Statistics Portugal (Instituto Nacional de Estatística)

Department of Demographic and Social Statistics - Labour Market Statistics Unit

Av. António José de Almeida, 5

1000-043 LISBOA

PORTUGAL

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.

All national territory was covered: Portugal (Mainland; Autonomous Regions of Azores and Madeira).

2007, 2011, 2016, 2022

Fieldwork periods:

• AES 2007: end of September 2007 to the end of December 2007
• AES 2011: 24/10/2011 - 13/02/2012
• AES 2016: 28/11/2016 - 07/03/2017
• AES 2022: 28/09/2022 - 09/02/2023

Not applicable.

Number, EUR.

Fieldwork period for AES 2022: 28/09/2022 - 09/02/2023

At European level:

Basic legal act: Regulation (EU) 2019/1700

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

At national level:

AES is registered in the National Statistics System

Not applicable.

At national level, the collection, processing and dissemination of data is carried out in accordance with the provisions of Law 22/2008 of May 13, namely Article 6, which establishes the application of the principle of statistical secrecy to data. Any statistical units, directly or indirectly identifiable, cannot be disclosed, not only due to the protection afforded by this principle, but also due to the protection of personal data that arises from the General Data Protection Regulation and other legislation on this matter.

The breach of statistical confidentiality is punishable not only disciplinary, but also criminally, as it constitutes professional secrecy, according to Article 32 of the SEN (National Statistics System) Law.

The treatment of statistical confidentiality consists in the elimination of individual identification, the variables used in the selection of the sample and those associated with the field work, in addition to the use of top / bottom coding and grouping in several variables, to minimize the risk of identification in order to obtain a file for scientific purposes.

The micro-data set of the AES 2022 is anonymized (no possible identification of individual units) by our methodological team. This data set is available for research purposes only, to registered researchers in the Ministry of Education and Science.

Concerning macro-data, for the indicators to be published, coefficients of variations will be calculated. Those proportions for which the associated CV are higher than the threshold defined internally for release purposes will be suppressed.

All statistical operations registered in the National Statistical System, methodologically certified, have an associated description of the products to be released, as well as a preannounced schedule, which contains, among others, the foreseen dates for publishing the results. Both calendars are available to the public in Statistics Portugal website.

The Statistics Portugal release calendar is available at https://www.ine.pt/xportal/xmain?xpid=INE&xpgid=ine_calendarios&xlang=en 

The following principles are included in Statistics Portugal's general policy for disclosing data to users, and will be followed by AES:

• Provide official statistical information free of charge on the Statistics Portugal website; 
• Provide objective, timely and punctual official statistical information, accompanied by the respective statistical metadata and, possibly, other information that facilitates its interpretation;
• The official statistical information should be impartial and disseminated simultaneously to all users;
• Provide official statistical information on pre-announced calendar established based on exclusively technical-regulatory criteria and taking into account the quality/up-to-date commitment;
• Disclose changes as far in advance as possible to the dissemination calendar and their justification, keeping the initial calendar accessible.

Every 6 years.

A Press Release, both in Portuguese and in English, was published on 17 October 2023, presenting and analysing the main results of the AES 2022 carried out by Statistics Portugal. This press release was accompanied by the availability of the main indicators at Statistics Portugal website.

Further publications to be released within the scope of the AES 2022 are under analysis.

The aggregated results of the survey will be available in a Database at Statistics Portugal website, under the theme "Education, Training and Learning".

Not applicable.

According to Article 6 of the National Statistical System Law (Law 22/2008 of 13 May), individual statistical data on natural and legal persons may only be provided for scientific purposes in an anonymized format, to accredited researchers, under a specific agreement between Statistics Portugal and the Ministry of Science and Technology.

The accreditation of researchers is done by the General Directorate of Education and Science Statistics.

Further publications to be released within the scope of the AES 2022 are under analysis.

Not applicable.

The main methodological aspects of the AES 2022 survey are described in a methodological document, which is mandatory for all surveys carried out in the context of the National Statistical System.

All methodological documents are freely accessible online, at Statistics Portugal website, complying with the principles defined in the “European Statistics Code of Practice”.

Not applicable.

