Offences and victims reported to the police

Offences and victims reported to the police
Topic: Social conditions, welfare and crime

Division for Income and social welfare statistics

National level, and by county, police district and municipality.

Annually. Published as soon as possible after the end of the statistical year, normally before the end of the first quarter in the consecutive year.

Some statistics are reported to international bodies such as the UN , the EU and Eurostat .

Collected and revised data are stored securely by Statistics Norway in compliance with applicable legislation on data processing.

Statistics Norway can grant access to the source data (de-identified or anonymised microdata) on which the statistics are based, for researchers and public authorities for the purposes of preparing statistical results and analyses. Access can be granted upon application and subject to conditions. Refer to the details about this at Access to data from Statistics Norway.

Raw data files from the Norwegian Police ICT Services (formerly the National Police Computing and Material Service (PDMT)) used for production of the annual statistics are stored permanently in a secured archive area. Raw data files formerly collected for half-yearly statistics were only kept until they were replaced by final annual files. Annual files used for the production of statistics on registered victims in the period 2004-2018 are also stored under this statistics area. The final processed data material (i.e. the material prepared and used in the statistics production) is stored as separate archive files and documented (in the system DataDok). The annual files that are used for the statistics production are also stored as SAS files, until they are no longer used.

The statistics on offences and victims reported to the police describes the registered level of crime from year to year. The statistics maps who is exposed to offences and who reports offences to the police, by different demographic background variables and type of offence. By mapping victims and reporters of offences, we wish to form a better basis for describing crime, and simultaneously obtain a wider understanding of the judicial system. Offences that include violations of persons are of special interest and have the best quality of data, and the statistics therefore place a particular focus on these victims and the types of offences they are subjected to.

The statistics are enabled as a result of a changeover from form reporting to electronic reporting from the police’s central criminal case register (BL/STRASAK, BL/STRASAK/PAL as from 2002). The statistics were first published in the Norwegian official statistics (NOS) for the year 1990.

Since 2004, the statistics on offences reported to the police have only been published as annual statistics. Until 2003, half-yearly statistics were also published, and quarterly statistics were published until 1998.

In 2005, the whole time series 1993 to 2002 was revised (see Control and revision under Production). This particular revision could not be carried out for 1992. Therefore, 1993 is now regarded as the first year of Statistics Norway's statistics on offences reported to the police. The principle division between crimes and misdemeanours has been discontinued in new Penal Code of 2005.

From 2019, the statistics on Offences reported to the police and Victims of offences reported to the police has been merged together as one statistic. Statistics Norway has in the past, in connection with an analysis project, published statistics on offences reported to the police (1995-1998) with information on victims (Gundersen 2000 , in Norwegian only). However, these two sets of statistics are not directly comparable due to considerable differences in the data basis, processing and definitions of statistical units. The statistics on victims 2004 are therefore regarded as new statistics from Statistics Norway. Statistics on victims of offences reported to the police have been developed and produced under commission by the Ministry of Justice and the Police. The production of statistics based on data for previous statistical years was considered, but has not been carried out for reasons of quality. As from 2009, new variables and regional statistics have been added to the statistics (see Definitions and Comparability over time and space under Background).

The statistics are now released with breakdowns of different types of offences which in principle are grouped in the same way as in the penal legislation, and are considered as a Standard of types of offences. Mainly, the standard corresponds to the divisions and terms in the current penal legislation, and classifies all offences included in the penal legislation and the special legislation (see Standard for types of offences in KLASS). From the year 2015 and onwards, the statistics is released with a new version of Standard for types of offences: Types of offences 2015, which is made as a consequence of the Penal Code of 2005 and the corresponding new codes for registering offences in all of the registers relevant to Statistics Norway’s crime statistics. The new standard replaces Standard for types of offences 2006. New breakdowns of offences are for certain tables in the StatBank made available with numbers dating back to 2003.

The principle division between crimes and misdemeanours has been discontinued in new Penal Code of 2005. Thus, from 2015 and onwards, this division is discontinued also in the statistics on offences reported to the police.

