European Commission - Eurostat

F4: Income and living conditions; Quality of life

European Commission - Eurostat

Unit ESTAT.F.4: Income and living conditions; Quality of life

L-2920 Luxembourg

Crime regional data are police-recorded offences. The data are collected at regional level for European Union Member States, EFTA countries, and potential EU members where NUTS3 is relevant and available. National data are presented in the dataset for countries where NUTS3 is not relevant (Cyprus, Luxembourg, Liechtenstein, Montenegro as the whole territory is one single NUTS region) or still not available (Bosnia and Herzegovina and Kosovo - this designation is without prejudice to positions on status, and is in line with UNSCR 1244/1999 and the ICJ Opinion on the Kosovo declaration of independence). 

From 2008 onwards, the statistics include police-recorded offences for homicide, assault, sexual violence, robbery, burglary, (of which) burglary of residential premises, theft, (of which) theft of motorized land vehicle.

Crimes are classified by the International Classification of Crime for Statistical Purposes (ICCS).

Regions are classified by Nomenclature of territorial units for statistics at level 3 (NUTS3).

Not applicable.

Each country is asked to provide details of any divergence from standard definition.

This information is found for each offence in the Crime metadata file - items 3.4 and 15.1

The statistical units is the police-recorded offence, means reported criminal act.

For administrative data, the statistical population for each statistic is the complete register (all the relevant records). 

European Union Member States, EFTA countries Iceland, Liechtenstein, Norway, Switzerland, EU Candidate countries Bosnia and Herzegovina, Montenegro, North Macedonia, Albania, Turkey, Serbia, and EU potential candidate Kosovo.

More details on geographical coverage of police data are provided in Crime metadata item 3.7.

The tables are available from 2008 onwards.

Not applicable

• Absolute number.
• Rate by 100.000 population size.

 Population size defined as resident population on January 1^st according to Eurostat database population figures.

Due to the changes in NUTS2016 and NUTS2021, crime and population data are not available in the same NUTS version for some years for a few countries. For example, crimes were provided with the later NUTS version for previous years, while population data were available with the previous NUTS version. Therefore, in these cases (some regions of BE, DE, NL, PL, NO) the ratio per hundred thousand inhabitants is not available. The ratio is also missing due to the unavailability of data population according to NUTS classification for previous years for Albania (population not available in NUTS3 for years 2008-2011), Serbia (population not available in NUTS3 for years 2008-2016) and Bosnia and Herzegovina.

All regional rates until 2021 were calculated from population data available in July 2023. The 2022 regional crime rates and any revisions of previous years were calculated from the population data available in February 2024.

The standard reference period for crime police-recorded offences is the calendar year.

The 2004 Hague Programme November: Strengthening freedom, justice and security in the European Union.

The Hague Programme, OJ C 53 of 3.3.2005, p.11 at

http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52005XG0303(01)

The 2009 Stockholm Programme: An open and secure Europe serving and protecting citizens.

http://eur-lex.europa.eu/JOHtml.do?uri=OJ:C:2010:115:SOM:EN:HTML

Communication from the  Commission to the European Parliament and the Council. Measuring Crime in the EU: Statistics Action Plan 2011- 2015

EUR-Lex - 52011DC0713 - EN - EUR-Lex (europa.eu)

 Data are shared with the Directorate for the EU regional and urban development - European Commission (europa.eu)

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.

Not applicable.

Data are released within 16 months after the reference year.

Release calendar - Eurostat (europa.eu)

In line with the Community legal framework and the European Statistics Code of Practice Eurostat disseminates European statistics on Eurostat's website (see item 10 - 'Accessibility and clarity') respecting professional independence and in an objective, professional and transparent manner in which all users are treated equitably. The detailed arrangements are governed by the Eurostat protocol on impartial access to Eurostat data for users.

Annual.

Ad hoc news releases 1-2 times per year on crime national data.

Crime and Criminal Justice webpage - Overview - Crime and criminal justice - Eurostat (europa.eu)

Crime and Criminal Justice - Statistics Explained on national data published in April-May every year.

Publication Crime section - Publications - Crime and criminal justice - Eurostat (europa.eu)

Tables are available on-line in Eurostat webpage Crime Database  and are updated annually. 

Not applicable.

Not applicable

Each country determines the methodology and is responsible for its documentation. Data are produced by official authorities such as the police, prosecution or the ministry of the interior. Each country determine which authority or statistical office that should collect, check, and send data and documentation.

