Data revision policy
When statistics is revised, an already published value or a figure is changed. This may happen e.g. because new and more reliable source data has become available, because of changes in methods, classifications and definitions or because of error corrections. Revisions are usually planned, and the users are notified in advance.
Statistics are generally revised to incorporate new and improved source information. Revision of published data is an essential part of producing high-quality statistics.
Data revision is a normal process in the production of European statistics to improve their quality. Revisions follow standard, well-established and transparent procedures.
The application of the Revision Policy in different statistical areas is an key aspect of "good governance practice" in official statistics and its importance is equal both at the national and at the international levels.
The Revision is an complementary part of the regular production process of many statistics, which contributes to the quality of produced statistical data, as well as their accuracy.
Bearing in mind that the mission of the State Statistical Office is production of high-quality and comparable statistical data, the Revision Policy constitutes an important instrument for improving the quality of statistical data.
The main objective of the Revision Policy is to achieve broader transparency of the overall procedure of data revision, and it will increase confidence in official statistics and the usefulness of statistical data.
Major revisions
Major revisions are substantial changes to the published data, due to various reasons: change of sources, updating of weights, etc. That is why it is necessary to make a return of the series, because otherwise there will be a break in the series.
The reasons for which Major revisions are made are the following:
- Availability of a new structured data source that is collected at a longer interval (eg every 5 or 10 years), such as censuses and other large statistical surveys.
- Update base year weightings of index series, typically every 5 years.
- Change in concepts, definitions and classifications used for the production of statistical data.
- Entry into force of a new legal act.
Usually, these revisions make major methodological changes in statistical research. Major revisions are expected and planned in advance. Users are informed in advance of upcoming major revisions and the change of the time series is announced.
1. Revision of data for the Population Census
The State Statistical Office carried out a return of the series of population estimates in the period between the 2002 Census and the 2021 Census. This activity of returning the population data series is a common worldwide practice after every census conducted. When using population estimate data, it is extremely important that you keep the following information in mind in order to use the most relevant data available from our website.
1.1. Methodological information
1. The data set in the 2002 Census Population Estimates folder (archived data), for the period December 31, 2002 to December 31, 2020, was calculated based on the data from the 2002 Census, taking into account the number of births, deaths, immigrants and displaced persons, according to the municipality where the persons lived and the year when the event occurred. All these data are obtained from administrative sources.
2. Unlike administrative data, censuses are conducted in the field and therefore certain differences in the number of the population and its distribution across municipalities are expected to appear between two censuses.
3. The difference in population numbers between the estimates made according to the 2002 Census and the results of the 2021 Census is due to unregistered emigration.
4. Therefore, the SSO performed a population assessment with a starting state from the 2021 Census, returning the calculations back to the 2002 Census.
5. Estimates return calculations are intended to provide more relevant population size data based on a more recent historical situation. That is, calculations based on data from the 2021 Census versus data from the 2002 Census.
6. Based on the available data, the SSO managed to identify displaced persons who are not registered in the administrative sources, in contrast to those persons who were not enumerated in the 2021 Census, and who are included in the total population.
7. Unfortunately, the SSO failed to find relevant sources of data for the year when the persons moved away.
8. Due to this shortcoming, the methodology used to estimate the population is based on a modified linear interpolation in which the modification refers to the intensity of migrations per calendar year, obtained from the statistics of other countries (known as mirror statistics).
9. As a consequence of these factors, the SSO calculated the number of the population in the period between the two Censuses only at the state level.
10. In addition, SSO calculated the total number of population and by statistical regions for the period 2017-2021.
1.2. Guidance to data users on the use of the revised population data series
1. Valid official population data are the data calculated based on the 2021 Census, shown in the Population Estimates according to the 2021 Census folder.
2. The lowest level for which this data is available is a statistical region, but only for the period from January 1, 2017 to January 1, 2021.
3. Data on population estimates according to the 2002 Census are archived historical data.
1.3. Revision of data from the Labour Force Survey (2017-2021)
In accordance with the published results of the Census of Population, Households and Dwellings in the Republic of North Macedonia in 2021, the State Statistical Office (SSO) revised the data from the Labour Force Survey (LFS) for the period 2017-2021. This revision was made in order to ensure greater accuracy and relevance of the statistical data.
Using the latest data series from the Census 2021 and applying the calibration method with SAS Macro Calmar, the data were adjusted according to the population by sex, five-year age groups and regions. This procedure made it possible to make new calculations and revise the data series for the period 2017-2021.
The new calculations and revised data are available on our website. These data are extremely important for labour market analysis, and we invite all interested parties to review and use them for their analyses and research.
2. Change of base year
The State Statistics Office, in order to follow the recommendations of Eurostat, in the Regulation on short-term statistics in 2024 changed the base year for all short-term surveys in the field of industry, prices, construction and trade. The new base year from January 2024 is 2021. All indices up to 2024 have been revised to the new base year and are published in the Maxstat database of the SSO.
Indices up to 2023 with base year 2015 are archived and are also available in the Makstat database in the respective areas.