Important users are the media, the research and education sector, NGOs, the Ministry of Justice and other government agencies and other agencies in the justice sector. The statistics also serve as a basis for information to anyone else interested in the state and development within the field of crime.

No external users have access to statistics before they are released at 8 a.m. on ssb.no after at least three months’ advance notice in the release calendar. This is one of the most important principles in Statistics Norway for ensuring the equal treatment of users.

The statistics on offences and victims reported to the police are one of the four regular principal statistics in the area of Crime and justice. The rest are statistics on Offences investigated , Sanctions and Imprisonments.

When using crime statistics, it is important to be aware that the various parts do not necessarily refer to the same population of offences within the same year. It can take a long time from an offence is reported to the police to when the investigation is concluded, and until a penal sanction is given and the sentence served; in some cases, it may take years. Generally, however, the offences that are included in the statistics on offences reported to the police in the same year or subsequently are represented in the statistics on offences investigated, sanctions and imprisonments (see the statistics on criminal prosecutions). The units that are included also differ in the various statistics. The tables below offences reported to the police are therefore not directly comparable to the other crime statistics. For an overall presentation of the crime statistics, see Kriminalitet og rettsvesen 2009 (in Norwegian only).

Other closely related statistics from Statistics Norway are: Survey of level of living, victimisation and fear of crime , Statres - Police and prosecution authority , Statistics on causes of death (murder).

Related external statistics

Until 2021The National Police Directorate (POD) published Kommenterte STRASAK-tall; the police’s report statistics (the individual police districts also publish their own figures for their respective regions), and since 2022 POD and The Higher Prosecuting Authorities

publishes the report Straffesaksbehandlingen.These reports contains figures for offences reported to the police on a national level, but must be regarded as operations statistics for the police and should not be interpreted as the official crime statistics. The police figures can vary somewhat from Statistics Norway’s figures based on the timing and criteria for extraction, as well as other processing routines and the groupings of offences in the statistics (see Standard classifications under Definitions).

Kripos publishes an overview of murders, but these statistics are organised based on a different definition of the sample than in Statistics Norway’s statistics on reported offences (including year of crime and not year of registration). In a number of cases, it is recommended that users use Kripos’ murder statistics. For 2005/2006 and beyond, Statistics Norway has carried out a review and audit of all murder cases (see Collection of data revision and estimation under Production). Subsequent to these audits, the number of murders is approximately the same in the statistics from Kripos and Statistics Norway. Kripos also publishes statistics on drugs, which include figures for the confiscation of drugs. Although confiscation relates to drug offences reported to the police, it is not possible to make direct comparisons between these figures.

The above statistics and reports are available from the police's website for numbers and facts.

The Norwegian Bureau for the Investigation of Police Affairs investigates cases where employees of the police or prosecuting authority are suspected of committing criminal offences in the course of duty. Offences which are investigated and completed by the Bureau without further prosecution or court proceedings are not included in the data basis for the statistics on offences and victims reported to the police and offences investigated. However, when such cases result in a sanction registered in the Criminal Record Register, they will be included in the statistics on sanctions. Furthermore, such cases are included in the statistics on imprisonments in the same manner as other criminal cases. A closer look at the offences registered by the Bureau in 2016 show that the vast majority of cases concern different types of misconduct and misuse of public authority (which in Statistics Norway’s crime statistics are classified as part of the group of offence Public order and integrity violations). The Bureau also investigates several other types of offences, including offences of violence and traffic offences. In their annual reports, the Bureau publishes statistics on offences reported and investigated over the course of the year.

The Norwegian Institute for Alcohol and Drug Research (SIRUS) publishes statistics on various forums, which include figures from Statistics Norway’s statistics on offences reported to the police, primarily in relation to drug offences and driving under the influence (cf. “drunk driving"´). The classification of the offences can, however, differ somewhat from the statistics published by Statistics Norway.

The Norwegian Institute of Public Health has the role of data processor for the Cause of Death Registry as from 1 January 2014, and is the publisher of causes of death statistics from the statistical year 2013.

The statistics are developed, produced and disseminated pursuant to Act no. 32 of 21 June 2019 relating to official statistics and Statistics Norway (the Statistics Act).

