Reference metadata describe statistical concepts and methodologies used for the collection and generation of data. They provide information on data quality and, since they are strongly content-oriented, assist users in interpreting the data. Reference metadata, unlike structural metadata, can be decoupled from the data.
Estonian Classification of Economic Activities 2008 (EMTAK 2008) based on NACE Rev. 2
3.3. Coverage - sector
The survey covers NACE Rev. 2 Sections F.
3.4. Statistical concepts and definitions
The total constructon volume index is based on the deflated value of construction production. The value of construction production = turnover from construction activity + construction for own use +/- changes in the stock of work in progress - value of work done by subcontractors. The production value of the reference quarter is compared to the production value to the same period of the previous year and the year-on-year percentage change is calculated. The indices at current prices are deflated with construction cost index to obtain the production volume index. Then indices are recalculated to the base 2021 = 100.
3.5. Statistical unit
Reporting unit - enterprise, observation unit - KAU.
3.6. Statistical population
The Statistical Business Register based list of active economic units which are active in the field of construction. There are about 16500 enterprises in the target population.
3.7. Reference area
The activities on the national territory as well as outside the national territory are included. Construction volumes in foreign countries accounted for 7% of the total volume of construction in 2024.
3.8. Coverage - Time
The construction production index as total was compiled for the first time in 1994. Since 1998 the production index was recalculated according to NACE Rev. 2.
3.9. Base period
Data is calculated on the base 2021 = 100.
Index
Quarter
6.1. Institutional Mandate - legal acts and other agreements
Regulation (EC) No 2019/2152 of the European Parliament and of the Council of 27 November 2019 on European business statistics, repealing 10 legal acts in the field of business statistics (Text with EEA relevance)
6.2. Institutional Mandate - data sharing
Data are transmitted to Eurostat.
7.1. Confidentiality - policy
The dissemination of data collected for the purpose of producing official statistics is guided by the requirements provided for in § 34 and § 35 of the Official Statistics Act.
Notifications about the dissemination of statistics are published in the release calendar, which is available on the website. On 1 October each year, the release times of the Statistical Database, news releases, main indicators by IMF SDDS and publications are announced in the release calendar (in case of publications – the release month).
8.2. Release calendar access
The publication calendar of news releases and the statistical publications are available on the web site of Statistics Estonia.
8.3. Release policy - user access
All users have been granted an equal access to official statistics: this means that the dissemination dates of official statistics have to be announced in advance and no user category (incl. Eurostat, state authorities and mass media) can have access to the official statistics (results of official statistical surveys) before other users. Statistical information is first published in the Statistical Database. In case a news release is published based on the same data, the information provided in the relevant news release is simultaneously published in the Statistical Database. Official statistics are available on the website at 8.00 a.m. on the date announced in the release calender.
Quarter
10.1. Dissemination format - News release
The news release „Construction” once a quarter. The news release can be viewed on the website.
10.2. Dissemination format - Publications
Not published.
10.3. Dissemination format - online database
Data are published under the heading „Economy/ Construction” in the Statistical Database
Construction volume indices are compared to the fixed base period, compared to the same period previous year and compared to the previous period, according to the construction type.
10.4. Dissemination format - microdata access
The dissemination of data collected for the purpose of producing official statistics is guided by the requirements provided for in § 34, § 35, § 36, § 37, § 38 of the Official Statistics Act. Access to micro-data and anonymisation of micro-data are regulated by Statistics Estonia’s Procedure for dissemination of confidential data for scientific purposes
10.5. Dissemination format - other
Data are sent to Eurostat both to be used in European aggregates as well as to be released as national data.
10.6. Documentation on methodology
A methodological description of the survey is available in the website of Statistics Estonia (ESMS metadata).
10.7. Quality management - documentation
Data on the quality of the population, sample and respondents are published in the Statistical Database.
11.1. Quality assurance
To assure the quality of processes and products, Statistics Estonia applies the EFQM Excellence Model, EU Statistics Code of Practice and the ESS Quality Assurance Framework (QAF). Statistics Estonia is also guided by the requirements provided for in § 7. „Principles and quality criteria of producing official statistics” of the Official Statistics Act.
11.2. Quality management - assessment
The quality assessment of the quarterly survey, which is the data source for most STS indicators, is based on quality requirements of the ESS (relevance, accuracy, timelines and punctuality, accessibility and clarity, coherence and comparability). The inspection of statistical survey, testing of statistical questionnaire and updating survey methodology, self-assessment of survey manager are carried out. The quality of this indicator is good. Weaknesses: indices are calculated basing on micro data having sizeable non-response. Strengths: length of time series; good consistency with SBS data.
