|
|
For any question on data and metadata, please contact: EUROPEAN STATISTICAL DATA SUPPORT |
|
|||
1.1. Contact organisation | Turkish Statistical Institute |
||
1.2. Contact organisation unit | Department of Annual Business Statistics Annual Services Statistics Group |
||
1.5. Contact mail address | Devlet Mah.Necatibey Cad. No:114 Room: 418 06650 Çankaya/ANKARA TURKEY |
|
|||
2.1. Data description | |||
The quality report prepared for the assessment of the Labour Cost Survey (LCS) 2012 comprises information pertaining to the six components of the European Statistical System. These are relevance, accuracy, timeliness and punctuality, accessibility and clarity, comparability and coherence. The first survey carried out by TurkStat dedicated to the labour cost was the Labour Cost Survey 2004. The second and third surveys were implemented 2008 and 2012 respectively. The LCS 2012 was started to be implemented in January 2013. The survey was structured as a four yearly survey to provide information in accordance with the Regulation No. 530/1999 of the European Council and Parliament and Regulation No. 1726/1999 and 1737/2005 of the European Commission. It has been the third survey carried out in line with EU requirements on the subject. The objective of this report is to assess the quality of the survey. The report will follow the contents of the Commission Regulation No. 698/2006 concerning evaluation of structural statistics on labour costs and earnings.
2012 LCS questionnaire has four main parts;
|
|||
2.2. Classification system | |||
Not available. |
|||
2.3. Coverage - sector | |||
Not available. |
|||
2.4. Statistical concepts and definitions | |||
Not available. |
|||
2.5. Statistical unit | |||
Not available. |
|||
2.6. Statistical population | |||
Not available. |
|||
2.7. Reference area | |||
Not available. |
|||
2.8. Coverage - Time | |||
Not available. |
|||
2.9. Base period | |||
Not available. |
|
|||
- |
|||
3.1. Source data | |||
The sample size of the local units were 22 883. The respondent local units were 17 806. No data from registers have been employed.
The distribution of the sample size was realized by the Compromise Allocation Method on the level of statistical region units and economic activity of local unit and the size class of the enterprises. Simple random stratified sampling without replacement was used in selection of local units. The size classes and the economic activity groups covered in the survey are as follows:
|
|||
3.2. Frequency of data collection | |||
[Not requested] |
|||
3.3. Data collection | |||
[Not requested] |
|||
3.4. Data validation | |||
[Not requested] |
|||
3.5. Data compilation | |||
[Not requested] |
|||
3.6. Adjustment | |||
[Not requested] |
|
|||
4.1. Quality assurance | |||
Not available. |
|||
4.2. Quality management - assessment | |||
[Not requested] |
|
|||
- |
|||
5.1. Relevance - User Needs | |||
The main international users of the survey presumed to be Eurostat, other EU institutions and international institutions such as OECD, ILO and IMF. Internal users may be conjectured as the researchers, employers’ associations, trade unions and governmental units such as Ministry of Labour and Social Security, Ministry of Industry and Trade, Central Bank of Turkey and media. |
|||
5.2. Relevance - User Satisfaction | |||
The satisfaction level of the users and further data needs will be obtained as the feedbacks and comments from the users are reached. |
|||
5.3. Completeness | |||
All mandatory variables were obtained. Hence the variable is calculated by using additional variables added to the questionnaire with this objective. |
|||
5.3.1. Data completeness - rate | |||
[Not requested] |
|
|||
- |
|||
6.1. Accuracy - overall | |||
[Not requested] |
|||
6.2. Sampling error | |||
The coefficient of variations for the variables mentioned in the Commission Regulation 698/2006 is given in the tables in annex. |
|||
6.2.1. Sampling error - indicators | |||
The coefficient of variations for the variables mentioned in the Commission Regulation 698/2006 is given in the tables in annex. Annexes: Coefficient of variation for "Annual and Hourly Labour Cost" by Economic Activity Section Coefficient of Variation for "Annual and Hourly Labour Cost" by Size Band of the Enterprise Coefficient of Variation for "Annual and Hourly Labour Cost" by NUTS1 Level |
|||
6.3. Non-sampling error | |||
The overall non-response rate of the survey is 6.0 %. The distribution of the total non-responded cases could be listed as follows; erroneous address (93.69 %), refusals (2.25%) and other reasons (4.06%). The highest rate of non-response comes from the errors in the addresses frame. Other reasons pertain to inability to contact due mainly to seasonality and inadequate address. |
|||
6.3.1. Coverage error | |||
The local units associated with the enterprises with 10 or more employees in the 2011 Business Register were used as the sampling framework in the LCS 2012. The business register is updated annually on the basis of information taken from Ministry of Finance and annual structural business survey results for the level of enterprise.