The statistics is part of the national programme for official statistics, main topic Social conditions, Welfare and Crime, topic Crime and Justice.

Not relevant.

The statistics cover all offences that were registered as reported by the police and the prosecution authority in the course of the statistical year.

The statistics provide information on the type and number of offences, police districts and scene of crime, as well as type of victim, sex and age of victims and place of residence (see Definitions).

From 1993 to 2014 the statistics contained both crimes and misdemeanours. The principle division between crimes and misdemeanours has been discontinued in new Penal Code of 2005. The definisions of the population and the offences are, however, equal before and after this change in 2015 (see under Defintions and Sources of error and uncertainty).

The victims part of the statistics includes all with the role of "victim" in all reported offences registered in the course of the statistical year. Each registered offence normally constitutes one unit in the material. Where the report includes more than one offence, all matters that can constitute a separate point in the charge/indictment are registered. Prior to 1994/1995, several offences could be registered as one offence (see Comparability over time and space).

The number and type of reported offences in the statistics on victims should be identical to the number and type in the statistics on Offences reported to the police in the same year. Where an offence entails several different victims, all of them are retained for statistical purposes.

The data basis is taken from the police’s central registration system, which covers all offences registered by the police districts (as well as special agencies and higher prosecution authorities) in the police’s databases for handling criminal cases (BL). Registrations in the local BL are transferred to STRASAK, from which Statistics Norway received its data basis extract until 2001. From the statistic years 2001/2002, Statistics Norway has received its data basis from the police’s analysis and leadership system (PAL). PAL contains data on all criminal cases after they have been transferred from STRASAK. The data source for the statistics is referred to as the criminal case register BL/STRASAK/PAL, and contains criminal case data that is transferred between these three systems on an ongoing basis.

See Sources of error and uncertainty under Sources of error and Definitions.

Collection of data

The data material is retrieved electronically from the Norwegian Police ICT Services in mid-January. The Norwegian Police ICT Services is responsible for the output of data from PAL (see Data sources and sampling). The police and prosecution authority will make subsequent corrections to this caseload after the end of the year, particularly in recently reported cases received immediately before the end of the year. Changes and structural updates in BL/STRASAK/PAL may also be made at the end of the year. The data basis for Statistics Norway is therefore not extracted until about two weeks after the end of the statistics period.

Editing

Editing is defined here as checking, examining and amending data.

The police registration systems contain some validity controls of data. In addition, Statistics Norway performs its own controls on some of the material. For example, missing information on scene of crime by municipality will be added to the municipality to the degree this is possible based on other information from the data basis. Duplicated (double entries) offences and victims are deleted, illogical valid values (for example dates) are revised or deleted.

Since 2005/2006, all murder cases have been controlled and revised (case by case) against information from Kripos (which at the end of the year has, for statistical purposes, retrieved data on all murder cases from the districts) and other register data from the Norwegian Police ICT Services.

From the statistic year 2011, all of the cases that were settled by the International Police Immigration Service (PU) during the year were deleted (which equates to around 500 offences in 2011). This has led to a relatively large reduction in the category "Special body" in the statistics that distributes the reported offences by police district, and fewer units in "Other crimes". For the previous years with PU-registrations (2005-2010), it was not possible for Statistics Norway to make this deletion of civil cases. From the statistic year 2015 the amount of deleted cases is considerably reduced, and the last five years (2018-2022) there are nearly any cases registered and settles by PU in the data material.

The statistics dating back to 1993 were completely revised in 2005. The reason for deleting cases at that time was that some cases were transferred from one police district to another. In such transfers, the same report had been registered in both police districts, meaning that they were double entries in STRASAK/PAL. The revision was carried out using information registered by the police itself and from the original data basis for the different years of the offences reported to the police. Where an offence had been registered several times, only the first registration was retained (see Comparability over time and space).

An overview of all controls and Statistics Norway's processing routines for the statistics up to the first half of 2002 is available in Notater 2000/13 "Statistikk over anmeldte lovbrudd og registrerte ofre. Dokumentasjon" (in Norwegian only).

