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.
The aim of the survey of construction activity is to provide a results of construction activities by the type of construction. For this purposes is collected the data of the value of construction works, the value of new contract (order), number of employees on the construction site and effective working hours. The collected data enable a monitoring of construction production, overview of the size and structure of the investments in the buildings and engaged the labor force in the construction sector. Data for the variable 110 (index of production in construction) are available from this survey.
3.2. Classification system
NACE Rev.2
Classification of Types of Construction - CC.
3.3. Coverage - sector
Reporting units of survey on construction activities are:
enterprises with 5 and more employees which are registered in sector F – construction (area 41, 42, 43) according to the classification NACE Rev.2 and
enterprises (KAU) which are not registered in sector F, but whose local units perform the construction works.
3.4. Statistical concepts and definitions
Value of construction works - the value of construction works on buildings and civil engineerings which is reporting units perfomed during the reference period with workers who are directly engaged for execution of work. The VAT is excluded as well as the cost for purchase of land, design and supervision.
New contract (orders) - total value of the contracted works in reference period which connected contractors and third party on conducting the construction work, including sub-contractors.
Number of employees - total number of employees which are works in reporting unit (including the owner work, regular partners in the unit and unpaid family workers), as well as the emloyees that are work outside of their unit and who are paid by reporting units. Shall include employees who are absent for a short period (sick leave, paid leave or special leave), as well as employees in the strike and who are absent for an indefinite period. It is also includes the employees with part-time work, seasonal workers, trainees and family workers who are on the payroll.
Effective working hours - total number of hours worked in the production of reporting units during the reference period.
3.5. Statistical unit
Quarterly survey of construction activities include reporting units that are in the Register of Business Entities registered in construction sector as well as KAU.
3.6. Statistical population
In a sample of 2025, of 612 enterprises from the field of construction, 121 enterprises with primary and secondary activity are included.
3.7. Reference area
The whole territory of Montenegro.
3.8. Coverage - Time
Quarterly survey of construction activity is implemented since 1952. During the last years there have been changes and amendments, but the main variables such as the value of construction works, number of employees and effective working hours were constantly provided. The last revisions were made in 2010 by introducing the questions about turnover, the value of spent building materials and spent fuel and electricity. In 2011 is introduced the question for number of employees and effective working hours on the construction site (divided on building and civil engineering's).
3.9. Base period
The base year is 2021.
Indices, percentage changes (%).
Quarter.
6.1. Institutional Mandate - legal acts and other agreements
The legal basis for the implementation of this survey is the Law on Official Statistics and Official Statistical System of Montenegro ("Official Gazette of Montenegro“, No. 18/12, 47/19 and 23/25) and Annual plan of statistical surveys.
Implementation of the construction activity survey in Montenegro also relies on the Regulation for the Integration of Business Statistics REGULATION (EU) 2019/2152 - EBS.
6.2. Institutional Mandate - data sharing
Data are regularly transmitted to Eurostat via eDAMIS.
7.1. Confidentiality - policy
National legislation:
According to the Law on Statistics and Statistical System of Montenegro, Article 54 the data collected, processed and stored for the purpose of official statistics shall be confidential, if it is on the basis of these data, either directly or indirectly identify reporting units which disclosing individual information.
European legislation:
Regulation (EC) No 223/2009 of the European Parliament and of the Council, of 11 March 2009, on the transmission of data subject to statistical confidentiality to the Statistical Office of the European Communities.
Government of Montenegro, on the 64th session, held on February 22, 2018, adopted Commitment on Confidence in Official Statistics which is available on web site.
7.2. Confidentiality - data treatment
Confidential data are treated according to policy rules (see point 7.1).
Data are nationally published only on the level at which they are not confidential.
Individual data represent an official secret.
8.1. Release calendar
Data are published 45 days after the reference period. Date of publication statements is defined in the Calendar of statistics of MONSTAT.
User access is achieved for construction statistics via the MONSTAT website. The data are available for all users at the same time. The data are published in a press release (Excel and Word), in the monthly statistical bulletin (Excel and Word) and as dynamic data (Excel).
Quarterly.
10.1. Dissemination format - News release
Data are available on the website Statistical Office of Montenegro.
10.2. Dissemination format - Publications
Data are published in the quarterly release "Construction activity" and in Monthly Statistical Bulletin.
A statement of Methodology is available on the Statistical Office website.
10.7. Quality management - documentation
Quality report of this survey is available on the web site.
11.1. Quality assurance
MONSTAT has officially accepted the European Statistics Code of Practice. European Statistics Code of Practice in Montenegro is available on the website.
