Statistical processing

Source data

The WTA are based on a combination of census and survey data. The WTA are compiled on the basis of three primary data sources:

1) The Register of Employment Statistics forming the basis for both:

• the Register-based labour force statistics RAS statistics
• and for the Establishment-related Employment Statistics ERE statistics

The WTA use the Register of Employment Statistics for obtaining data on jobs and persons employed at end-November as well as annual data on aggregate payroll costs (compensation of employees) and annual paid hours of work for employees.

The structural data incorporated in the 2013 WTA, which are made up by the Register of Employment Statistics (the basis for the register-based labour force statistics, RAS and the Establishment-related Employment Statistics at the end of November 2013), are based on a special version exclusively for internal use in the Working Time Accounts, National Accounts and Industrial Accounts Statistics, where the former methods and sources are applied. This internal version is used for the purpose of avoiding breaks in the statistics mentioned.

Employment Statistics for Employees BfL

The Employment Statistics for Employees contain monthly data on jobs, paid hours of work and total wage and salary costs relating to employees throughout the year. The data are used in the WTA for projecting compensation of employees, hours worked, employment, primary and side line jobs for employees during the year.

The three above-mentioned statistics are compiled on the basis of eIncome:

2) The Structural Earning Statistics

The Statistics on Earnings are used in the WTA for converting paid hours of work into hours worked by employees during the year. Where the above-mentioned structural statistics set the level for the statistics in the WTA, the short-term statistics are used for describing the development throughout the year.

3) The Labour Force Survey LFS

The Labour Force Survey is used for describing the development in the number of hours worked during the year.

Furthermore, the Labour Force Survey is also used for measuring the effect derived from each day of absence from work during the Easter holiday on the distribution of hours worked between the months March and April. Subsequently, the effect derived from the Easter holiday is calculated by counting the number of Easter days of absence, falling in each of the two months during each year.

The Labour Force Survey is also applied in undertaking projections of employment and jobs for self-employed and assisting spouses during the period, following the latest November-statistics of the Register of Employment Statistics. Finally, the LFS is applied in describing how many more hours of work were performed by self-employed and assisting spouses compared to hours worked by employees.

For self-employed and assisting spouses, the development in employment and jobs is calculated as a steady development from one structural statistics to another (employed persons in the Register-based Labour Force Statistics and the number of jobs in the Establishment-related Employment Statistics). However, rolling annual statistics from the Labour Force Survey are used for projections in the period following the latest structural statistics (i.e. after the end of November 2011).

The basis for calculating hours worked by self-employed persons and assisting spouses are hours worked per job for employees. These hours are enumerated by the number of jobs for self-employed persons and assisting spouses and adjustments are made on the basis of the number of more hours worked by self-employed persons and assisting spouses compared to hours worked by employees according to the Labour Force Survey.

Average employment (and average number of jobs) over the year is estimated as an average figure of average employment during the 4 quarters of the year (respectively average number of jobs of 4 quarters). Against the background of the projections, it is possible to compile preliminary annual statistics for the period following the latest structural statistics.

In deciding which data sources to apply in compiling the WTA, attention is centred on the major advantages provided by each individual statistics. For example, register-based data are used to ensure complete coverage in the calculation of employment, number of jobs, aggregate payroll costs and paid hours of work. Register-based short-term statistics are used for describing the development throughout the year in the same variables. Information from the wage and salary system of the business enterprises is used to convert paid hours of work into hours worked during the year. Personal interviews are used to obtain information on the distribution of hours worked during the year as well as information on the groups that are not covered by the registers.

The Working Time Accounts are exclusively based on existing data sources, which are subsequently converted to the concepts used in the WTA. The WTA is flexible in its choice of primary sources, which can be replaced by other sources, if these have proved to be more accurate. The choice of primary source decides the amount of data editing necessary. When it comes to integrating all the sources, however, all the concepts are consistent in conforming to international standards and every variable fulfils the requirement of the system for the WTA.

Frequency of data collection

Is not relevant to this statistic as the WTA only make use of already existing statistics (i.e. no new reports).