Estimations

Based on the data material from the police’s registers, the statistics counts four units (see Definitions of the main concepts and variables under Definitions);

• The number of offences reported to the police with a specification of characteristics; type of victim, type of offence, group of offence and scene of crime (county, municipality, size of municipality and centrality (-2017).

• The number of unique victims (persons) registered as a victim of one or more offences during the year, and the number of these persons with different characteristics, such as age, place of residence and number of offences they have been subjected to. For victims subjected to more than one offence during the year, the principal offence is attributed based on an approximate indicator on the basis of the minimum and maximum sentences, as well as other criminal procedure characteristics of the offence.
• The number of unique victims of violence and maltreatment, and the number of these victims with different characteristics, such as age. For victims subjected to more than one incicent of violence and maltreatment during the year, the principal offence is used (as for victim above).
• The number of unique victims of property theft, and the number of these persons with different characteristics, such as age. For victims subjected to more than one property theft during the year, the principal offence is used (as for victim above).

A victim (person) is identified based on different information about him/her – e.g. date of birth, personal ID number, sex and name (as opposed to other legal entities, which are identified by organisation number and name). Where there is insufficient information on a person or enterprise/organisation, the victim is categorised as a person if a minimum of personal information is registered.

In addition, calculations of percentages and rates per 1 000 inhabitants are made (also for different ages, genders and regions). All rates per 1 000 inhabitants are calculated with figures for the population as of 1 January in the statistics year and in statistics that use a two-year average, the population on 1 January in the latter of the two statistic years is used (see the population statistics). As a general rule, age is calculated based on date of birth and the date the offence was committed.

Employees of Statistics Norway have a duty of confidentiality.

Statistics Norway does not publish figures if there is a risk of the respondent’s contribution being identified. This means that, as a general rule, figures are not published if fewer than three units form the basis of a cell in a table or if the contribution of one or two respondents constitutes a very large part of the cell total.

Statistics Norway can make exceptions to the general rule if deemed necessary to meet the requirements of the EEA agreement, if the respondent is a public authority, if the respondent has consented to this, or when the information disclosed is openly accessible to the public.

More information can be found on Statistics Norway’s website under Methods in official statistics, in the ‘Confidentiality’ section.

The statistics contain personal data and information on punishable offences that is considered to be highly sensitive. Statistics Norway therefore administers the data basis for crime statistics with great caution.

The statistics are primarily published at a higher regional level (in total and the whole country), and when classified at lower levels (county and municipal) the offences are only published in larger categories – e.g. by group of offence.

Marking cells with few units is not normally necessary on account of anonymity (and neither is it normally done in other crime statistics that are published using the same principles). When releasing statistics with few units, the protection of privacy is, however, considered on a case by case basis. In the statistics on offences reported to the police classified by municipality, in addition to a two-year average, a lower limit has been set for the number of units in the release (fewer than 5). The choice of groups and lower level of detail, as well as exclusion by marking categories with few units, is also made based on an assessment of the data quality and the statistics’ suitability for comparison over time and place (see Coherence with other statistics under Background).

There are many factors that can have a bearing on comparability over time and place in the various crime statistics, and a broader explanation of the problem here is given in Kriminalitet og rettsvesen 2009 (in Norwegian only).

Annual statistics on offences reported to the police are, subject to changes in registration practice and mergers of police districts, comparable from 1993 (see Chapter 5). Quarterly figures were published from 1991 to 1999, and half-yearly statistics from 1999 to 2003, in addition to annual statistics. Since 2004, Statistics Norway has only published annual statistics. The revision of all years dating back to 1993 (carried out in 2005, see Offences reported to the police. 1993-2001. Revised figures. ) was only for the annual statistics, and no new tables were created for the police districts (owners) in the years 1993-2001 or for scene of crime (county) in the years 1993-2000.

Many regional standards are very stable over a long period of time, but regroupings of these also occur – which in some cases is a determining factor in the content of the statistics. One example of a directly visible change took place after the reorganisation and 50 % reduction in police districts in 2002. Most of the districts were given a new name and new geographic borders, and it is almost impossible to compare the content in both the data basis and the statistics that are classified according to the old and new geographic areas (see Offences investigated 2002).