For the systematic and standardized approach to quality, the Quality Management Framework was adopted by MONSTAT. Clear intention is expressed to take further steps towards full compliance with the European Statistics Code of Practice.
11.2. Quality management - assessment
Quality is monitored throughout all stages of data production, by the responsible statisticians, according to specifications for concrete surveys. Data collection, data capture and processing are usually subject to this monitoring (similarly as timeliness and punctuality).
In the framework of the organization of statistical surveys, selected quality aspects are part of preparing staff involved in statistical production.
Users of data are: Eurostat, Ministry of Finance and other Ministry, sectors inside MONSTAT, the companies engaged in construction, other general public.
12.2. Relevance - User Satisfaction
Customer satisfaction survey is not yet available at the level of survey. All users request was fulfilled and based on that, we have concluded that users are satisfied with our survey.
12.3. Completeness
The survey is done on the sample and includes the enterprises that are registered in sector F (according to the Classification NACE Rev.2) and the data from this survey are harmonized with the REGULATION (EU) 2019/2152 EBS.
13.1. Accuracy - overall
The results of quarterly survey of construction activity are based on a sample of reporting units and are subject to the usual types of errors associated with sampling techniques, as well as non-sampling errors, measurement errors, processing errors and non-response. Sampling error for value of construction works was calculated using SAS procedure Proc survey mean for part which was sample based.
13.2. Sampling error
A quarterly survey of construction activity is conducted on a sample basis. Included are all enterprises and KAU with 5 and more employees who are registered in sector F according to the classification NACE Rev.2. The reporting units from class 1 (5-49 employees) are selected through a stratified random sampling method. Reporting units from classes 2 and 3, >50 employees is full coverage. In class 1, there are 85 reporting units, in class 2 are 32, and in class 3 are 4 reporting units. The sampling error of the estimated value of construction activity, calculated as CV, is below 10% per quarter and is precise enough to be published.
13.3. Non-sampling error
Coverage error is the divergence between the frame population and the target population. We didn't make the estimation of the impact of entrepreneurs.
Measurement errors are errors that are occurring during the data collection and caused that the recorded values of variables be different from the true ones.
Processing errors. When the data have been collected, they pass through a range of processes before the final compilation (estimates, etc.) are produced: coding, keying, editing, imputing, weighting, tabulating, etc. Errors introduced at these stages are called processing errors.
Non-response errors occur when the survey fails to get response to one, or possibly to all questions of the survey. Typical response rate is around 75%.
The unit non response rate is calculated as the ratio of the number of units which did not responded to the total number of eligible units (unweighted rate). The item non response rate is calculated as the ratio of the eligible units which have not responded to a particular item and the in-scope units that are required to respond to that particular item.
We perform the imputation on the level of field and on the level of questionnaire. Imputation on the level of field is conducted by the missing variable we take an average of that variable from previous quarter. The non-response on the level of questionnaire solved by imputing the data based on the previous quarter (the average value).
14.1. Timeliness
The data are published 45 days after the end of the reference period.
14.2. Punctuality
Deadlines for publishing have been complied in accordance with the calendar of release.
15.1. Comparability - geographical
Not applicable.
15.2. Comparability - over time
During the last years the ammendments have been done, but the main indicators such as the value of construction works, effective hours worked and number of employees were constantly collected. The data are comparable since 2010.
15.3. Coherence - cross domain
In some cases, data from other areas of economic statistics can be used for cross-checking of data. If the differences between results from survey and other sources are significant, we made double checking of data.
15.4. Coherence - internal
The data are internal coherent. Higher level of aggregations are derived from the detailed indices according to a well-defined procedures.
Financial resources required for the implementation of this survey are defined by the budget of Statistical Office of Montenegro.
17.1. Data revision - policy
Revision Policy prepared by Statistical Office of Montenegro is available on web site.
The revision policy is applied nationally and in transmissions to Eurostat.
17.2. Data revision - practice
Significant revisions have only been made to reflect changes in methodology. We update figures all the time in the new press release with the remarks. The biggest revision happens when we change the base year.
Routine revisions, related to replacing estimated data with the subsequently reported data, are not carried out, because all data are available according to the deadlines.
If an error occurs, corrected data are published in the next issue of First release. Data changed in relation to the first release are disseminated with a comment.
Size of revisions has not been calculated yet.
Use of benchmarking and domain (indicator) specific characteristics of revisions is not available.
