Turkish Statistical Institute (TurkStat)

Short Term Statistics Department.

Turkish Statistical Institute (TURKSTAT) Devlet Mah. Necatibey Cad. No:114 06420 Çankaya / ANKARA

Quarterly Short Term Construction Statistics Survey.

NACE Rev. 2. clasification, CC classification (classification of Types of Construction).

The activity coverage is Section F of NACE Rev. 2 and CC. Construction enterprises with 20 and more employees are included.

List and definition of variables: Production index is produced by adjusting hours worked index by productivity coefficient obtained from construction sector’s gross domestic product at constant prices.

Planned changes in information collected:Studies are going on to change the calculation method of construction production method.

Accounting conventions: Data are recorded for the reference period (quarter).

Reporting unit: Enterprise.

Observation unit(s): Kind of activity unit.

Construction enterprises with 20 and more employees. The number of enterprises surveyed quarterly in 2016 is 8905.

 Geographical area covered: Turkey

Time series for this indicator according to NACE Rev. 2 is available from 2005.

 2010=100

Indices and percentage changes.

Quarter

Legal basis: Statistical Law of TURKSTAT no. 5429, of 10th November 2005 is the legal basis of all statistics.

COUNCIL REGULATION (EC) No 1165/98 of 19 May 1998 concerning short-term statistics.
REGULATION (EC) No 1158/2005 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 6 July 2005 amending Council Regulation (EC) No 1165/98 concerning short-term statistics.

COMMISSION REGULATION (EC) No 1503/2006 of 28 September 2006 implementing and amending Council Regulation (EC) No1165/98 concerning short-term statistics as regards definitions of variables, list of variables and frequency of data compilation

REGULATION (EC) No 1893/2006 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 20 December 2006 establishing the statistical classification of economic activities NACE Revision 2 and amending Council Regulation (EEC) No 3037/90 as well as certain EC Regulations on specific statistical domains.

Obligation on units to provide data: Compulsory surveys. Fines will be given in cases of non-response or deliberate reporting of wrong information according to the Statistical Law no. 5429.

Planned changes in legal basis, obligation to respond and frame used: None

Data are transmitted to Eurostat.

All statistics collected and published by TURKSTAT are governed by the Turkish Statistical Law no. 5429, of 10 November 2005.
Under the terms of the Statistical Law:

1. TURKSTAT has the independence to determine the methodology and release dates of statistics for which it is responsible.
2. Data are collected only for statistical purposes and confidentiality is assured.
3. The confidentiality of individual responses is guaranteed.
4. The TURKSTAT logo is used on all statistics published under the law.
5. The public dissemination of economic statistics is required.

If the number of statistical units in any cell of the data table formed by aggregating the individual data is less than three, or if one or two of the statistical units are dominant even if the number of units is three or more, the data in the concerned cell is considered confidential. Confidential data can be published only when combined with other data so as not to allow any direct or indirect identification. All measures to prevent any illicit access, use or revelation of the confidential data shall be taken by the institutions and organisations producing official statistics.
The confidential data compiled, processed and preserved for the production of official statistics cannot be delivered to any administrative, judiciary or military authority or person, can not be used for purposes other than statistics or as an instrument of proof. Confidential data can be accessed only by persons involved in the production of official statistics, to the extent that they are needed for performing their duties properly.

At the end of each year TURKSTAT publishes a release calendar providing the release dates for the coming year. This is also disseminated on TURKSTAT’s internet website (www.turkstat.gov.tr).

http://www.turkstat.gov.tr/ingtakvim/tkvim.zul#tb1

The data are released to all interested parties by issuing the news release “Construction Turnover and Production Indices”. Simultaneously, the news release is disseminated on the Turkish Statistical Institute (TURKSTAT) website http://www.turkstat.gov.tr.

There is no internal governmental access to data before their release to the public.

Series are transmitted to Eurostat at the same day with the press release. The transmission is done via EDAMIS by using GESMES coding system.

Quarterly.

The results of the indices are published in the news release “Construction Turnover and Production Indices”.

The news release “Construction Turnover and Production Indices” is available on the TURKSTAT web site http://www.turkstat.gov.tr

The index is published at the 2-digit level for the split of building and civil engineering.The following indices are compiled.

Table-1: Seasonally and Calendar Adjusted Turnover Index and Percentage Changes (2010=100)

Table-2: Seasonally and Calendar Adjusted Production Index and Percentage Changes (2010=100)

Table-3: Construction Turnover and Production Indices (2010=100)

The results of the indices are published in the news release “Construction Turnover and Production Indices”.

The news release “Construction Turnover and Production Indices” is available on the TURKSTAT web site http://www.turkstat.gov.tr

Online database is available at http://www.turkstat.gov.tr/PreTabloArama.do?metod=search&araType=vt

Not available.

Series are transmitted to Eurostat at the same day with the press release. The transmission is done via EDAMIS by using GESMES coding system.