On 1 January 2016, Norwegian police were organised into new regional districts, where 27 police districts were reduced to 12. The internal process of organising the new districts was executed gradually during the course of 2017 and 2018, but as of year 2016, Statistics Norway release statistics by the new police districts in StatBank. The new police districts are mainly organised by merging the former districts, which on an aggregated level are comparable to the former division of 27 districts. However, in the new organisation, the former Midtre Hålogaland PD is divided by the county border between new Nordland PD and Troms PD. Thus, comparisons across the new and former district are not possible for the districts in question. However, derived from information about owner district by the previous organisation, as well as information about the criminal cases belonging in terms of police branches and scene of crime, Statistics Norway have been able to allocate all criminal cases after the new organisation of police districts. All offences with a branch code for the police stations in Harstad, Iberstad and Salangen have been ascribed to Troms PD as the owner district. Furthermore, offences with a scene of crime in Troms County and with branch codes for Evenes, Tjelsund and Skånland police station, as well as "Joint Operational Unit" (Fellesoperativ enhet) and "Eco Team" (Økoteam), have been ascribed to Troms PD as the owner district. Remaining offences registered with the former Midtre Hålogaland PD as the owner district have been ascribed to Nordland PD. For a detailed overview of the new police districts, including their relation to municipalities, see Statistical Classifications and Code Lists.

From 1 January 2020, there have been major changes in the division of counties and municipalities, due to regional reform. The changes are described in this article (only in Norwegian). This also means that the police districts, which in turn consists of a whole number of municipalities, have changed. See Classification of county, Classification of municipalities and Classification of Police district.

Changes in regional standards that are not directly visible take place when, for instance, a municipality changes over to a different county, when two municipalities merge or when a municipality has population changes that result in a different classification in size of municipality (see Definitions).

The most central changes in the police’s registration system and practice that are of significance to the development of offences reported to the police took place around 1994 and in 2002.

New registration practice from 1994/1995

Where a report includes more than one offence, all matters that can constitute a separate point in the charge/indictment are registered. This has been the practice since 1994/1995. Prior to this, the practice was that only one offence was registered when:

• More than one offence was committed in one act against one victim. The most serious was registered.
• More than one of the same type of offence was committed by the same person within a period of time against one victim.
• One offence was committed against a group of victims.
• The first offence was committed in order to pave the way for the second.

However, more than one offence was registered when an offence was committed with the purpose of hiding another offence, which was also the practice after 1994/1995.

STRASAK has, for a long time, been under development with regard to quality assurance and registration details. This development has led to the data basis for the statistics containing more information than previously. The transition in 1994/1995 to registering all offences is of the greatest significance, and resulted in the number of offences reported to the police increasing in the statistics without this necessarily being related to more offences actually being reported.

Changes to STRASAK in 2002

Towards the end of 2002, the police’s criminal case register, STRASAK, underwent a major change. There are three factors in particular that have a specific bearing on the statistics on offences reported to the police:

1) Merging of police districts; as from 1 January 2002, the former 54 police districts were merged into 27 police districts. This means that the procedure of adding a value for the municipality of crime (see above) can be applied with a fewer number of police districts.

2) Duplicate registrations; In October2002, acentral serial number instead of the previous local serial number was introduced. In the long term, this will ensure that transferred cases will no longer be re-registered in the new police district, and such duplicate registrations (see Collection of data revision and estimations) will be reduced from 2003.

3) Scene of crime abroad; some offences reported to the police are committed abroad, but some are reported inNorway. Prior to 2002, no information on these was available in the material for the crime statistics. These offences were therefore previously included in the statistics and allocated a municipality of crime or county of crime according to the routine described above. As from 2002, it is possible to identify these offences, and they have been retained in the statistics with the scene of crime given as abroad.

Municipality statistics

As from 2010, Statistics Norway publishes detailed municipality statistics on offences reported to the police (2007-2008). Statistics Norway has also previously published statistics on offences reported to the police in the individual municipalities of crime (Styrings- og informasjonshjulet 2008 (in Norwegian only). Old (2003-2004 to 2006-2007) and new statistics are relatively similar and relatively comparable for the categories that have the same type of basis and classification. Due to the quality of the data, however, we have decided not to publish new figures for districts for the sample boroughs.