12.1. Relevance - User Needs
Ministry of Economic Affairs and Communications
Bank of Estonia
Media
Researchers and students
Enterprises / businesses
Eurostat
12.2. Relevance - User Satisfaction
Users‘ suggestions and information about taking them into account are available on the website
12.3. Completeness
In compliance with the rules (regulations).
13.1. Accuracy - overall
The type of survey and data collection methods ensures sufficient coverage and timeliness.
When calculating the indicator estimates, the statistical data analysis and editing of outliers are applied. When estimating the error, the influence of non-response, imputation of missing values and misclassification on the precision of estimates was taken into account.
The sampling error is small. In general the data are not revised after they are first published.
13.2. Sampling error
The data are based on a survey. Data are collected from all construction enterprises with 20 or more employees. For small enterprises (with 1 to 19 person employed) a stratified simple random sampling is used. For calculation estimates Horvitz-Thompson estimator is used.
Due to the probability sampling the standard error estimates have been computed for all aggregated data. In 2024, the coefficient of variation of variable "total turnover from construction activity” varies between 4,4% to 6,1% by quarters.
13.3. Non-sampling error
It is compulsory to respond to the survey. The response rate is about 65% 50 days after the end of the reference period. Missing responses are requested by telephone contacts followed up by e-mail reminders. The data of non-responded enterprises with more than 19 persons employed and essential smaller enteprises are imputed based on administrative data or data of the previous period. To compensate the non-responded enterprises with less than 20 persons employed, recalculation of weights is used. The data of non-responded enterprises were replaced by the respective average of responded similar enterprises. The impact of editing, coding, imputation and modelling errors is not calculated yet. In the case of imputation, the data are mostly influenced by the differences in definitions between administrative and statistical data.
At the time when data are provided to Eurostat and published nationally the response rate is the same.
There are negligible non-sampling errors concerning coverage, multiple listings and data collections. Improvement actions taken include registry updates, careful logical and accounting controls and constant contact with the reporting units. The multiple listings are precluded because all of the enterprises are identified by a unique number. For data collection eSTAT (the web channel for electronic data submission) is used. eSTAT is also used to monitor the completion of questionnaires and includes instructions and arithmetic and logical controls. The questionnaire cannot be submitted, if it is not correctly completed. Therefore, the item non-response occurres very seldom.
14.1. Timeliness
The data are released at the latest 60 days upon the end of the reference quarter (T+60).
14.2. Punctuality
The data have been published at the time announced in the release calender.
15.1. Comparability - geographical
Comparable within EU countries.
15.2. Comparability - over time
The data are comparable across time.
15.3. Coherence - cross domain
Data are related to statistical activities 21405 „Quarterly National Accounts” and 20300 „Annual Financial Statistics of Enterprises”.
The data are also confronted with data from VAT declarations of the Estonian Tax and Customs Board.
15.4. Coherence - internal
Indices are internally coherent. Higher level aggregations are derived from detailed indices according to predefined procedures.
The cost and burden are determined for statistical survey in total and it is not possible to calculate for separate indicators. The survey covers whole business sector and the information for the next variables are collected: turnover, persons employed, hours worked, wages and salaries, stocks and investments.
The assessment of cost and burden covers the statistics for NSI as well as STS needs.
In 2024 total burden (respondents): 31390 hours per year.
Total cost (NSI): 4700 hours per year.
17.1. Data revision - policy
The data revision policy and notification of corrections are described in the dissemination policy of Statistics Estonia
Vintages are stored but not published, they are available by request. No special revision calendar exists and revisions are not published.
The same revision policy is applied to data released nationally and transmitted to Eurostat.
17.2. Data revision - practice
Data may be revised when the data for next quarter are added. Data may be revised due to belated questionnaires or new corrected versions from reporting units are taken into consideration. Regular revisions are quite small and usually do no affect totals. If the data are revised then it is marked in footnotes under the tables.
In case of irregular revisions, related to methodological improvements, users have to be simultaneously notified of planned revisions and the publication of the results. Users have to be notified of the release date of revised data three months in advance.
