Comparability

Comparability - geographical

No comparable results from other countries are available. Though, some countries have statistics on absence which are based on other sources. This could be information from The Labour force survey or information on sickness benefits.

Comparability over time

The statistics on absence reflects the structure of absence which means that the statistics reflects the level of absence for different groups of employees at a certain period of time - in this case on a yearly basis. The statistics doesn't take structural changes into account - changes that might have happened within groups of employees between two years. New enterprises are established, old enterprises close down while some enterprises change their primary activity. Furthermore, within the enterprises some significant changes of the staff might have taken place, for instance dismissals or changes in the characteristics of the employees. These so called structural changes between two years are some of the terms of structural statistics in general and specific of the statistics of absence. Comparisons between years should fundamentally only be made with reservations.

Furthermore the quality of the data from enterprises, municipalities, governmental institutions etc. are expected to improve over time. An increase in absence from one year to another might to some extent be caused by improved registrations of absence. Comparisons between years concerning the earliest years should in general be made with caution.

Reallocations of occupation codes: In general, larger or smaller displacements within occupation groups may occur when enterprises and public institutions change the occupation of their employees in the registration.

• Municipal and regional government 2021:* Due to the strike of nursing professionals in summer 2021, the quality of the monthly earnings data is not satisfactory for these groups. Statistics Denmark has tried to identify the jobs that have been affected by the strike. The weeks that the strike occurred are excluded from data for the identified jobs.

Defective registrations from The National Church: The first of January 2018 The National Church changed earnings system and system for registering absence. This has had the impact that the registrations of absence for 2017 have been defective. Consequently, only a smaller part of The National Church is included in the statistics of 2017. This will have an impact on the industries and occupations of which the employees of The National Church are a part. This concerns the governmental sector and especially the industry ‘Activities of membership organizations’ and the occupation ‘2636 Religious professionals’. In 2018 registrations are gradually improved which is reflected in the industries and occupations of which the employees are a part.

Changes in maternity leave in the general government sector: Periods of maternity and adoption leave in the general government sector are based on data from the register of benefits during sickness or in connection with childbirth, as a number of the respondents have not reported any periods of maternity leave. In 2017 the system for administrating the benefits in connection with childbirth was changed. Final data were not received before the publication of the statistics of absence 2017, which means that the published data concerning maternity and adoption leave for 2017 are uncertain. It is for instance uncertain whether the definition of maternity and adoption leave is comparable to previous years.

The new data includes a variable that states the extent of the absence compared to 37 hours. This makes it possible to calculate the number of hours of absence more precisely and has the impact that it is now possible to include part time maternity and adoption leave. The use of the new data has especially for men caused an increase in paternity and adoption leave. In 2018 the system for delivering the data is not yet completed which means that the published data are still uncertain.

Changes in the classification of education: Between 2016 and 2017 there have been changes in the education nomenclature. These changes mainly concern the sector Government including social security funds. The main change is associated with one specific education group which was previously placed in Medium-cycle higher education. In 2017 this education group is instead placed in Short-cycle higher education.

Automatic data validation and feedback to corporations and organizations: In 2017 automatic and manual feedback concerning the statistics of earnings has contributed to an increase in the number of fulltime employees in the sector corporations and organizations. This has the impact that more enterprises can be included in the statistics of absence.

Reallocation of certain local units between sectors: In 2017 further privatization of some parts of the sector general government has led to reallocations with regards to some occupation groups and industries. This means that some units that previously belonged to the sector general government now belongs to the sector corporations and organizations.

Reallocations of certain occupation codes: Automatic and manual feedback has entailed an improved quality with regards to occupation codes, which has brought about a few changes within some occupation codes in the sector corporations and organizations in 2017.

During 2016 the sector of enterprises and institutions has been evaluated thoroughly. This has caused a change in some enterprises and institutions sector compared to previous years.

The municipality of Copenhagen has in 2016 changed their pay roll system and system for registering absence which means that comparisons between 2016 and previous year should be made with caution. When systems are changed, periods of absence might be registered twice or with different dates in the two systems. These periods have been deleted when possible. Furthermore, in the old system half of the periods of absence were registered without the actual hours of absence and the hours were calculated assuming that a full-time employee works five days a week and 7,4 hours a day. In the new system only a small part of the periods of absence are registered without hours of absence. In the data for 2016 a part of the periods of absence have more than 7,4 hours of absence a day. The calculate hours of absence for the previous years might consequently be underestimated.