When using statistics at such a level of detail, it must be borne in mind that there may be great variation, particularly for the municipalities and groups of offences with few cases. Data quality and registration method, particularly in major cases, vary and can lead to considerable differences between otherwise identical municipalities – and from year to year. In both the new and old statistics, certain consideration has been given to this type of statistical random variation by creating a two-year average, but only for the largest groups of offences and the municipalities that have more than four units.

Introduction of the new penal code on 1 October 2015

After the new penal code (Penal Code of 2005) came into effect on 1 October 2015, the raw data from the police registers are changed. There are quite distinct differences in structure, language and content between the Penal Code of 1902 and the Penal Code of 2005, which is even more evident in the corresponding codes used for registering offences in the police’ central registration system. Code connected to both the new and the former penal code is used after 1 October 2015, due to the new legislation not having retroactive effect. Offences may be registered and processed long after they have been committed. This means that the Penal Code of 1902 and 2005 will appear in the data of all crime statistics for many years to come.

Criminal cases with violations of the provisions dating before the introduction of the new penal code (registered by former codes and appellation), will not always be possible to specify and classify in the same way as criminal cases with a violation of a corresponding provision from the Penal Code of 2005. Certain types of offences in Standard for types of offences 2015 will therefore exclusively contain violations of offences stemming from new penal legislation, and mainly consist of offences committed after 1 October 2015. In the standard, these will be indicated with “(from 1.10.2015)” in the classification name. Older offences which cannot be specified in the same way are placed in “other or unspecified” types of offences. The content of the statistics – and the possibilities of doing comparisons on the most detailed levels of the standard – will for certain types of offences therefore be mainly based on offences committed after 1.10.2015. This type of break in the time series will vary over time for the different types of offences, depending on which crime statistics you use. When such breaks in the time series are identified, the general recommendation is to use figures on a more aggregated level, which means less specified figures with a smaller degree of time series break. Se Standard for type of offences 2015 in KLASS and the article “The New Crime Statistics” (in Norwegian only).

Other changes

Releases from 2005 and onwards use the figure as per 1 January in the statistics year in calculations where population figures are included in the statistics on offences reported to the police (as opposed to previously, when the figure as per 31 December in the statistics year was used). This will be the practice for all statistics released from this point in time, also for previous years.

As from 1 January 2005, Kripos and the PU can register and investigate their own criminal cases. The category “special body" was therefore introduced in the statistics from 2005, classified by police district. This category includes the new district units, together with Økokrim (which has been included in the statistics since 1995).

Changes to legislation that affect the content of the statistics directly and considerably shall, as far as Statistics Norway is aware of such changes, be highlighted in the tables in the years where such changes have taken place – and where this is relevant to corresponding statistics over time. For examples of this, see Offences reported to the police 2006 (for example on the abolition of the Vagrancy Act).

Since the statistics year 2005/2006, all murder cases have been controlled and revised (case by case) against information from Kripos and register data from the Norwegian Police ICT Services (see Collection of data revision and estimations). Some cases have subsequently been changed from “other type of case" to “reported murder", but the opposite has occurred more often. After this revision, the murder figures can therefore be somewhat lower than previous years.

Merger of the offence and victim statistics from 2019

From 2019, the statistcs on Offences reported to the police and Victims of offences reported to the police are data processed, quality assured and published as one statistic. Compared to previous years, certain adjustments are made in the data processing. These changes include the choosing of offence code in cases where an offence is registered with more than one offence code, as well as the choosing of the principal offence for victims (persons) in cases where a person has plural victimisatons of offences with equal severity. However, these changes will only affect offence codes (and age) in a maximum of one hundred cases, are are therefore not considered as a break in the time series.

Subject to subsequent changes, more detailed information on both the old and the new data basis, variables and definitions are described in Notater 2000/13 “Statistics on offences reported to the police and registered victims. Documentation” (in Norwegian only). StatisticsNorway has relatively comprehensive documentation of the data basis and the statistics on the individual years, and information on this can be obtained from the contact persons specified for the statistics.