The survey questionnaire asks for information on the current economic activity, the region and actual number of employees in the local unit. The results were checked with the frame information. Misclassification in terms of economic activity was observed in the survey. The overall misclassification rate was 33.9 %.
It is not possible to obtain under coverage, since there is no other external source to compare the frame information. |
|||
6.3.1.1. Over-coverage - rate | |||
Regarding to over coverage, 16.2 % of the sample was out of scope. The distribution of over coverage by reasons is as follows;
Other reasons could be inactive local units, size of the enterprise is out of scope etc. |
|||
6.3.1.2. Common units - proportion | |||
[Not requested] |
|||
6.3.2. Measurement error | |||
The collection process of the questionnaire is performed by the enterprises via a web based software tool. The software for the data entry is designed to check the consistency at several points and not to allow entering wrong or missing data. Item non response was not allowed. Therefore imputation techniques were not used. The checks embedded in data entry code mainly focused on;
|
|||
6.3.3. Non response error | |||
The overall non-response rate of the survey is 6.0 %. The distribution of the total non-responded cases could be listed as follows; erroneous address (93.69 %), refusals (2.25%) and other reasons (4.06%). The highest rate of non-response comes from the errors in the addresses frame. Other reasons pertain to inability to contact due mainly to seasonality and inadequate address. Annexes: Non-response rates by Economic Activity Sections Non-Response Rates by Size Band of The Enterprise |
|||
6.3.3.1. Unit non-response - rate | |||
[Not requested] |
|||
6.3.3.2. Item non-response - rate | |||
[Not requested] |
|||
6.3.4. Processing error | |||
As the data were controlled from the regional offices, data were either validated or sent back to regional offices after the application of second level checks on all variables of questionnaire. Correction or approval from the establishment was required for non validated cases. The second level checks applied in the centre cover the ones embedded in data entry code with more precise lover and upper levels and also cover checks on;
The information concerning the total annual hours actually worked have not been the one for which the enterprises documented well. Hence, additional questions were directed to calculate the figure pertaining to the variable. Average daily actual working hours, average weekly normal working hours, total annual (paid and unpaid) overtime hours, number of days actually not worked by the type of absence were the additional questions of the survey used for this purpose. |
|||
6.3.4.1. Imputation - rate | |||
No imputation were apllied. |
|||
6.3.5. Model assumption error | |||
No model were used in the LCS 2012. |
|||
6.4. Seasonal adjustment | |||
[Not requested] |
|||
6.5. Data revision - policy | |||
[Not requested] [Not requested] |
|||
6.6. Data revision - practice | |||
[Not requested] |
|||
6.6.1. Data revision - average size | |||
[Not requested] |
|
|||
- |
|||
7.1. Timeliness | |||
The results of the LCS 2012 were released on 30 December 2013. The delivery of data to Eurostat was in February 2014 that is before the deadline foreseen in the requirements of regulations and arrangement, 18 months after the end of the reference period. |
|||
7.1.1. Time lag - first result | |||
[Not requested] |
|||
7.1.2. Time lag - final result | |||
[Not requested] |
|||
7.2. Punctuality | |||
The timetable applied throughout the implementation of LCS 2012 was as follows;
|
|||
7.2.1. Punctuality - delivery and publication | |||
[Not requested] |
|
||||||
- - |
||||||
8.1. Comparability - geographical | ||||||
The LCS 2012 is applied in the local units associated with enterprises employing 10 and more employees as set in the relevant regulations/arrangements.
The compilation of the data was based on local units and enterprises as defined in the Council Regulation no. 696/93. The units were stratified by size of the enterprise, NACE and statistical region units.
