Verian Deutschland; registered as Mantle Germany GmbH; formerly known as Kantar Public

Verian Deutschland, unit: labour research and vocational training

Verian Deutschland - Mantle Germany GmbH

Landsberger Straße 284

80687 München

Germany

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.

Germany

AES 2007 (2/03/2007 - 7/07/2007)

AES 2010 (2/03/2010 - 30/06/2010)

AES 2011 (national name: AES 2012) (1/03/2012 - 30/06/2012)

AES 2014 (13/05/2014 - 14/07/2014)

AES 2016 (1/07/2016 - 26/10/2016)

AES 2018 (16/07/2018 - 2/12/2018)

AES 2020 (24/07/2020 - 08/11/2020)

AES 2022 (12/07/2022 - 20/03/2023)

Not applicable.

Number, EUR.

2022 AES fieldwork: 12/07/2022 - 20/03/2023

At European level:

Basic legal act: Regulation (EU) 2019/1700

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

At national level:

- 

Not applicable.

• Data protection concept and technical and organizational measures: see Points 1 to 4 in Annex "Data protection concept: Organization and implementation of data protection"
• Certificates: see Point 5 in Annex "Data protection concept: Organization and implementation of data protection"

Data and documents are released for academic research and academic teaching via GESIS (https://www.gesis.org/en/home).

Data and documents are transferred to GESIS after publication of the national trend report.

Bundesministerium für Bildung und Forschung (BMBF), Berlin (o.J.): Adult Education Survey (AES 2022 - Germany). GESIS Datenarchiv, Köln. ZA8771 Datenfile Version 1.0.0, doi:10.4232/1.14234

(to note: The DOI is reserved for the AES, but can only be used once the data has been published.)

The data are released only for academic research and teaching. The data including study description and documentation at variable level are published in an analysis portal (currently ZACAT-Online Study Catalogue). Data can only be downloaded by registered users. The data are made available via password-protected online accesses for which registration is required. The user may only process and use the data provided for the purpose of academic research and teaching. Any processing or use for other purposes requires a separate prior written agreement. The user must take appropriate technical and organizational measures to ensure that only the data recipient has access to the data provided. Unauthorized disclosure of the data provided to third parties is not permitted.

Every 6 years.

Not yet.

Trend report (Publication at the beginning of 2024): Bundesministerium für Bildung und Forschung (BMBF) (2023): Weiterbildungsverhalten in Deutschland 2022. Ergebnisse des Adult Education Survey - AES Trendbericht, Bonn 

(URL: probably published on BMBF’s homepage)

GESIS (additionally to the German dataset there are tables provided for data users).

Not applicable.

Microdata will be available in the beginning of 2024 at Data Archive hosted by GESIS.

German AES-data is used as basic information especially on further education of individuals and therefore, in a couple of national statistical reports, e.g. "Bildung in Deutschland", "Bildung und Forschung in Zahlen", "Datenreport" of the Federal Statistical Office (destatis, Statistisches Bundesamt), "BIBB-Datenreport zum Berufsbildungsbericht".

Not applicable.

GESIS short version of the Fieldwork Report (will be published together with the data at the GESIS website).

Not applicable.

See attachment.

Comment: In the national survey for Germany the respondents were asked directly which gender they belong to. The following three categories were asked: male, female, divers. A total of two people stated "diverse" as their gender. These two people were excluded from the EU dataset. The case numbers between the national dataset and the EU dataset therefore differ.

Our certifications: ISO 9001, ISO 20252, and ISO 27001 certification

Systematic activities: weekly monitoring of fieldwork progress, interviewer control, interview duration checked, data editing and validation (output quality proved and checked), sample structure checked (weighted and unweighted), punctuality (deadlines met, e.g. distance of time to the last AES), scientific advisory committee for quality assurance.

The German AES tries to serve two objectives:

1. to continue the national line of reporting on the participation in adult learning as represented by the series of BSW (Berichtssystem Weiterbildung, Reporting System on Continuing Education and Training) as well as (national) AES surveys and reports since 1979, and
2. to comply with the requirements of the AES requesting comparable data at European level.

According to preliminary evaluations, the German AES as implemented in 2022 successfully meets both objectives.

Measurement for FED activities cannot be taken directly from the AES-Manual (which only defines a very general variable here). Instead, it needs a specific module of questions reflecting the structure of the national educational system (as given in the German qualification frame (DQR)).