After undertaking a thorough review of various data basis and crime statistics, and dialogue with the Norwegian Police ICT Services, Statistic Norway believes that the revision that was carried out in 2005 has led to a considerable improvement in the quality of the statistics on offences reported to the police. However, there is some uncertainty in connection with the precision of deleting cases, and whether the revision has led to the deletion of duplicate registrations to the same extent throughout the period.

Offences that lack information on municipality of crime are, by means of internal controls, allocated a municipality of crime to the degree this is possible. Offences are largely investigated in the police district where the act took place. It can therefore be assumed that the offence has been committed in the relevant police district unless there are indications to the contrary. To the degree the police district does not cross one or more municipal or county borders, the offence is deemed to have been committed in this municipality/county. Where the municipality of crime is added using this method, it is considered to be relatively certain. This has been the practice in all previous years of offences reported to the police, but such partial non-response has had increasingly less significance in recent years.

Statistics Norway has no possibility of controlling the non-response of units where there is incomplete registration in individual police districts. Partial non-response (where certain variables are not given for some units) occurs, but with the aid of internal controls in BL/STRASAK these are limited to less important variables.

Statistics on offences reported to the police are based on registrations in the first stage of the legal system’s proceedings. This means that the quality is tested and controlled to a lesser degree, and is thereby more uncertain than in the other parts of the crime statistics (see Coherence with other statistics under Background). An analysis of offences of which the investigation was completed in 2010 (Stene 2020), the scope of offence code alteration over the course of an investigation is proven to be quite high for certain types of offences.

Due to the statistical units for offences are chosen on the basis of case numbers and split-case numbers in the police registers, Statistics Norway in some instances has to choose a representative for the offence in case of different offence codes. These cases are liimited to a couple of hundre cases per year. The principles of choosing offence codes were carried out in somewhat different ways for the previous statistics on Offences reported to the police and Victims of offences reported to the police, which prompted a harmonisation of these cases during the data processing of the victims statistics. During the joint data processing of the two statistics in 2019, this problem has been solved by adjusting the previous processing principles. However, the scope of such cases are so limited, and offence code variation spread across so many different types of offences, that these changes will not affect the statistics in any significant manner.

As from 2002, the data basis for the statistics is extracted annually from the same register in the same way and at roughly the same time. Likewise, Statistics Norway’s processing of the data basis should be the same every year. This will eliminate sampling errors. The principles are the same, but practical and technical changes in extraction and processing may be made which could potentially entail adjustments of the content of the sample that is used for statistical purposes.

Definition limitations (see Definitions) can have a bearing on the sample, without this necessarily being regarded as a sampling error. Investigation cases, such as fires, missing person reports and suspicious deaths are also registered in the criminal case register. If one of these cases is not registered with an offence at the time of extraction, the case is not deemed to be a criminal case reported to the police, and is not therefore included in the statistics.

Offences and criminal cases without a registration date in the statistical year at the time the data is taken from the police’s central registration system (normally approx. January 14th), will not be included in the data basis. An offence which at a later point of time is re-registered as an offence with registration date within the statistical year will therefore not be included in either data basis for the statistics on Offences reported to the police (or Victims of offences reported to the police). Albeit, these kinds of re-registered offences will, when settled by the police and prosecution authority, be included in the data basis for the statistics on Offences investigated.

Various tests on different data basis for the time period 2008-2011 indicate that these late-registrations have led to 1-2 per cent of the offences actually reported to the police not being a part of the statistics annual data sample. Statistics Norway has not obtained the data basis for the mapping of this phenomenon before the statistical year 2008. However, from the technical possibilities and knowledge of registration routines in the Police’s central criminal case register (BL), as well as police statistics, Statistics Norway has reasons to presume that the scope of such late-registrations were as many or more in the years prior to 2008.