The European Statistics Code of Practice, a self-regulatory instrument whose main purpose is to improve trust and confidence in official statistics produced and disseminated by the statistical authorities of Member States, Candidate Countries, EFTA Members and Eurostat, reinforcing their independence, integrity and responsibility and to enhance the quality of European Statistics. (https://www.ine.pt/xportal/xmain?xpid=INE&xpgid=ine_inst_codconduta&xlang=en)

Information Security Management System (ISMS), aligned with the best international practices, namely NP ISO / IEC 27001: 2013, is comprised of a set of policies and procedures that are now available to all Statistics Portugal’s procedures, and which allow the operationalization of the System. (https://www.ine.pt/xportal/xmain?xpid=INE&xpgid=ine_inst_sgsinformacao&xlang=en)

The following are noteworthy strategic documents to Statistics Portugal and made available in its Portal:

• The 2019 edition of the Quality Chart (only in PT version), which formalizes Statistics Portugal’s assumed public commitment in relation to the quality and credibility of the official statistics it produces and disseminates, to the public service it provides to the society – making it clear to all information providers, users of statistical information and to the public in general – also expressing the commitment towards information security;

• The Information Security Policy (only in PT version), which sets the general principles by which Statistics Portugal carries its mission, to the assets it manages within the scope of the ISMS, following all requirements within NP ISO/IEC 27001:2013, the applicable legislation, regulation and recommendations of the ESS (European Statistical System) and EUROSTAT in what specifically concerns information security;

• The Statistical Confidentiality Policy (only in PT version), which replaces the former Statistics Portugal's Confidentiality Chart and is part of the ISMS and formalizes the public commitment of compliance with the Principle of Statistical Secrecy assumed by Statistics Portugal as the central body responsible for the coordination and development of the national statistical activity;

• The Personal Data and Privacy Protection Policy (only in PT version), which aims to supply the providers of data information about the nature of the collected data, its intended purpose and how the data are treated.

The Dissemination policy of Statistics Portugal lays down the fundamental principles governing the dissemination of official statistics, directly or indirectly produced under its responsibility. It should have as main reference the applicable principles of the National Statistical System: technical independence, statistical confidentiality, quality and accessibility. (https://www.ine.pt/xportal/xmain?xpid=INE&xpgid=ine_inst_pdifusao&xlang=en)

To ensure process and data quality several procedures are carried out:

• The questionnaire and the methodological document follow a procedure of approval within the National Statistical System. This procedure is supervised by a coordination team and implies the consultation of the questionnaire and methodological document by all departments, where opinions and recommendations are collected and adopted, if relevant.
• Data collection is carried out by interviewers (telephone interviews) or via a web platform, using specific computer programs backed up with validation rules (for guarantying the quality of responses, coherence between related questions and the observance of the routing), instructions for the interviewers/respondents and with explanations about the technologies which are eventually more difficult to be understood by respondents.
• Interviewers follow a training action before the beginning of data collection, backed up with support documentation, information about the necessary procedures to be taken and rules about what should be considered in each question, always following Eurostat recommendations.
• During the data collection, the interviewers work, and the response rate are periodically supervised, the latter according to previously defined goals.
• Before sending the database to be grossed-up by the Department of Methodology and Information Systems, the data is submitted to validation procedures both by the Department of Data Collection and Management and the Department of Demographic and Social Statistics, where therefore, confirmations with team-work, (re)codifications and corrections can be made to the data collected.

• Initially, upon registration, the data was automatically validated through a set of validation processes previously included in the registration computer application.

• A critical analysis of the global coherence of the recorded information was carried out, through the use of Business Intelligence Software (BIS) on the copy of data from the operational environment. For the questionnaires to be considered finalized, all inconsistencies of the "Error" type must be resolved.

• In a second phase of quality control, data analysis was carried out in Business Objects (BO), to check, for example, the questionnaire paths and validation rules, and possible inconsistencies in relation to expected values. 

• Finally, the data was processed and analysed using SPSS software. After concluding the fieldwork, a validation of the national database was undertaken, consisting of sample control and a set of computer procedures aiming at identifying registration gaps, coding problems and possible inconsistencies in the values. 