NFE activities also have to be measured in a way adapted to the understanding of such activities at the respective national level. This is possible, and the German case seems to support the impression that the CLA based measurement of NFE is accepted as reference.

INF activities were (repeatedly) not exactly measured in the way the AES-Manual suggested. The "tries" of informal learning as given in the AES-Manual were changed in favour to "intentional activities" (see question F130a in the German questionnaire). A recommendation of changing the indicator on the European level is still wishful.

Information according to the scientific community is collected at GESIS (general link to the data base: https://dbk.gesis.org/dbksearch/home.asp), where the national data is provided for users. There is a project network implemented of very experienced persons in this research domain and especially with the project. Additionally to the project network the scientific advisory committee is implemented for quality assurance purposes and to meet the demands according to the study from the scientific community. 

Before the AES 2016 there had been a study on adult education indicators in Germany which concluded with some good advice to improve the AES for national scientific purposes.

See 12.1 above.

All variables as required by the legislation are covered.

Not applicable.

The 2022 AES can be assessed as being successful regarding the overall accuracy.

The precision of an estimator, usually expressed by its standard error, depends, among other things, on the sample design, which for example defines the size of the sample, the selection probabilities, or multi-stage sampling including clustering. Some sample designs imply an increase of the standard error of the estimators and thus a decrease of their precision (design effect). A measure for the design effect is the effectiveness of the sample design indicating how much the sample size, i.e. the actual sample size, is decreased by implementing the corresponding design or otherwise stated how much lower is the effective sample size. If there is no design effect actual and effective sample size are the same. If there is a design effect, actual and effective sample size differ. The actual sample size of the target persons aged 25 to 69 is 7,270. As the effectiveness of the sample design showed to be 90.42 % the effective sample size amounts to 6,574.

The accuracy of the sample completely complies with the requirements of Eurostat and the EU regulations. The precision requirement for Germany is 0.72 for participation rate in non-formal education and training, age 25-69, which is met (0.61). The precision requirement for participation rate in formal education and training, age 18-24 is 0.96, which is also met (0.89).

With regard to basic information on the population see Annex "AES Germany: Methods (Extracted parts of the German report)" in section 10.7.

The coefficient of variation (CV) is calculated as the ratio of the standard error to the mean of each group multiplied by 100. We want to point out, that the CV is not ideal for percentage values as it is calculated with the percentage value in the denominator. Therefore, although the value of the standard error is symmetrically distributed around the percentage share with the maximum standard error at a percentage share of 50%, i.e. it is identical for values with the same distance to 50% above it or below it, e.g. for 60% and 40%, the CV is not symmetrical: The lower the percentage value the higher the CV (see also manual of the quality report, pages 36, 40-41, 121).

See table 13.2.1 “Sampling errors - indicators for 2022 AES key statistics” in annex “DE - QR tables 2022 AES (excel)”. All standard errors shown in table 13.2.1 include the design effect.

The design effect is taken as being constant (= 90.42%) and estimated using the variance of the weighting factors of the whole sample with 7,270 persons. This means, all standard errors shown are multiplied by 1.11 (=100%/90.42%). For this reason the confidence intervals given are conservative estimates, i.e. rather broader.

The software programs used for all the operations are Excel and SPSS.

In Germany, missing information is not imputed in the AES 2022 (unlike in the AES 2016) for the NFEPAIDVAL. Due to the filter specification (see sheet "Information" in "DE - QR tables 2022 AES (excel)") for the calculation of "Average amount paid by a participant for all the expenses related to the non-formal learning activities, age 18-69 - EUR": NFEPAIDVALx > 0), the costs per person are underestimated. Any missing information is included in the calculation as information on the costs of € 0.

See items 13.3.1 - 13.3.5 below.

Not part of the sampling frame: persons not living in private households.

Not interviewed: persons not able to answer the questionnaire in German.

Other ineligible includes unknown ineligibility, e.g. address unknown.

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

Not applicable.

Measures taken to prevent measurement errors: 

CAPI/CAWI:

• the entire programmed questionnaire (including plausibility questions) was tested and a data set of test interviews was checked for filtering, 
• concurrent validation of answers with a possibility to correct or specify given answers directly in the interview, 
• the question order is telling a story and it is easy to follow the interview, 
• interviewer/respondent hotline for all questions.

Additionally used for CAPI only:

• visualizing lists (paper) were additionally used for questions with a long list of answers in the interview, 
• every interviewer gets profound material with explanatory notes about the study and the characteristics and importance of the study, 
• control of interviewers due to quality assurance.