•  More details on methodological rules are available in Crime metadata item 10.6 on

– Stage of data collection of police data for number of offences
– Counting unit of police data
– Counting rules for offences (use of principal offence rule, counting rule for multiple/serial offences of the same type, counting rule for when an offence is committed by more than one person)

• Crime and Criminal Justice Statistics - Methodological guide for users This methodological guide presents how EU Statistics on crime and criminal justice are collected, which types of data are included, how they are classified in an international context and which rules should be applied for counting offences. 
• EU guidelines for the International Classification of Crime for Statistical Purposes — 2017 edition  This booklet presents the structure of the ICCS, its classification principles and its relations to existing classifications, to facilitate the implementation of the ICCS. It provides a brief overview of concrete organizational and technical tasks for a successful implementation of the ICCS at the national level, which should engage all crime data users and data providers. In addition, an implementation strategy through the creation of a correspondence table, a tabulation of all offence categories in the ICCS linked to all offences on the national level, is briefly described.
• Guidelines for the production of statistical data by the police This guidelines by the United Nations Office on Drugs and Crime (UNODC) for the production of statistical data are aimed at providing advice to police departments on the collection, production and dissemination of high-quality statistical data that can assist them in performing and monitoring their core functions, improving the measurement of access to justice and promoting the implementation of ICCS.

Not applicable.

Data are provided to Eurostat by a contact designated by each country. Notably, this contact can be another authority than the national statistical office. Each country determine its own procedures for gathering data (see point 10.6.).

No assessment has been made.

The need to provide information on the development of crime in the European Union was recognised in the Hague Programme adopted by the European Council in 2004.

Some EU institutional users communicate statistical needs to Eurostat on a regular basis.

As crime have impact on urban and regional economic and social development and has significant variation within countries, the General Directorate for Regional and Urban Policy was interested in crime regional data in order to increase effective measures against crime, including improving capacity of police force, through the Cohesion Policy funding.

No user satisfaction survey has been done.

Data is missing for some countries and some crimes, for some or all years.

As regards the 36 countries for which NUTS3 is available, only one country (MK) have never provided regional data. All the requested NUTS3 regions were provided by all the 31 relevant countries except for EL and IT that provided data at NUTS2 level for one region (EL30 and ITG2), and NO that provided the NUTS2 level for one region (NO06) in 2008-2019 and since 2020 didn’t provide data for one region (NO0B1). For countries with only one NUTS3 (CY, LU, LI, ME), the national value, if available, is disseminated.

The NUTS classification is still not available for BA and XK. However the national value, if available, is disseminated in the dataset.

Some regional data are not disseminated according to the validation rules (see 18.4 Data validation).

In the table below, the number of countries for which data are disseminated in the regional crime dataset is shown by crime and year, for EU countries, EFTA countries and potential members (BA and XK included).

In the table below, the number of crimes (out of seven) which are disseminated in the regional crime dataset is shown by country and by year.

Principally, police-recorded offences indicates the registration and handling of cases by the police.

Relative to the administrative data used, the accuracy is largely unknown.

Relative to total crime occurrence, it is generally considered that administrative crime statistics underestimate the total amount, as a certain volume remained unreported.

Not applicable - data are from administrative records.

Data from administrative records are likely to contain some random errors (mistakes) as well as systematic errors (bias).

The data for the reference year Y are requested to be provided by the 15th of September of year Y+1. After the validation checks have been applied to the data, the countries are consulted again to revise any inconsistencies or to provide additional reference metadata. Data are disseminated within 16 months from the reference year.

No delay in 2024.  

Comparability of crime data among countries is affected by many factors, such as differences in:

• Legal system and criminal justice system
• Legislation, criminal law and legal definitions
• Organisation and efficiency of police
• Recording system, reporting system, and production system
• Relation between national crime definition and international statistical definition
• Statistical unit and statistical population
• Statistical definition, reference time, counting methods and calculation methods
• Reporting rate for the crime

The comparison can be made between regions in the same country as the data is not affected or less affected by these factors.

See the list of factors in 15.1. Changes in any of the factors can affect the comparability over time.

Comparisons can be based upon trends as the characteristics of the recording system of the countries remain fairly constant, in case of a break in a series is indicated with a ‘b’ flag.

Not examined.

The internal coherence of the data is checked with basic validation rules (see section 18.4. Data validation). Moreover, sum of regional data is checked with the national data.

However, sum of regional data may differ from national figures due to differences in the data source or in the timing of the data collection.

Not estimated.

To further specify the general Eurostat revision policy, the following revision policy has been established for the crime domain. Revisions to the data are possible at any time, but only based on new data received from national contact.