The main users of the AES are:

Internal users (from the National Statistical System):

• Statistics Portugal 
• Regional Directorates of Statistics of Região Autónoma dos Açores (SREA) and Região Autónoma da Madeira (DREM) that act as delegations of Statistics Portugal for regional data collection purposes;
• Directorate-General for Education and Science Statistics

Other national users: 

• Ministry of Education; Researchers; Press

 Other European/International users: 

• Eurostat
• OECD
• UNESCO

Statistics Portugal Dissemination Policy foresees a regular monitoring and assessment of the quality and accessibility of disseminated data, in order to ensure continued improvement of dissemination activity.

This process involves:

• The different types of users are regularly surveyed on their satisfaction regarding products and services made available by Statistics Portugal;
• User comments, suggestions and complaints are viewed as real opportunities to further improve the dissemination activity of Statistics Portugal;
• User comments, suggestions and complaints are duly treated in accordance with the procedures established in Statistics Portugal's Quality Charter;
• User comments, suggestions and complaints, where relevant, lead to the adoption of measures to further improve the activity of Statistics Portugal, particularly in the dissemination function.

All mandatory variables were included in the microdata file sent to Eurostat.

Optional variables on COVID-19 impact were collected voluntarily at national level.

Not applicable.

Random errors are essentially caused by the respondents and by the records of the interviewers. As for systematic errors, they are caused by the non-responses of some households/individuals, which are compensated in the final weights which integrates a factor of non-responses and calibration.

• First step: determination of an initial weight, based on the Horvitz-Thompson estimator, in which the weights of each unit are given by the inverse of the respective selection probability.
• Second step: application, to the weight calculated in the 1st step, of a correction factor for non-responses, to compensate for the effect caused by these on the sample size.
• Third step: correction of the weights determined in the previous phase by applying the margin adjustment method, so that the distribution of employees weighted by the values of the variables considered in the adjustment is identical to the structure of the corresponding universe.

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

A non-sampling error is an error in survey estimates which cannot be attributed to sampling fluctuations. Such errors can either be coverage errors, measurement errors, non-response errors, processing errors or model assumption errors.

For information on the sampling frame see section 18.1.

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

Not applicable.

At Statistics Portugal, data collection is done by external interviewers. The interviewer’s work is carried out in a self-employment basis in the data collection activity. The training of the interviewers and supervision staff was undertaken, including training sessions. The interviewers participated in a training session before the start of data collection, backed up with support documentation, information about the necessary procedures to be taken, and rules about what should be considered in each question, also following Eurostat's recommendations. An interviewer's manual, distributed before the training sessions, and other training and data collection materials, including a FAQ's document, were prepared. The interviewer’s manual covers the following points: detailed information and guidelines about all subjects covered in the training and the questionnaire; information about the necessary procedures to be taken and rules about what should be considered in each question; explanatory notes concerning objectives, procedures, main concepts, flows, and the answering criteria of the questions, always following Eurostat's recommendations. Once the fieldwork began, and after training, interviewers had a few days to familiarise themselves with the questionnaire and the data collection computer application before starting work.

See the following sections.

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

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

As the fieldwork developed, the management and results of the interviews, as well as their contents, were monitored by the supervisors and the team responsible for the project. Furthermore, there was always a very close monitoring with the interviewers, on a daily basis, by the supervisors.

Several measures were implemented, considering the problems faced throughout the fieldwork:

• Clarification of the respondents' doubts by internal staff, using telephone and e-mail.
• Addresses: registration and thorough analysis of proposed changes of address; checking whether the dwellings with an advance official letter returned for address reasons were the subject of a proposal to change the address.
• Analysis of contacts made with dwellings classified as "inaccessible" or as "absent" to confirm if the interviewer's effort was being adequate (distribution over the days of the week and contact times) and provide internal support to change this status.
• Official letters: contacts with households to reverse unsuccessful interviews; checking if, after sending the 2nd copy/reminder/refusal letter, the dwelling was contacted again, as well as analyses of the situations in which the interview with a selected respondent was not carried out/concluded.
• Checking if the notes and observations recorded were consistent with the data collected and with the contacts and appointments made and whether the registered notes would require internal intervention (e.g., sending a 2nd copy of the official letter, sending a reminder or a refusal letter, contacting the household, change of address, etc.).
• Checking if dwellings were contacted according to what was scheduled and if the contacts were evenly distributed between working days and weekends and diversified as well (if dwellings were contacted outside working hours on working days, for the contact to be successful) and identification of the reasons for the absence of contact with dwellings that should have already been contacted.
• Analysis of refusals, identifying which dwellings interviewers should invest more effort to reverse refusals, verifying refusal incidence by interviewer and contacting dwellings to reverse refusals and reschedule interviews.
• Reallocation of sample dwellings to other interviewers whenever justified; transition of interview mode when requested by the household and as a management measure to promote an increase in the response rate.
• Reallocation of sample dwellings and interview mode transitions to increase response rates; in mode transitions, was checking if dwellings have a telephone number (for CATI) or a valid e-mail address (for CAWI).
• Monitoring interviews for the detection of systematic errors and the respective alert/correction with the interviewers; promote internal support interventions and/or rectifications of the interviewer's procedures to achieve the expected results.
• Checking the existence of interviewers without daily exports and proceed with the respective alert.