Therefore data conform to regulations and are comparable with other countries' data. |
||||||
8.1.1. Asymmetry for mirror flow statistics - coefficient | ||||||
[Not requested] |
||||||
8.2. Comparability - over time | ||||||
The difference between LCS 2012 and the previous surveys is that the statistics is published on the new industry classification standard only. TurkStat used NACE Rev.1.1 as an industry classification for surveys of LCS 2004 and 2008. LCS 2012 is published according to NACE Rev. 2 and cannot be compared directly with the 2008 and 2004 publications. |
||||||
8.2.1. Length of comparable time series | ||||||
[Not requested] |
||||||
8.3. Coherence - cross domain | ||||||
1. Coherence with Labour Force Survey The Figure 1 in annex compares the hours actually worked expressed in NACE Rev.2 sections in the Labour Force Survey (LFS) and the Labour Cost Survey in 2012. LFS data is calculated for employees work in establishments with 10 and more employees to ensure the comparability with LCS. The average annual actual hours for LFS were calculated using ‘weekly actual working hours’ estimation of LFS with the assumption of two weeks of annual vacation. The results of the LCS and the LFS were compared in terms of the values of “annual hours actually worked per employee”. The discrepancy between LCS and LFS in the values pertaining to the hours actually worked may stem from differences in the methodology used in these two surveys. The most important difference comes from the fact that there is an unregistered employee fact in the country which is reflected in LFS but not in LCS.
2. Coherence with Structural Business Survey The Figure 2 in annex, compares the wages and salaries expressed in NACE Rev.2 sections in the Labour Cost Survey and the Structural Business Survey (SBS) in 2012. It is observed that wages and salaries were found to be higher in the LCS than that in the SBS in general. The differences between the LCS and the SBS still owes to the differences between the two surveys in terms of the methodology. Two main methodological differences might be cited as size coverage and observation unit. While enterprises with less than 10 employees are excluded from the LCS where the observation unit is local unit, all sizes are covered in SBS whose observation unit is the enterprise. [Not requested] Annexes: Figure 1 Actual Hours Worked in LFS vs LCS Figure 2 Wages and Salaries per Employee in SBS vs LCS |
||||||
8.4. Coherence - sub annual and annual statistics | ||||||
[Not requested] |
||||||
8.5. Coherence - National Accounts | ||||||
The average annual growth rate of the variable ‘hourly labour cost’ could not be compared with the average annual growth rate of the LCI at the detail of NACE breakdown because of using different industry classification of LCS 2008 and LCS 2012 surveys. However, overall comparison is given below (Table). It is also not possible to compare the variables requested in the Regulation with National Accounts results, because National Accounts data is not available for the period beginning from 2006 due to ESA 95 revising studies.
Table: Average Annual Growth Rates of Hourly Labour Costs: LCI vs LCS
|
||||||
8.6. Coherence - internal | ||||||
[Not requested] |
|
|||
- |
|||
9.1. Dissemination format - News release | |||
The press release for LCS 2012 may be reached from http://www.turkstat.gov.tr in Turkish and English. |
|||
9.2. Dissemination format - Publications | |||
The bilingual publication containing estimations in table and graphic format and methodological information was released both in printed and electronic formats. Additional information is provided on request to the users in table format. The survey results are provided to respondents only on request. |
|||
9.3. Dissemination format - online database | |||
The selected tables and metadata from the publication is also available at TurkStat’s home page in Turkish and English. |
|||
9.3.1. Data tables - consultations | |||
[Not requested] |
|||
9.4. Dissemination format - microdata access | |||
[Not requested] |
|||
9.5. Dissemination format - other | |||
- |
|||
9.6. Documentation on methodology | |||
The metadata pertaining to the LCS 2012 is available in the publication and is available at http://www.turkstat.gov.tr. |
|||
9.7. Quality management - documentation | |||
[Not requested] |
|||
9.7.1. Metadata completeness - rate | |||
[Not requested] |
|||
9.7.2. Metadata - consultations | |||
[Not requested] |
|
|||
[Not requested] |
|
|||
- |
|||
11.1. Confidentiality - policy | |||
[Not requested] |
|||
11.2. Confidentiality - data treatment | |||
[Not requested] |
|
|||
- |
|
|||