New classification of education: In 2015 the Denmark Statistics' education statistics started using a new classification of education called DISCED-15, which replaced the previous, DUN and Forspalte1. This causes a break in the time series, thus the series FRA021 is discontinued. The new series for earning by highest level of education is called FRA021A, from where tables for 2015 and beyond may be extracted. DISCED-15 is the Danish version of the ISCED 2011 classification, which was developed by UNESCO in consultation with relevant international organizations and partners, such as Eurostat and the Organisation for Economic Cooperation and Development (OECD). The ISCED-classification is meant to contribute to the production of even more reliable and comparable international statistics on education, reflecting the ongoing evolution of education systems worldwide. Fore more information see https://www.dst.dk/da/Statistik/dokumentation/Nomenklaturer/ny-uddannelsesklassifikation-DISCED-15

Displacements of certain occupation groups: In 2014 the wage and salary administrations for the municipal- and regional sector were able to report occupation codes for their employees. Previously the wage specification of employee was used to decide the occupation codes. This has caused displacements within a few occupation groups in the municipal- and regional sector from 2014. These displacements still occur in 2018. This is also an issue for the governmental sector where displacements within some occupation groups have occurred in recent years.

Changed sector definition from 2013: This publication presents the Statistics of Absence for the first time according to the new European national accounting manual ESA2010. The naming of sectors has changed from Private sector to Corporations and organazations, from governmental-, the municipality- and the regional sector to Government including social security funds. In addition to the changed naming the substantial definitions of the sectors are changed. (See News from Statistics Denmark 2014 no. 270[www.dst.dk/nytudg/18045] and News from Statistics Denmark 2014 nr. 321[www.dst.dk/pukora/epub/Nyt/2014/NR321.pdf]. Publications according to the new sector definition are not comparable to previous publications according to the old sector definition.

Changes in the education nomenclature Between 2013 and 2014 , there are many changes in education nomenclature This gives some differences between some education groups in 2014. The changes concern all sectors.

Central government sector 2013 Periods of maternity and adoption leave in the central government sector are based on data from the register of benefits during sickness or in connection with childbirth, as a number of the respondents have not reported any periods of maternity leave. In connection with moving the administration of benefits concerning childbirth from the municipalities in December 2012 some registrations concerning childbirth have not been correct. The published maternity leave for the central government sector for 2013 is consequently based on data where some of the periods have no been correct. The published maternity leave for 2013 is estimated to be underestimated. This is probably especially the case for the women's maternity leave.

Municipalities 2013: Due to the lockout of certain groups of teachers in spring 2013, the quality of the monthly earnings data is not satisfactory for these groups. Statistics Denmark has tried to identify the jobs that have been affected by the lockout. The weeks that the lockout occurred are excluded from data for the identified jobs.

Changes in then DISCO-nomenclature, 2010: Up to and including 2009 the grouping by occupation is based on the variable DISCO-løn. From 2010 the new variable, DISCO-08, is used instead. DISCO-08 is a revised version of DISCO-løn. There is no unique translation between DISCO-løn and DISCO-08. For this reason earnings by occupation is not comparable between 2010 and previous years.

Since 2005 data for the central government sector include absence of less than whole days, unlike the previous years. Consequently, the real change from 2004 to 2005 is considerably lower than the numbers indicate.

Data on absence in The City of Copenhagen municipality in 2005 are not as reliable, as it has been difficult to match reports on absence with employment data.

From 2006 the method of linking periods of absence has been improved concerning holidays. This has an influence on the number and length of the periods of absence.

In connection with the reform of municipalities upper secondary schools were in 2007 transferred from the municipal sector to the governmental sector. Furthermore, due to the reform of the municipalities the match between the absence periods and their belonging job has caused great problems. Consequently, comparisons between years should only be made with caution. In 2008 it has been especially difficult to attach absence periods to its belonging jobs in the health sector.

Change of systems to register absence data might cause problems with the quality of the reported periods of absence and consequently the results of the units concerned might be uncertain. This has been the case for parts of the local government sector from 2010 and forward.

Due to a strike within some municipal areas in the spring of 2008, the quality of the monthly data reports on earnings is not satisfactory for some groups of employees. This might have an impact on the number of absence periods that it has been possible to connect to the belonging job.

Coherence - cross domain

KRL, an actor in the municipal sector, publishes a statistic on absence for the municipality and the regional sector based on the same data that the statistics from Statistics Denmark is based on. However, the methods and definitions differs and consequently the results of the two statistics are not comparable.

Moderniseringsstyrelsen (The Agency for the Modernisation of Publlic Administration) publishes statistics on absence for the central government sector based on the same data. Due to differences in methods, results of the two statistics are not comparable.

Danish Employers' Confederation (DA) and Danish Employers' Association of the Financial Sector (FA) publish statistics on absence for their members. The data submitted to these two organizations are the basis for the statistics of absence together with the absence data submitted to Statistics Denmark.

Coherence - internal

Different systems are used for registering absence. This is the case between the different sectors but also within individual sectors. In the private sectors enterprises have for instance the possibility to use a wide variety of systems. This means that the absence is not necessarily registered consistently. There are, however, prepared similar guidelines for registration of absence for all sectors in order to ensure a consistent registration as possible.