The quality indicators used in the data revision practice are: MAR (Mean Absolute Revision) and MR (Mean Revision). The calculated values for the last 20 quarters (IQ 2020 to IVQ 2024) PVC index are:
Annual growth rates for original data series (YoY) MR = 0,3% , MAR = 1,4%
Annual groth rates for calendary adjusted data series (YoY) MR =0,3%, MAR = 1,4%.
18.1. Source data
The data is based on a survey. Data are collected from all construction enterprises with 20 or more employees. For small enterprises (with 1 to 19 person employed) a stratified simple random sampling is used. There are about 16500 enterprises of which 5% (800) are included in the survey.
Data from statistical activities 20007 "Financial Statistics of Enterprises (quarterly)“, 20300 “Annual Financial Statistics of Enterprises“ and 20401 “Construction Cost Index“ are used.
18.2. Frequency of data collection
Quarterly.
18.3. Data collection
Data are collected through eSTAT (the web channel for electronic data submission). eSTAT is also used to monitor the completion of questionnaires. The questionnaires have been designed for completion in eSTAT by the respondents themselves and they include instructions and controls. The questionnaires and information about data submission are available on Statistics Estonia’s website. Data are collected with the quarterly statistical questionnaire „Majandustegevus.Ehitus”.
18.4. Data validation
In order to ensure data quality, verification procedures are carried out: data verification at the data entry stage and error correction according to data checking protocol; the data completeness and the rate of non-response and missed data are determined; secondary editing (analysis of the primary data set and aggregate data) is exercised to identify and correct outliers. The statistical results are compared to the previous period, and with the corresponding period of the previous year.
Before transmission to Eurostat, the structure of the data file and format is checked in detail.
18.5. Data compilation
Non- response is treated by imputation with the VAT data, with the data of the previous period or with the average of the stratum. The value index is deflated using a construction cost index to produce a volume index. The production index is a Laspeyres chain index.
18.6. Adjustment
The production value data are deflated with the construction cost index.
The time series are calendar and seasonally adjusted and trend- cycle are produced, using the TRAMO/SEATS method in JDemetra+.
Calendar adjusted data: Data in which the influence of varying number of working days has been eliminated. The calendar effect in construction data is not detected.
Seasonally adjusted data: Direct method is used for the seasonal adjustment. Model is fixed for the current year. The model is chosen after detailed analysis: 1 regressor (Trading days) are used, outliers are detected and corrected, national holidays calendar is used.
Diagnostic statistics are used to control the quality of seasonal adjustment. The significance levels, the number of outliers and statistics on residuals (residual normality, residual asymmetry, etc.) in data series are controlled.
Revisions: Seasonal adjustment models and parameters are revised and fixed for one year when the data of the fourth quarter is available.
More information about the adjustment is found in added template.
The total constructon volume index is based on the deflated value of construction production. The value of construction production = turnover from construction activity + construction for own use +/- changes in the stock of work in progress - value of work done by subcontractors. The production value of the reference quarter is compared to the production value to the same period of the previous year and the year-on-year percentage change is calculated. The indices at current prices are deflated with construction cost index to obtain the production volume index. Then indices are recalculated to the base 2021 = 100.
Reporting unit - enterprise, observation unit - KAU.
The Statistical Business Register based list of active economic units which are active in the field of construction. There are about 16500 enterprises in the target population.
The activities on the national territory as well as outside the national territory are included. Construction volumes in foreign countries accounted for 7% of the total volume of construction in 2024.
Quarter
The type of survey and data collection methods ensures sufficient coverage and timeliness.
When calculating the indicator estimates, the statistical data analysis and editing of outliers are applied. When estimating the error, the influence of non-response, imputation of missing values and misclassification on the precision of estimates was taken into account.
The sampling error is small. In general the data are not revised after they are first published.
Index
Non- response is treated by imputation with the VAT data, with the data of the previous period or with the average of the stratum. The value index is deflated using a construction cost index to produce a volume index. The production index is a Laspeyres chain index.
The data is based on a survey. Data are collected from all construction enterprises with 20 or more employees. For small enterprises (with 1 to 19 person employed) a stratified simple random sampling is used. There are about 16500 enterprises of which 5% (800) are included in the survey.
Data from statistical activities 20007 "Financial Statistics of Enterprises (quarterly)“, 20300 “Annual Financial Statistics of Enterprises“ and 20401 “Construction Cost Index“ are used.
Quarter
The data are released at the latest 60 days upon the end of the reference quarter (T+60).