Quality assessment: The response rate meets the expected target.

Measures taken to reduce unit non-response: 

• an invitation letter including a recommendation of the Federal Ministry and up to two reminders were sent to the target persons
• to increase the participation of persons with migration background the invitation letter and reminder letters were sent in Turkish, Polish and Russian besides German
• a post-paid incentive of 20€ was paid
• CAPI fieldwork took place also in the evening and on Saturdays to be able to better reach and interview hard to reach persons (e.g. employed persons).
• use of very experienced and well trained interviewers
• specific interviewer instructions were handed out to the interviewers with additional information on the concept of the study as well as on specific questions
• a hot line via email or telephone informed or supported the target persons with regard to all questions or problems they had

As a result, there was hardly any bias in regional characteristics as well as regarding age, sex, and nationality. A bias, however, could be observed in terms of education, in particular the lower and medium education level was under-represented. Using calibration weights the bias was eliminated.

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

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

Data entry and coding control mainly via CAWI or CAPI, i.e. computer assisted. For detailed information see Annex "AES Germany: Methods (Extracted parts of the German report)" in section 10.7.

Main errors detected in the post-data checking according to detailed information on educational attainment. As described in chapter A1.5 in Annex "AES Germany: Methods (Extracted parts of the German report)" in section 10.7., in the German questionnaire about 450 questions were asked in order to give the information needed on ISCED-level, on the one hand, and, on the other hand, to better describe the German situation. Plausibility checks resulted mainly from open-ended questions.

No model calculated.

No deviation.

About 9 months from the last day of the reference period to the release of the first report, trend report ("Trendbericht").

About 14 months from the last day of the reference period to the release of the big report with first deeper analysis. As the data is used by the academic community in Germany it is not possible to name the final date.

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

Not applicable.

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

No additional variables related to COVID-19 were collected.

Not applicable.

The sample source and the survey design were changed in the German AES. In the AES 2016 it was a random route sampling method, in the AES 2022 it is a register-based sample.

In 2022, in addition to the CAPI mode, a CAWI mode was also implemented. Initially, all respondents received an invitation to the CAWI and only those who do not participate are sent to the CAPI field. The changes were introduced via an experiment in the national AES 2020 in a controlled way. Then, the former and the new design were tested and compared. The results showed no break.

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

Not applicable.

See table 15.3 “Coherence - cross-domain” in annex “DE - QR tables 2022 AES (excel)”.

Not applicable.

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.

Staff involved in administering and analysing the survey all in all about 3 FTEs for 2 years.

The total of 9,818 interviews is divided into 6,842 CAWI interviews and 2,976 CAPI interviews.

About 187 interviewers realized the 2,976 CAPI interviews.

Including all questions, national and EU mandatory: 36 min.

We are not able to give an estimate time on a measured basis for EU questionnaire. Due to our experience we'd estimate the questionnaire would have about 20 minutes of average interview duration.

Not applicable.

Not applicable.

Not applicable.

Stratified random sample, register-based sample of target persons.

Two-stage sampling:

• First stage: Selection of PSUs
• Second stage: Selection of target persons as respondent (FSUs)

For details see Annex "AES Germany: Methods (Extracted parts of the German report)" in section 10.7.

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

Every 6 years.

CAWI first, after second reminder CAPI.

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

In computer assisted web and personal interviews (CAWI, CAPI), posterior data cleaning and editing is hardly necessary. Various data checks carried out with the CAWI and CAPI script before the launch of fieldwork and regular data checks during fieldwork, with a programmed syntax, ensured that filtering mistakes were minimized. Some data checks, especially plausibility checks, had to be done manually. Furthermore open answers had to be inspected (e.g.: “other, please name”) and then be classified to values of closed-ended questions if possible.

The ex-post coding of the open answers regarding the International Standard Classification of Education 2011 (ISCED 2011), the Classification of Occupations 2008 (ISCO 08), the Classification of economic activities Rev. 2 (NACE Rev. 2) provided by respondents was done by a small team of professionals specialized in the coding. This method of coding was chosen in order to ensure a very high quality of the coding, with a minimization of coding mistakes and a very high degree of harmonization and coherence. 

For further information see chapter A1.5 in Annex "AES Germany: Methods (Extracted parts of the German report)" in section 10.7.

See GESIS short version of the Fieldwork Report (will be published together with the data at the GESIS website).

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

Not applicable.

Not applicable.

None.