EPSO AST/158/26 Data & Statistics: 423 Places Explained

The Largest Field of the New AST Data Competition

EPSO published the notice of competition EPSO/AST/158/26 in the Official Journal on 6 October 2026 (C/2026/4963). It recruits assistants at grade AST 3 in three fields, and field 1 — Data processing, analytics and statistics — is the largest, with 423 successful candidates sought. The other two are librarianship and information science (243) and records and archives management (298). You may apply for one field only.

Applications close on 11 November 2026 at 12.00 (midday), Brussels time, via the EPSO competition page. By that deadline you must also upload a scan of a valid identity card or passport; every other supporting document is due by 11 March 2027 at 12.00.

Do not confuse it with EPSO/AD/426/25, the AD 7 data management and knowledge competition that closed in December 2025. That one recruited administrators to design data governance. This one recruits assistants to collect, clean, validate, analyse and publish data — the production line of EU statistics and analytics.

Who Can Apply

On top of the general conditions — EU citizenship with full rights, military obligations fulfilled, character requirements — you need one of two combinations on the closing date:

1. Post-secondary studies of at least two years in a listed area, plus three years of relevant professional experience after the diploma. A diploma in a listed area gained after a bachelor's in any field — postgraduate certificate, master's, PhD — also qualifies, whatever its length.
2. Secondary education giving access to post-secondary studies, plus six years of relevant professional experience.

The listed areas are broad:

• mathematics, statistics, actuarial science or data science;
• computer science and informatics, including AI;
• engineering, any field;
• economics, applied economics, business administration or finance;
• geography including geographical information systems, or geoinformatics;
• physics, chemistry or biology;
• social sciences, if the studies included quantitative methods, survey methodology, computational social science or data analysis;
• linguistics, if the studies included corpus linguistics, computational linguistics or text/data analysis.

Field 1 is also the only field with a list of accepted experience areas beyond Annex II — thirteen of them, from programming and database management to data governance, data journalism, dashboards and GIS tools. The general rules pro-rate each job by its share of relevant tasks: above 75% it counts fully, over 50% to 75% at 75%, 25% to 50% at 50%, below 25% not at all.

The Tests — and the Only One That Ranks You

Scoring runs in order: reasoning first; the field MCQ only for those who passed reasoning; then a ranking by MCQ score. EUFTE scripts are marked, and eligibility checked, only for the top of that ranking — in principle no more than 1.5 times the places, about 634 candidates for field 1, more only if ties at the cut-off require it.

Two things to plan for. The ranking test is in Language 2, where the notice asks only for B2. And it allows 80 seconds per question. A numerate candidate will be tempted to work answers out; at that speed, the definitions in the rulebook have to be recognised, not reconstructed.

See our guides to the field-related MCQ and the EUFTE. The numerical reasoning gate should be comfortable for this profile, but it shares a combined pass mark with abstract reasoning, and verbal reasoning has a gate of its own.

What the Field MCQ Draws On

Annex II lists 17 typical duties. They fall into three blocks, and each block points to a different kind of preparation. The mapping below is our guidance, not text from the notice.

Producing and validating statistics (duties 1–6, 10, 11)

Collecting data from administrative, survey, geospatial and web sources; exchanging data and metadata with Member States; cleaning, validating, producing tables, indicators and time series; reporting on quality. Inside the EU institutions that work is governed by Regulation (EC) No 223/2009 on European statistics.

The statistical principles (Article 2(1)) are six: professional independence, impartiality, objectivity, reliability, statistical confidentiality and cost effectiveness. Statistical confidentiality means the protection of confidential data on single statistical units, implying the prohibition of their use for non-statistical purposes and of their unlawful disclosure.

The quality criteria (Article 12(1)) are where questions are easiest to set and easiest to miss:

• relevance — the degree to which statistics meet current and potential needs of users;
• accuracy — the closeness of estimates to the unknown true values;
• timeliness — the period between the availability of the information and the event or phenomenon it describes;
• punctuality — the delay between the date of release and the target date by which the data should have been delivered;
• accessibility and clarity — the conditions and modalities by which users can obtain, use and interpret data;
• comparability — the measurement of the impact of differences in concepts, tools and procedures across areas, domains or time;
• coherence — the adequacy of the data to be reliably combined in different ways and for various uses.