Not applicable.

Data dissemination at national level took place in October 2023.

8 months

Not applicable.

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

Not applicable.

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

Some additional variables/information related to COVID-19 were collected, see table 15.1 "Comparability - geographical" in annex "PT - QR tables 2022 AES (excel)".

Not applicable.

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

Not applicable.

See table 15.3 “Coherence - cross-domain” in annex “PT - 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.

The average duration of interview was 41 minutes in CATI and 44 minutes in CAWI. Average duration of interview is quite similar between modes; still, the highest average interview duration was observed in CAWI mode.

Not applicable.

Not applicable.

Not applicable.

The Portuguese AES 2022 survey is a stand-alone survey, registered in the National Statistics System.

The sampling frame was selected from the National Dwellings Register, which in turn uses information collected in the Census. The geographical coverage refers to the total territory. It is constituted by private dwellings of usual residence and excludes collective households and institutions (approximately 1% of total population). Its size is approximately 1.4 million dwellings of usual residence.

The sampling frame follows a stratified and multistage sampling scheme where the primary units (PSU – Primary Sample Unit) are constituted by aggregation of 1 km² grid cells (to contain at least 150 main residence dwellings) and selected with a probability proportional to the size of the number of main residence dwellings.

Sample units are dwellings. Up to two individuals per dwelling were interviewed, aged between 18 and 69, and selection was made by using the last birthday method (last person in the household to have celebrated their birthday).

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

Every 6 years.

Regarding data collection preparation and implementation, Eurostat's guidelines were followed.

Two data collection modes (CATI, CAWI) were made available, starting with CAWI and continuing, for those dwelling units not responding through this mode, with CATI. The assumption is that a mixed-mode approach allows for a higher number of interviews to be obtained, as it is possible to switch modes according to the respondent's preference. At any time during the data collection period, it is possible to answer the survey online, or in other preferred interview mode, as the transition between modes was available.

In addition to the advance letter, as the fieldwork was being carried out, reminders were sent to those dwelling units that had not yet responded, reinforcing the call to answer to the survey.

For those dwellings for which an email address of the representative was available, informative and reminder emails were also sent. These emails were sent to different segments of the population and with different texts, depending on their situation (for those who had not responded to the survey at all, as well as for those who had started to respond to the survey, but had not yet completed it).

Several weeks after the survey was made available on the web, part of the sample that did not respond online was transferred to CATI. However, whenever they wished, respondents could go back to completing the survey online.

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

Data collection was carried out by interviewers (telephone) or via a web platform, using a specific computer programme backed up with validation rules (for guarantying answers quality, coherence between related questions and the observance of the routes), instructions for the interviewers/respondents and with explanations on questions which are eventually more difficult to be understood by respondents.

Before sending the database to be grossed-up by the Department of Methodology and Information Systems, the data have been submitted to validation procedures both by the Department of Data Collection and Management and the Department of Demographic and Social Statistics, where confirmations with teamwork, (re)codifications and corrections could, therefore, be made to the data collected.

Additionally, several coherence validation checks were performed. Necessary corrections have been made accordingly, to ensure full compliance with the rules applied to data for all Member States. 

No imputations were made for missing data.

The annual estimates for resident population as at 31/12/2022 were used, according to the 2021 Census final results, which correspond to the reference period of this statistical operation.

None. See also table 18.5.1 “Imputation - rate” in annex “PT - QR tables 2022 AES (excel)”.

Not applicable.

Not applicable.

None.