Methodology and sources used in statistical compilation are disseminated on the Turkish Statistical Institute (TURKSTAT) website (http://www.turkstat.gov.tr/).

Not available.

The quality of statistics is assessed according to five quality criteria: relevance, accuracy, timeliness and punctuality, accessibility and clarity, coherence and comparability.The quality of statistical information and its production process is ensured by the provisions of the European Statistics Code of Practice.

The quality of the data corresponds requirements of accuracy, timeliness and puctuality, comparability and compatibility, statistical reports used to collect statistical data are tested.

The main users of statistical information are Eurostat, State institutions and agencies, international organisations, the media, research and business communities, students. The users' needs are satisfied without a breach of the confidentiality principle.

Not available.

IPC is compiled for building and for civil engineering, the distinction between building and civil engineering is based on the CC classification. The contents and level of detail of the indicator comply with EU requirements. The series are available from the year 2005.

The field application of Quarterly Construction Sector Questionnaire is carried out by the Regional Offices of TURKSTAT. The data are collected quarterly. The data entry of the questionnaires is performed by the enterprises via a web based computer software. The software for the data entry is designed to check the consistency at several points and not to allow entering wrong or missing data.

After taking IPC results, they are analyzed by using SAS and excel programs together. Imputation is used to compensate the non-response; this is made by means of previous replies of the enterprises.

Data are revised for late questionnaires which cannot be included in index calculation and for revision of the data source due to incorrect reporting. Revision is made also due to the updating of the base year and changes in statistical methods, concepts, definitions and classifications. Production index values are revised, in the case that production data of construction sector gross domestic product are revised.

1. Index values are revised for previous quarter in the current quarter.

2. Index values are revised for each quarter in the last quarter of the year.

First estimations are consistent with the revised indices calculated at the end of the year. The Average Percent Absolute Revision is about 0.003 and  this result is within the limits of confidence.

Data source, and results are regularly assesed and validated.

The sampling method is cut off sampling and sampling error is not calculated for construction statistics.

Enterprises the main activity of those is construction and those have 20 and more employees are taken into account.  The sampling method is cut off sampling. Sampling error is not calculated. Representation ratio of construction sector is good enough for index compilation. There is not coverage problem in the frame. The survey is cumpolsory. Response rate is approximately 99%. Several checks are implemented at micro and macro level. Errors made during data editing are recognized with automatic checks. The imputed value is not within a plausible range, so no imputation error is calculated.

After data enterance, the data is controlled by regional offices stuffs and they contact with the respondents for corrections. Several analyse tables are used for checking the data and the index finally, e.g.terror program is used. If any inconsistences are found the data is verified at micro level again.

Non-responding units are followed up by telephone contacts or visits by field officers.

Item non response is not implemented.

Timeliness: Preliminary results are available on the 95th day after the reference period.

The field application of Quarterly Construction Sector Questionnaire is carried out by the Regional Offices of TURKSTAT. The data are collected quarterly. The data entry of the questionnaires is performed by the enterprises via a web based computer software.  The software for the data entry is designed to check the consistency at several points and not to allow entering wrong or missing data. 

The survey is conducted according to the following schedule.

1. Quarter (January, February, March) field application in April-May
2. Quarter (April, May, June) field application in July-August
3. Quarter (July, August, September) field application in October-November
4. Quarter (October, November, December) field application in January-February

The news releases are published according to the announced release calendar.

The studies are going on to change the calculation method of Construction Production Index to improve timeliness.

Whole Turkey is covered. Statistical information is comparable with EU countries.

There is no break in the series. Indices are comparable over time.

The index of production in construction (IPC) is calculated by using productivity corrected hours worked aproach. In this approach construction production index is obtained by adjusting the hours worked index by the productivity coefficient obtained from the construction sector’s gross domestic product at constant prices.

Comparisons are made with national accounts and made with previous STS-values (previous month and year).

Because one of the data source of construction production index is national accounts, the results show cohorent tendency.

The total for production in construction is consistent with building and civil engineering for gross data. Seasonally and calendar adjusted figures of Production Index has been produced by using direct approach independently from the seasonal adjustment of its sub-series.

The average time spent by respondents to fill statistical report is 15 minutes.

Revision policy

Our institutional revision policy is published under the Revisions part of the MetaInfo on TurkStat official website. Related links are given below:

• http://www.turkstat.gov.tr/UstMenu.do?metod=metabilgiic&prm=3
• http://www.turkstat.gov.tr/UstMenu.do?metod=metabilgiic&prm=1
• http://www.turkstat.gov.tr/UstMenu.do?metod=metabilgiic&prm=2

Types of revisions

• routine revisions (including release/revision calendar),
• major revisions (e.g. base year changes; benchmarking/confrontation with other statistics),
• non-scheduled revisions (error handling and information to users)

Vintage database is available only for in-service using.

