLEVEL,
• Problems with misclassification of sub-category "other" in multi-choice variables (FEDREASON, FEDOUTCOME, NFEREASONx, NFEOUTCOMEx) as more detailed description of main sub-categories were missing. Solved by correction according to respondent description of answer category "other", in the next AES the methodology should be improved. 
• All possible measurement errors were verified by the interviewer with the respondent, the errors were corrected.

• Systematic testing of the new and problematic variables and questions on central and regional level,
• Simplification of wording in questionnaire,
• Using the introductory texts and the simple guidelines in questionnaire,
• Preparation of detailed 2022 AES national methodological manual for interviewers,
• Using the visual cards in order to reduce possible errors in questions with numerous answer categories,
• Training of interviewers and video presentations of guidelines for interviewers,
• All methodological issues that arose during the data collection were consulted with central level. Adopted solution and instructions were distributed to all interviewers,
• Systematic testing of electronic version of the questionnaire and software for data collections,
• Implementation of build-in checks and controls into the electronic version of the questionnaire and software for data collection,
• Manual of checking rules integrated to Blaise software for data collection were prepared for interviewers,
• According to instructions, all errors and warnings had to be verified. In case of acceptation, interviewers had to give explanation of acceptation,
• The collected data were checked continuously at the regional level. Basic checks and controls were performed on partial data by regional staff each month and sent to the central level for further data controls and analyses.

See table 13.3.3.1 “Unit non-response - rate” in annex “SK - QR tables 2022 AES (excel)”.

See table 13.3.3.2 “Item non-response rate” in annex “SK - QR tables 2022 AES (excel)”.

No evidence of processing errors in the final dataset transmitted to Eurostat. Before transmission to Eurostat, the dataset was validated by the data validation service STRUVAL and CONVAL. All error messages left in transmitted dataset have been verified and approved.

No specific models used to define the target of estimation.

The date of dissemination at the national level is the end of October 2023.

Not applicable as there are only one, directly final set of results/statistics.

T+11 months

16 days ahead.

See table 14.2 “Project phases - dates” in annex “SK - QR tables 2022 AES (excel)”.

Not applicable.

See table 15.1 “Deviations from 2022 AES concepts and definitions” in annex “SK - QR tables 2022 AES (excel)”.

Not applicable.

• Sample of individuals,
• Existence of substitution sample in order to reduce possible non-response rate,
• Use of proxy answers in order to reduce possible non-response rate,
• Preference of computer assisted data collection mode,
• Introduction of modularisation for better orientation and easier use of the questionnaire, 
• Introduction of new approach for data collection of multiple-choice variables. In 2016, we used yes/no checklist for data collection of multi-choice variables. This method was quite criticised, so we decided to use single checklist in almost all multi-choice variables.
• Redesign of variables FEDNBHOURS, NFENBHOURS. While in AES 2016, variables on instruction hours were surveyed by a single question, in AES 2022 we divided variables into four individual questions. Respondents were gradually asked about the number of weeks, weekly number of hours and length of instruction hours,
• Redesign of variable NFE. In AES 2016, we designed variable NFE as multiple-choice variable with yes/no checklist. According to recommendation in AES 2022 manual, we change the concept for identification of four types non-formal learning, and we used an individual question for each type of non-formal learning activity,
• Collection of variable NFENUM for each type of non-formal learning by separated question and not only by one question as it was in AES 2016,
• Redesign of module 'obstacles to participation in education and training'. In AES 2016, due to negative way of asking the questions, it was considered as the weakest part of the questionnaire. Therefore we decided to change of philosophy of the asking questions, use only one word question (why?) and to convert the proposed question into answer category.

See table 15.2 “Comparability - over time” in annex “SK - QR tables 2022 AES (excel)”.

Not applicable.

AES results for a given data collection round are based on the same microdata and results are calculated using the same estimation methods, therefore the data are internally coherent.

At the moment, there is no exact information on cost.

The staff involved in administering the survey (without the interviewers) was 1.25 in full time equivalent.

Number of interviewers involved in field work was 11.6 persons in full time equivalent.

The average time used for answering the survey was 32 minutes.

The increase of sample size and voluminous questionnaire have negative impact on respondent's burden and costs for fieldwork.

Not applicable.

Not applicable.

Not applicable.

See table 18.1 “Source data” in annex “SK - QR tables 2022 AES (excel)”.

Every 6 years.

• Manual of checking rules for possible errors,
• Build-in field level checks,
• Build-in logical checks and controls,
• Build-in checks for potential coherence issue among variables,
• Every methodological issue/error that arose during the data collection was consulted with central level. Solution and instruction were distributed to each interviewer.

• The first stage of data validation at regional level was the competency of the Department of sample survey statistics in Banská Bystrica. The role of this department was to process data of all AES 2022 questionnaires, to perform checks and validation at the regional level, the finalization of the regional dataset and its transmission to the Central Statistical Office. Besides the number of build-up checks in electronic questionnaire and Blaise software for data collection and processing, the set of validation rules were also prepared outside the Blaise. The set of checks were prepared in SAS enterprise guide for verification correctness of values, identification of potential logical relations and potential coherence among variables. During the fieldwork, the data was controlled continuously each month. Final regional dataset was sent to central level for further checks and validation.
• The second stage of data validation was performed at the central level on the Labour and Education Statistics Department. The final regional dataset was analysed and verified in SAS enterprise guide by set of validation rules. The set of global checks covered the completeness checks, coding correctness checks, routing/filter checks, format checks, logical checks and checks for coherence issue among variables. If error was detected, information was verified at the regional level. After validation, the final national datafile for transmission to Eurostat was compiled.
• The last stage of validation was performed on the final national dataset transmitted to Eurostat by the data validation service STRUVAL and CONVAL.

• sex x age classes (18-24, 25-39, 40-59, 60-69)
• education (low, medium, high) x age classes (18-24, 25-69)
• labour status (employed, unemployed) x age classes (18-24, 25-69)

No imputation was used.

See table 18.5.1 “Imputation - rate” in annex “SK - QR tables 2022 AES (excel)”.

Not applicable.

Not applicable.

None.