The European Statistics Code of Practice (2017 edition) rests on 16 principles covering the institutional environment, statistical processes and statistical output — numbered 1 to 15, because coordination and cooperation was inserted as Principle 1bis. The output principles mirror the criteria above: 11 Relevance, 12 Accuracy and Reliability, 13 Timeliness and Punctuality, 14 Coherence and Comparability, 15 Accessibility and Clarity. The ESS Quality Assurance Framework (version 2.0) turns each principle into methods and tools.

For data and metadata exchange (duty 2), the standard is SDMX — Statistical Data and Metadata eXchange — whose 3.0 technical specifications include VTL, the Validation and Transformation Language used to express validation rules. For national accounts indicators and time series, the reference is ESA 2010, Regulation (EU) No 549/2013. Our statistics specialist guide goes deeper into Regulation 223/2009.

Protection and disclosure control (duty 12)

The notice asks you to apply data protection and confidentiality/disclosure-control rules and cybersecurity safeguards. For the EU institutions the data protection text is Regulation (EU) 2018/1725, not the GDPR: it governs processing of personal data by Union institutions, bodies, offices and agencies. Statistical confidentiality comes from Article 2(1)(e) of Regulation 223/2009, above. Our data protection guide covers the EUDPR in detail.

Tools, machine learning and AI (duties 7–9 and 13–17)

Exploratory and standard statistical analysis, geospatial analysis, sampling and risk indicators, visualisations and dashboards, scripts and queries in Python, R and SQL with basic testing and version control, machine-learning techniques with quality and interpretability checks, new technologies including AI, and generative AI through effective prompting. The notice names these skills but not how they will be examined. For the AI duties, which ask for implementation "in line with the relevant guidance", the Commission's own reference is its Communication C(2024) 380 of 24 January 2024, "Artificial Intelligence in the European Commission (AI@EC)".

Practical advice: split preparation time between the rulebook, which produces questions with single correct answers drawn from a text, and the tools, where Annex II tells you what to be fluent in.

Four Pairs That Look Alike

1. Timeliness against punctuality. Timeliness looks back to the event; punctuality looks at the promised date.
2. Accuracy against reliability. Accuracy is a quality criterion of Article 12; reliability is a statistical principle of Article 2 — and they are paired again in Code of Practice Principle 12.
3. Sixteen principles, fifteen numbers. Principle 1bis is the reason.
4. EUDPR against GDPR. Personal data in EU institutions fall under Regulation 2018/1725.

How to Prepare Between Now and the Tests

• Read Annex II line by line and tick off each duty against a text or a tool.
• Learn Articles 2 and 12 of Regulation 223/2009 verbatim. They are short and they generate questions.
• Prepare in your Language 2. EUR-Lex publishes Regulation 223/2009 and every other act in all official languages.
• Keep the reasoning gates safe. A fail there ends the competition before the MCQ is scored.
• Check your candidate account at least every three calendar days, and keep your EU Login email valid until the reserve list is published.

If an MCQ question looks wrong, you have 3 calendar days from the day after the test to complain through your candidate account; MCQ results cannot be reviewed.

The sister fields of the same notice are librarianship and information science and records and archives management. For how this competition fits the wider digital recruitment, see our IT specialist guide.

Sources

• Notice of competition EPSO/AST/158/26 — OJ C/2026/4963, 6.10.2026
• EPSO competition page
• Regulation (EC) No 223/2009 on European statistics
• European Statistics Code of Practice
• Regulation (EU) 2018/1725
• Regulation (EU) No 549/2013 (ESA 2010)
• SDMX

Eligibility, tests, scoring and deadlines are quoted from the notice of competition. The mapping of duties to legal texts and standards is our preparation guidance.