In 2012 and the years after, there were far fewer late-registrations than in previous years. The annual average in the periode 2012-2014 was about 1 200 offences (1 000 of which were crimes), amounting to only 0.3 per cent of all offences (0.4 per cent of all crimes) in the published statistics. Derived from the average in the time period 2008-2011, there were annually about 7 000 offences (4 800 of which were crimes) subject to late-registration, thus not included in the published statistics.

The scope of such late-registrations is unevenly spread out on different groups of offences and types of offences. Larcenies and other types of offences for profit are rarely subject to late-registration, and this methodological phenomenon therefore does not significantly influence the use of the statistics on these types of offences. For other groups of offences, however, these late-registrations have a bigger impact on the figures, especially when figures from 2011 and before are compared to statistics from later than 2011. For example, the following group of offences were the subject a of significant scope of late-registration: nearly 1 500 more narcotic offences, 1 300 more traffic offences, 500 more offences of violence and 150 more sexual offences compared to the average number of late-registrations in 2012 and 2013. When attempting to characterise the actual development for the number of offences reported to the police during these time periods, one must account for the fact that these statistics up to and including 2011 are "missing" a number of offences in the given magnitude. Even though the change in routines for late-registration has significance for the statistics comparability over time, this phenomenon is not considered to be an actual source of error in this statistic. This is because the criterias which are used draw the data sample for these statistics has been constant for the whole time period.

Statistics Norway does not have data to precisely quantify the impact of these registrations on how many victims (persons) in the reported offences have not been included in the statistics. As stated in About the statistics on offences reported to the police, we know that there is a relatively high proportion of narcotic offences that have not been included in the statistics of reported offences in the years before 2011. This does not affect the statistics on victims, in particular, because these offences rarely have registered victims. We also know that property theft, which constitute a very high proportion of all offences with victims, and where three out of four of the reported offences are registered with persons as victims, have seldom been re-registered. Changes in the scope of late registrations for other groups of offences, such as offences of violence and sexual offences, could have a somewhat greater impact – especially when the figures from 2011 and previous years are compared with the statistics for recent years.

Although the scope of the victims, i.e. everyone with the role of «victim» in all reported offences, should be somewhat greater, especially in the years before 2011, this indicates that the late registrations have a smaller impact on the statistics on victims of offences reported to the police than on the statistics containing only the reported offences.

The crime statistics only include registered crime, i.e. the offences and victims that are identified by or are made known to the police. Police resources and priorities, and people’s tendency to report offences to the police are of great significance to what acts become a part of the crime statistics. An unknown number of offences are never registered. Such unregistered offences are often called dark/missing figures. Different surveys on self-reported crime and exposure to crime indicate that the crime that is not registered is extensive, and will vary according to type of offence (see Survey of level of living conditions ). What crime is registered over time is also affected by changes in legislation and regulations. These phenomenons are, however, not considered as "errors" in the statistics, but elements that affect them. For more details of this, see Kriminalitet og rettsvesen 2009 , kapittel 3 (in Norwegian only).

Statistics Norway can, to a certain extent, quality assure the data basis, but must - to a greater degree than in the statistics on perpetrators - use the information as it is in the police register at the time of extraction. Statistics Norway has a reasonably clear impression that the police registration of victims does not follow such strict requirements for precise and correct identification of role, or filling in of information on the victim compared to the registration of perpetrators. This means there is greater uncertainty in relation to the level of accuracy of the statistics on victims, both with regard to the scope of victims and whether the right person is identified as the victim in the case. However, the personal information on persons designated as victims in 2004-2010 had a very high level of completion - and by using several variables and sources there are very few victims who have an unknown value for the most central demographic variables (sex, age and place of residence) in the statistics.

Statistics Norway has few possibilities for controlling non-response errors of units in the event of insufficient registration of known and unknown offences and victims. There are, for instance, relatively large differences between the police districts’ shares of offences that have no information on victims. Although there are other conditions that are the main reason for these discrepancies between the districts, we cannot rule out some of the differences being due to varying registration practices. A varying practice in the registration of victims will thus result in a varying non-response of known victims in different parts of the country. Partial non-response (when some variables are missing on some units) occurs, but with the aid of internal controls in BL/STRASAK these are limited to less important variables.