Data collection

Is not relevant to this statistic as the WTA only make use of already existing statistics (i.e. no new reports).

Data validation

Data are already checked for errors in the primary statistics. In the WTA further checks, troubleshooting and debugging are carried out. This is partly based on the information from the producers of the input sources, partly systematic (mostly figurative) controls the internal consistency between variables and over time, and by comparisons with other published statistics. Finally developments are systematically being discussed with stakeholders from other statistics, including in the context of short term statistics interest groups (where data is typically assessed on the level of News from Statistics Denmark, which for the quarterly statistics will say aggregated to 10 industrial groups and 2 sector groups). When developments look strange, then the reason behind this is being analysed, with possible assistance from producers of input statistics, and if errors are detected then corrections are included in the system. In some cases, the statements lead to corrections in the primary statistics.

Each series are examined quarterly by evaluating the series over time (monthly from 2008 onwards) level:

• employees: 127-industrial grouping * 2 sector groups
• self-employed and assisting spouses: 36 industrial groupings

In connection with the annual estimates error detection is carried out also by gender and full time / part time.

Calculations to illustrate the internal consistency in the WTA (assessed accordingly over time at the above level):

• Hours worked per employee
• Hours worked per employed
• Hourly wage (compensation of employees / hours worked for employees)
• Wage per employee (compensation of employees / number of employed workers)

Comparisons with published statistics:

• National Accounts (employment, hours and compensation of employees)
• Employment of Employees (full-time jobs and compensation of employees)
• Public employment statistics
• Indices of average (hourly) earnings in the private and public sector
• Labour Force Surveys (employment and development in hours worked)
• Number of persons employed in the construction industry

It is possible in the system for the Working Time Account to enter corrections in the input data sources as well as in the output data sources of the Working Time Account before they are disseminated. It is also possible to make feedback to those responsible for the input data sources.

Data compilation

WTA is a statistic that is based on the integration of several input data sources / existing statistics. In the WTA a lot of development is continuously taken place in the form of revisions due to new data input sources, data breaks in existing input sources, revisions in line of industry codes, new sector codes, new or changing needs of users (national and international), need for projecting data back in time to periods of the current system does not cover, etc.

A special workflow that does not fall on a regular basis throughout the year, are descriptions of the method, answering various questionnaires (about strategy, comparability to other areas of statistical methodology development, etc.), updating of publishing calendars, formation of specific data to the STS (e.g. weights, changes to population coverage and groups of STS transmissions, changes to seasonal adjustment, etc.). This workflow can involve work and coordination between employees in different offices or departments. These tasks may well just start with an external query and end after analyses and response. For more comprehensive changes may be needed for explanatory analyses and preparation of memoranda for internal use, to Eurostat or to the website. This will typically happen in the incorporation of new annual structural data, and major revisions (where data is typically revised throughout the time series length). If the elucidations results in recommendations to revise (all or part of) the WTA begins a new project. In the following description, the above mentioned operations are not described.

Overview: Structural data constitutes the basis of the variables conceptual boundaries and levels. Thus, the WTA use annual data at the micro level (job level) from the Register of Employment Statistics. Earnings statistics are in the WTA used for converting paid hours of work to hours worked. Short-term statistics is used for distributing structural annual data over the months. Thus, employment from the Register-based labour force statistics (RAS statistics) and jobs and aggregate payroll costs (compensation of employees) and annual paid hours of work for employees from the Establishment-related Employment Statistics (ERE statistics) is distributed to a monthly basis with the help of BFL's monthly data. Employment Statistics for Employees (BfL) are also used to predict job, employment, compensation of employees and hours worked by employees in the period after the latest available annual structural data. The LFS is used to describe the distribution of hours worked during the year. Furthermore, the LFS is used for projecting the number of employed and jobs for self-employed and assisting spouses in the period after the latest annual structural data. Finally, the LFS is used for describing how many hours self-employed and assisting spouses work compared to employees.