For calendar-adjusted data, MAR, RMAR, MR and MaxAR were calculated for TOTAL (F) on year-on-year growth rates for the period 2021-Q1 to 2025-Q4 (n=20), based on the first release and the latest release, separately for routine revisions only and all revisions:
Series
Period
Variant
MAR
RMAR
MR
MaxMAR
W
2021-Q1–2025-Q4
Routine revisions only
2.16
0.40
-0.91
7.65
W
2021-Q1–2025-Q4
All revisions
2.04
0.44
-0.49
5.47
Y
2021-Q1–2025-Q4
Routine revisions only
2.26
0.44
-1.11
7.36
Y
2021-Q1–2025-Q4
All revisions
1.91
0.42
-0.65
6.15
18.1. Source data
Data are collected on a sample of reporting units, including the Statistical Business Register (SBR) in MONSTAT, which are registered according to sector F by NACE Rev. 2 classification. Class 1 (5-49 employees) includes 85 reporting units, Class 2 (50-249 employees) includes 32, and Class 3 (>250 employees) includes 4 reporting units.
18.2. Frequency of data collection
Quarterly.
18.3. Data collection
Data are collected through questionnaires (by post, email).
18.4. Data validation
In order to improve the quality of data we check that the coverage and response are as required; checking the illogicalities (large increase or decrease in value of construction works, a large change in the new contracts compared to the previous quarter) and ect. We contact the reporting units if it’s necessary.
18.5. Data compilation
The data are collected on a sample of reporting units whose results represent a quarterly estimation of construction activities in Montenegro. Horvitz-Thompson estimation of totals is calculated according to the standard procedure for a stratified sample with random selection of units within a stratum. The estimation of totals is calculated by summing the weighted values from the sample, where the weight for units from the same stratum is equal to the quotient of the number of units in the stratum and the number of units in the sample. The original weights are adjusted for non-response. Allocation is done with the Neyman algorithm.
18.6. Adjustment
The data is working day and seasonally adjusted.
More information about the seasonal adjustment please sees in Annex.
The aim of the survey of construction activity is to provide a results of construction activities by the type of construction. For this purposes is collected the data of the value of construction works, the value of new contract (order), number of employees on the construction site and effective working hours. The collected data enable a monitoring of construction production, overview of the size and structure of the investments in the buildings and engaged the labor force in the construction sector. Data for the variable 110 (index of production in construction) are available from this survey.
4 June 2026
Value of construction works - the value of construction works on buildings and civil engineerings which is reporting units perfomed during the reference period with workers who are directly engaged for execution of work. The VAT is excluded as well as the cost for purchase of land, design and supervision.
New contract (orders) - total value of the contracted works in reference period which connected contractors and third party on conducting the construction work, including sub-contractors.
Number of employees - total number of employees which are works in reporting unit (including the owner work, regular partners in the unit and unpaid family workers), as well as the emloyees that are work outside of their unit and who are paid by reporting units. Shall include employees who are absent for a short period (sick leave, paid leave or special leave), as well as employees in the strike and who are absent for an indefinite period. It is also includes the employees with part-time work, seasonal workers, trainees and family workers who are on the payroll.
Effective working hours - total number of hours worked in the production of reporting units during the reference period.
Quarterly survey of construction activities include reporting units that are in the Register of Business Entities registered in construction sector as well as KAU.
In a sample of 2025, of 612 enterprises from the field of construction, 121 enterprises with primary and secondary activity are included.
The whole territory of Montenegro.
Quarter.
The results of quarterly survey of construction activity are based on a sample of reporting units and are subject to the usual types of errors associated with sampling techniques, as well as non-sampling errors, measurement errors, processing errors and non-response. Sampling error for value of construction works was calculated using SAS procedure Proc survey mean for part which was sample based.
Indices, percentage changes (%).
The data are collected on a sample of reporting units whose results represent a quarterly estimation of construction activities in Montenegro. Horvitz-Thompson estimation of totals is calculated according to the standard procedure for a stratified sample with random selection of units within a stratum. The estimation of totals is calculated by summing the weighted values from the sample, where the weight for units from the same stratum is equal to the quotient of the number of units in the stratum and the number of units in the sample. The original weights are adjusted for non-response. Allocation is done with the Neyman algorithm.
Data are collected on a sample of reporting units, including the Statistical Business Register (SBR) in MONSTAT, which are registered according to sector F by NACE Rev. 2 classification. Class 1 (5-49 employees) includes 85 reporting units, Class 2 (50-249 employees) includes 32, and Class 3 (>250 employees) includes 4 reporting units.
Quarterly.
The data are published 45 days after the end of the reference period.
Not applicable.
During the last years the ammendments have been done, but the main indicators such as the value of construction works, effective hours worked and number of employees were constantly collected. The data are comparable since 2010.