The revision practice effectively corresponds to the revision practice of the domain listed under sub‑concept 17.1 (data revision – policy). All reported errors (once validated) result in corrections of the disseminated data. Reported errors are corrected in the disseminated data as soon as the correct data have been validated. Data are only published once they are deemed to be sufficiently complete for all data providers. Whenever new data are provided and validated, the already disseminated data are updated. Aggregates and components are revised at the same time.

Data can be produced by police, prosecution or the ministry of the interior.

The source data type is record in administrative registers.

Annual.

In each country, figures are collected in a single file from different territorial units or from a centralised data source, depending on the statistical organisation of the country. The files are sent to Eurostat by a common procedure, the electronic Data file Administration and Management Information System (EDAMIS).

The data are checked for completeness, internal consistency, and consistency over time and coherence with other relevant data sources. For internal consistency the following are checked at national level:

- Large revisions (new figures for earlier years)

- Large variation between years

- Large change from previous to current reference year

- Inconsistencies between data and metadata

Moreover, at regional level:

-the data should cover all regions in the whole territory of criminal law jurisdiction, or the same total area as the reported national crime figure (UN-CTS). Thus, countries are expected to provide data for all regions at NUTS 3 level.

-the consistency between regional and national figures is checked: sum of regional crimes should be equal to the national figure reported in the UN-CTS. 

The 2008-2017 data were published only if the difference between the sum of regional and the national data was less than ± 20 %, while since 2018 data, countries are invited to resolve the outstanding amounts by using the residual category for the region (unknown region or outside the country) or by providing explanations.

Until 2022 data collection (2021 data) Assault (level 3 - ICCS02011 including ‘Serious assault’ and ‘Minor assault’) was collected in regional data while Serious Assault (level 4 - ICCS020111) was collected in the UN-Crime Trends Survey; in ths case, the validation rule required that the sum of regional data was not lower than the national data.  Theft of a motorized land vehicle (level 4 -ICCS050211) was collected in regional data, while at national level the ICCS code ICCS05021 ‘Theft of a motorized vehicle or parts thereof’ (‘including ‘Theft of a motorized land vehicle’, ‘Illegal use of a motorized land vehicle’, ‘Theft of parts of a motorized land vehicle’ and ‘Other theft of a motorized vehicle or parts thereof’) was collected; in this case, the sum of regional data should not be higher than national data. Since 2023 data collection (2022 data) Assault and Theft of a motorized land vehicle have also been collected at national level and the full consistency have been checked in the validation process.

Any issue revealed by the checking is pointed out to the national contact, by email. The national contact may either confirm the validity of data or send new data. No editing, estimation, imputation, weighting, seasonal adjustment, or other statistical modifications are carried out by Eurostat.

Comments on NUTS3 classification:

EL: Regions EL301 to E307 were not provided, the figures were provided at NUTS2 level E30, as data are kept according to the territory of police agencies.

IT: With regard to Sardinia region, the available data refers to the provinces of Cagliari, Nuoro, Oristano and Sassari, on the borders before the institution, with regional law 12 Jul 2001 n. 9, of the provinces of Olbia-Tempio, Ogliastra, Carbonia-Iglesias and Medio Campidano, subsequently suppressed (available since NUTS2006). For these, and for the new province of Sud Sardegna established by regional law on 4 February 2016 (available since NUTS2021), the new Prefectures have never been activated (Prefectures have the responsibility of validating the data within their own areas). The sum of the available crime data of the four provinces of Cagliari, Nuoro, Oristano and Sassari, on the pre-2001 borders, therefore corresponds to the total of the Sardinia region. Thus, crime data can be provided only at NUTS2 level.

NO:With the regional and municipal reform, Norway went on 1 January 2020 from 18 to 11 counties, and from 422 to 356 municipalities. The time series before and after 2020 therefore contains breaks due to major changes in the division of counties and municipalities. The counties with the same name will have time series that extend before and after 2020, although these time series may also contain breaks due to changes in the boundaries between the counties. Data on NO0B1 Jan Mayen cannot be provided as Jan Mayen is not a county or municipal in the Norwegian crime statistics, while data on NO0B2 Svalbard due to small numbers are included in the residual code NOZZZ Extra-Regio NUTS3. NOZZZ Extra-Regio NUTS 3 includes the following: offences outside of Norway, unspecified categories and Svalbard. The sum of the regional data should then match the national value.

Comments on inconsistency with national data or missing data for crimes:

Intentional Homicide

CZ: 2018 regional data also includes attempted intentional homicide.

HU: ICCS categories have only been corresponded from 2017 to crime categories of the Hungarian Criminal Code, therefore we cannot provide ICCS-based data for earlier years.