More detailed description of work:

The necessary data sources are loaded.

Data is aggregated, integrated and projected: BFL and the Structural Earning Statistics are used on micro-level, in a preliminary step: As data from Structural Earning Statistics is imported, more updated workplace information from Employment Statistics for Employees (BfL) is transferred to data from the Structural Earning Statistics (including information on line of industry and sectors). Data from different sources are aggregated (to the extent this have not already be done so when loading the data). Any previous corrections to the input data (BFL or RAS / ERE) is included. This is followed by a calculation of factors for the number of paid hours actually worked in relation to the number of hours paid per years. These factors are transferred to relatively detailed structural data from the Register of Employment Statistics. Factors are calculated (based on BFL) to distribute annual data. Further factors are computed (based on LFS) to describe the distribution of hours worked during the year. Enumeration factors calculated for BFL level to the level of The Register of Employment Statistics end of November statements (RAS, ERE Statistics) for variables paid hours of work, main job and side line jobs and salary for employees. The calculated projection factors (and in the initiation of the WTA also factors to project from 2009 to 2008), which describes the development of jobs, employment, compensation of employees, hours worked for employees over the year. Data from different sources are integrated and projected.

Aggregated publishing data are formed: The most current version of the structural data from The Register of Employment Statistics is found. Non-corrected data for jobs, hours worked and compensation of employees are gathered in a single table. Non-corrected data is integrated with corrected data. Only data for the latest version of the reference period are included. Data is enriched with various aggregate line of industry (Dst's Std-groups) and it is ensured that the variables and formats match names and formats used for publishing purposes and troubleshooting. The WTA is broken up on detailed sectors by means of end of November statements (Register-based labour force statistics (RAS statistics) / the Establishment-related Employment Statistics (ERE statistics)). Based on these data (not least for the national accounts industry aggregates) data sets are generated to the National Accounts at various levels of aggregation.

Data are seasonally adjusted, debugged and corrected: Initially an overall quality checks are implemented to see if something is wrong, just to be sure there is not a need for new input data deliveries. If necessary, the process stops here and the responsible person for the input data is contacted to find out what the problem is about / possibly to secure new supplies, and it is agreed, if necessary, who makes the adjustments and how. Because Easter can have very serious consequences for the development of hours worked, data are seasonally adjusted. Actual data, trading day adjusted data and seasonally adjusted data are also formed on Business Short Regulation (STS) groups so that these may also be used for troubleshooting. Actual and seasonally adjusted data are error detected in detail, mainly due by means of graphics. In addition, consistency checks are carried out (between variables, over time and in comparison to other statistics). This also takes place primarily based on graphics. Where deviations look strange any corrections are calculated for input sources and/or output data. It is checked whether the cause is an industry shift. Responsible of the input statistics are contacted in order to carry out analysis. If it turns out that there is a need for corrections, these can be incorporated to the various input sources, or they can be incorporated to the WTA output data. Cyclical interest group meetings are held. This may reveal the need for additional troubleshooting and corrections.

Data are analysed and disseminated: The story in the data are found for the most recent data. If necessary notes on the subject site will be updated. Special features are described for use in eg News from Statistics Denmark and Documentation of statistics. Documentation of statistics in Danish and English are updated. Danish and English figures, tables, stories and explanations in News from Statistics Denmark, Statistical News, Yearbook, STS, etc. are updated and proofread. Various data deliveries to Macro Database, various series for the STS, deliveries to internal users (including National Accounts and various stakeholders in relation to the short term statistics forum) as well as deliveries to external users are formed, controlled at an overall level and delivered. Answers to questions from the press, questions from the national accounts, Eurostat, Labour Movement's Business Council (Arbejderbevægelsens Erhvervsråd, AE), Danish Industry (DI), Danish employers' association (DA), the trade unions (LO), the Productivity Commission, ministries, or just one-off requests from individuals or organizations / companies. Any press contacts are reported.

Data are archived.

Adjustment

No corrections of the data are carried out beyond what has already been described during data validation, data processing and seasonal adjustment.