Revision calendar

For each statistics, information on revision publication date for routine and major revisions is published via revision information form on the Metadata section of the related statistics. For example; revision information form of Construction Turnover and Production Indices can be accessed from the following link: http://www.turkstat.gov.tr/indir/revizyon/KDIIDB_IHI_ICUE_ENG.pdf

Data are revised for late questionnaires which cannot be included in index calculation and for revision of the data source due to incorrect reporting. Revision is made also due to the updating of the base year and changes in statistical methods, concepts, definitions and classifications. Production index values are revised, in the case that production data of construction sector gross domestic product are revised.

1. Index values are revised for previous quarter in the current quarter.
2. Index values are revised for each quarter in the last quarter of the year.

In seasonal and calendar adjusted data; the model, filters, outliers and calendar variables are re-identified following the publication of the last bulletin of each year. Along the next year, parameters and factors are re-estimated in each bulletin period. The effects of this revision, which is sourced by the re-estimation of seasonality and calendar factors, is reflected on the data of last three years excluding the current year.

Coherence of revision policy: This revision policy is applied to the data both released nationally and transmitted to Eurostat.

The news release indicates that the data are preliminary and are subject to revision. Any major changes in methodology are announced at the time of the change and the data compiled using the revised methodology is released. Very little revisions occur.

Major revision

In the case of major revision, it is announced to the public according to the “Instruction on Methods and Principles Regarding Revisions on Statistical Data Produced by Turkish Statistical Institute” within the scope of the Article 11-1, Article 11-2 and Article 11-3. The link of the instruction is given at: http://www.turkstat.gov.tr/UstMenu/yonetmelikler/Rev_Yonerge_Yayim_27.02.2014_ENG.pdf

Benchmarking

Benchmarking method is not applied to this statistics.

No numeric information exists on the size of typical revisions.

Revision policy is the same for each domain (construction total (110), building (115) and civil engineering (116) production indices.)

Type of source: Statistical survey.

Construction production index is produced by adjusting the hours worked index by the productivity coefficient obtained from the construction sector’s gross domestic product at constant prices.

Frame on which the source is based: Construction enterprises which have 20 and more employees are selected from the business register. At the beginning of every year, the register used is updated in terms of SBS respondent enterprises. Construction enterprises which have less than 20 employees are ignored.

Criteria for stratification: Not relevant.

Threshold values and percentages: Construction enterprises which have 20 and more employees are selected.

Frequency of updating the sample: Every year

Periodicity of data collection: Quarterly

Questionnaires used in the survey: Quarterly Short Term Construction Statistics Survey.

Planned changes in national questionnaires: None

Data collection media: The data entry of the questionnaires is performed by the enterprises via a web based computer software.

Planned changes in data collection methodology: None

Appropriate validation procedures are applied during the whole production process of statistics, and they are in line with rules provided by Eurostat, i.e. starting from selection of respondents to compiling indices and transmission to Eurostat. Data validation system incorporates many tools for data control and imputation.

Estimates for non-response: The variables of the enterprises whose data cannot be obtained due to several reasons are imputed. Previous data of the related enterprise is used for the imputation of the non respondent enterprise.

Estimates for grossing-up to population levels: None

Type of index: Quarterly productivity corrected hours worked approach is used in the index calculation. It is not a chained index. Laspeyres index is used.

Method of weighting and chaining: The construction sector’s gross domestic product is used.

Planned changes in production methods: Studies are going on to change the calculation method of construction production index.

Method Used

Turkstat  carries out the seasonal adjustment of Production Indices, using TRAMO-SEATS methodology based on ARIMA (Autoregressive Integrated Moving Average) model estimation developed by the Banco de Espana and also suggested by Eurostat. The software that is used for the application of this method  is TRAMO-SEATS for Windows (TSW).

Seasonal Adjustment Process

The process of seasonal and calendar adjustment of Production Index begins at the end of each year with determination of the specification of models of the next year.  This specified model structure is kept fixed throughout the year to adjust seasonally and/or calendar effects. At the end of the year, just like the previous year, specification of econometric estimation models for the following year is determined. The identified process repeats itself in a cyclical manner each year.

Total Production Index and subtitles according to Classification of Types of Construction (CC) as building construction and civil engineering works have been seasonally and calendar adjusted. All of the Construction Production Indices contain both seasonal and calendar effects.

Direct or Indirect Approach

Seasonally and calendar adjusted figures of Production Index has been produced by using direct approach independently from the seasonal adjustment of its sub-series. Therefore, no consistency exists between aggregation levels.

Revision Policy

Seasonal adjustment procedure is subject to revisions over time because of the re-estimation of seasonal component as new observations are added. These revisions  are implemented on the data of the last three years excluding the current year.

Publication

Adjusted data has been published in 2 different ways.

•  “Calendar adjusted” data is obtained by subtracting calendar effects including country-specific moving and fixed holidays from raw data.
•  “Seasonal and calendar adjusted” data is obtained by subtracting seasonal and calendar effects from raw data. 

Consistency of Different Aggregation Levels

Series are adjusted by using direct method so consistency does not exist between aggregation levels.

None.