NL:  In the Police registration ICCS0101 (Homicide) includes attempts of homicide so the total number of offences is much higher than the actual number of homicides in The Netherlands. Therefore we do not use police registration for determining the number of homicides. Data not provided.

PL: Data for ICCS0101 is not available with breakdown by NUTS3

LV: In 2019 regional data collection (2018 data) the method for compiling crime data, was different than for UN-CTS. As it was suggested by Eurostat we used already publicly available data as data collection covered time series from 2008 it would be to big of a burden to involve data provider on crime data (number of offenses). Thus, some double counting might be present. 

Assault

So far, 5 countries (BE, BG, CZ, DE, and SK) provided the same data or very small differences for ‘Assault’ in regional data and ‘Serious assault’ in national data.

BE: ICCS02011 cannot be splitted in two (minor vs. serious) because it's not always possible to determine if the aggression is minor or major : structured data do not always allow to make the distinction / we have the info about certain types of injuries (e.g. genital mutilation) but we don't always have information on the severity of injuries.

CZ: Serious Assault (ICCS020111) and Assault (ICCS02011) are the same values, because TSK (tactical-statistical classification of the Police of the Czech Republic) includes only Intentional bodily harm (TSK 151).

DE: Until 2019 regional data are provided for serious assault.

EL: Data for this crime are available at regional level since 2021.

HU: ICCS categories have only been corresponded from 2017 to crime categories of the Hungarian Criminal Code, therefore we cannot provide ICCS-based data for earlier years.

Robbery

FR: Since 2016, in regional data there are some offences that have been included in national data since 2019.

HU: ICCS categories have only been corresponded from 2017 to crime categories of the Hungarian Criminal Code, therefore we cannot provide ICCS-based data for earlier years.

PL: The differences between national and regional 2018-2021 data are due to the fact that they are captured from two different recording systems.
The national data are obtained from the system which records data according to the Police units registering the proceedings. Data from this system can be divided by ICCS categories but cannot be divided by NUTS3.Regional data are obtained from the system which records data according to administrative division. Data from this system can be divided by NUTS3 but not all ICCS categories can be delivered by this system. 

NO: The differences between regional and national 2018-2019 data for the ICCS0401 are due to unspecified place and Svalbard being excluded from the regional data delivery.

Burglary

EE: Burglary is not distinguished in our laws (such offences are counted among thefts). Although the police have data on how crimes were committed such data are not complete and have no legal meaning.

HU: No matching category in the Hungarian Criminal Code, which means that we couldn’t provide data on Burglary (0501)

LV:  In 2019 regional data collection (2018 data) the method for compiling crime data, was different than for UN-CTS. As it was suggested by Eurostat we used already publicly available data as data collection covered time series from 2008 it would be to big of a burden to involve data provider on crime data (number of offenses). Thus, some double counting might be present. 

LT: In 2018 international crime classifications were not yet incorporated in Lithuania’s Register of Criminal Offences. Hence, we cannot provide 2018 and older data.

PL: The differences between national and regional 2018-2021 data are due to the fact that they are captured from two different recording systems.
The national data are obtained from the system which records data according to the Police units registering the proceedings. Data from this system can be divided by ICCS categories but cannot be divided by NUTS3.Regional data are obtained from the system which records data according to administrative division. Data from this system can be divided by NUTS3 but not all ICCS categories can be delivered by this system. 

Burglary of residential premises

PL: Data for ICCS05012 is not available with breakdown by NUTS3

EE: Burglary of residential premises is not distinguished in our laws (such offences are counted among thefts). Although the police have data on how crimes were committed such data are not complete and have no legal meaning.

LV: In 2019 regional data collection (2018 data) the method for compiling crime data, was different than for UN-CTS. As it was suggested by Eurostat we used already publicly available data as data collection covered time series from 2008 it would be to big of a burden to involve data provider on crime data (number of offenses). Thus, some double counting might be present. 

HU: No matching category in the Hungarian Criminal Code, which means that we couldn’t provide data on Burglary of private residential premises (05012).

IS: This data has been provided since 2022.

Theft

PL: The differences between national and regional 2018-2021 data are due to the fact that they are captured from two different recording systems. The national data are obtained from the system which records data according to the Police units registering the proceedings. Data from this system can be divided by ICCS categories but can not be divided by NUTS3.Regional data are obtained from the system which records data according to administrative division. Data from this system can be divided by NUTS3 but not all ICCS categories can be delivered by this system. 

HU: ICCS categories have only been corresponded from 2017 to crime categories of the Hungarian Criminal Code, therefore we cannot provide ICCS-based data for earlier years. Theft 0502 includes Plundering (Section 366, excluding cases when violence or threat is applied), Theft (Section 370) and Larceny of Motor Vehicle (Section 380, excluding cases when violence or threat is applied).

NO: The differences between regional and national 2018-2019 data for ICCS0502 are due to unspecified place and Svalbard being excluded from the regional data delivery.

RO: Since 2019, figures include Theft (art.228 Penal code), Aggravated theft (art.229 Penal code), Stealing for personal use - Total (art.230 Penal code). Offences solved by the police / declined to the Prosecutor's Office. For previous years the data provided included 'Burglary'. Source: General Inspectorate of Romanian Police within the Ministry of Internal Affairs.

SK: The national statistical data on thefts also include data of non-departmental units other than the police. This non-departmental units provide only overall data, do not sort it into the regions. These include offices such as: Military Police, Prison Guard, Railway Police, Customs Criminal Office and that is the reason why there are differences between regional and national statistics.

Theft of motorized land vehicle

So far, out of 27 countries that provided both 'Theft of motorized land vehicle' in regional data collection and ‘Theft of motorized vehicle or parts thereof' in national data collection, 13 countries provided the same value for ICCS05021 ‘Theft of motorized vehicle or parts thereof’ in national data collection and ICCS050211 ‘Theft of motorized land vehicle’ in regional data collection, while 14 countries (BE, BG, CZ, DK, HR, LT, LV, NL, AT, PL, RO, FI, SE, CH) were able to distinguish the two crimes and provided in the regional data collection for the period 2018-2022, or at least for some years, a slightly lower number according to the more specific code.

BE: there is difference between the 2 datasets. In UN-CTS (disseminated in crime_off_cat), the figures for the more general code 05021 are provided and in the regional crime data, the more specific figures for the code 050211. This explains why there is a smaller number in the regional crime data.

CZ: we reported ‘Theft of a motorized vehicle’ data in the regional data collection and ‘Theft of a motorized vehicle or parts thereof’ in UN-CTS disseminated in (crime_off_cat).

EE: 'Theft of a motorized land vehicle' (050211) corresponds to our data and data has been provided, but we are not able to provide data for 'Theft of a motorized vehicle or parts thereof' (05021) as a whole (i.e, we do not have separate data on theft of parts of a motor vehicle - they are covered by 0502 'Theft').

EL: Theft of a motorized land vehicle is not separate from Theft of a motorized vehicle or parts thereof.

ES: The type of theft of vehicle parts is not available. Thus, the data provided in indicator ICCS 050211 corresponds to those indicated in indicator ICCS 05021.

LV: In 2019 regional data collection (2018 data) the method for compiling crime data, was different than for UN-CTS. As it was suggested by Eurostat we used already publicly available data as data collection covered time series from 2008 it would be to big of a burden to involve data provider on crime data (number of offenses). Thus, some double counting might be present. 

LT: In UN-CTS data collection (disseminated in crime_off_cat), the crime is provided by the code ICCS05021 (Theft of a motorized vehicle or parts thereof).In regional data collection the crime is provided by the code ICCS050211 (Theft of a motorized vehicle).Since we have filled out the data according to the codes provided in questionnaires the numbers in UN-CTS and Regional data differ.

HU: No matching category in the Hungarian Criminal Code, which means that we couldn’t provide data on Theft of a motorized land vehicle (050211).

MT: For 'Theft of a motorized land vehicle' and 'Theft of a motorized vehicle or parts thereof’, it will not be possible to provide since the national category includes the theft of vehicle part within the category which includes all theft from vehicles.

NL: In the regional data collection we collected code ICCS050211 for the crime ‘Theft of a motorized vehicle’. In UN-CTS (disseminated in crime_off_cat) we used code ICCS05021.

PT: For the crime "Theft of a motorized land vehicle" in both national and regional data it is included the crime ‘Theft of a motorized land vehicle’ (level 4 -ICCS050211)". It is not possible to disaggregate data related to "parts thereof".

SI: The provided data under ICCS05021 is in great measure identical to data under ICCS050211.

SK: We cannot distinguish theft of a motorised land vehicle separated from theft of a motorised vehicle or parts thereof. We registered theft of motorised land vehicle only.

FI: in UN-CTS data collection (disseminated in crime_off_cat) we provide data for code ICCS05021, in regional data collection we provide data for code ICCS050211.

TR: The data is given for ICCS 050211 level for 2018, 2019, 2020 and 2021 year in UN-CTS.

EU totals are not published in the dataset.

No adjustments are made.

None.