Average (mean) gross weekly earnings by Welsh local areas and year (£)
None

Metadata
 High level information
 Summary information
 Statistical quality information
 Weblinks
 Keywords
 Open Data
Title
Average earnings data by Welsh local areasLast update
26 October 2016Next update
October 2017Publishing organisation
Welsh GovernmentSource 1
Annual Survey of Hours and Earnings, Office for National StatisticsContact email
economic.stats@wales.gsi.gov.ukDesignation
National StatisticsLowest level of geographical disaggregation
Local authoritiesGeographical coverage
WalesLanguages covered
English and WelshData licensing
You may use and reuse this data free of charge in any format or medium, under the terms of the Open Government License  see http://www.nationalarchives.gov.uk/doc/opengovernmentlicenceGeneral description
These data show average gross weekly earnings in pounds for the local authorities and aggregations thereof in April of the years shown. The data relate to fulltime employees on adult rates whose pay for the survey period was not affected by absence.The local authority aggregations are not yet available as we did not have access to these figures in advance of the publication. These will be updated as soon as possible.
Data collection and calculation
The figures are taken from the Annual Survey of Hours and Earnings (ASHE) run by the Office for National Statistics (ONS). In 2004, the ASHE replaced the New Earnings Survey (NES) by introducing a new methodology into the calculation of earnings data. This new methodology applies weights to the results to take account of the structure of the population in terms of age, gender and occupation and area of workplace (the latter being in London and the South East or elsewhere in the UK). The NES data for 1997 to 2003 have been reworked to provide a backseries of earnings data using the new methodology.There were further changes to the ASHE methodology in 2005 as a result of the introduction of a new questionnaire. 2004 data have been reworked to be comparable with this new methodology, but it has not been possible to do this for earlier years. Thus there is a discontinuity in the data that must be taken account of when making comparisons in earnings data over time. To help with this, there are two estimates for 2004, one on the previous basis (referred to as excluding supplementary survey information) and one on the new basis (referred to as including supplementary survey information).
In 2007 and 2008, there was a sample reduction of around 20 per cent. The sample reduction has been designed to be biggest in those industries where earnings exhibit lower levels of variation.
A new automatic coding system for occupations was introduced in 2007. The main impact of this has been to move a number of jobs away from the top occupational groups to other occupational groups. This has tended to lower the average earnings in the top occupational groups and to lower earnings overall. Partly in response to the change to the sample design, an additional weighting stratum has been introduced for those large enterprises which submit electronic returns to the survey (special arrangements). There has been no reduction in the sample amongst these enterprises. The ONS has produced a version of the 2006 ASHE results which includes the automatic occupational coding change and the special arrangements weighting stratum. This enables yearonyear comparisons to be made which take account of these two changes. These are referred to as 2006 (consistent with 2005) and 2006 (consistent with 2007).
In 2009 the original sample size was reinstated.
Please note that due to small sample sizes it is also not advisable to make yearonyear comparisons at a local authority level. To get around this, a set of aggregations of local authorities has been created, for which comparisons over time can be considered more robust. These aggregations group similar local authorities together and are wholly enclosed within the NUTS2 boundaries in Wales that are used to determine European Funding allocations. Figures for the two NUTS2 areas (West Wales and the Valleys and East Wales) are also included.
As the results come from a survey, the results are samplebased estimates and therefore subject to differing degrees of sampling variability, i.e. the true value for any measure lies in a differing range about the estimated value. This range or sampling variability increases as the detail in the data increases, for example regional data are subject to higher variability than the Great Britain or United Kingdom data.
Where the estimate for any cell in this dataset is not considered to be precise (due to excessive sampling variability), a cellnote is included against the estimate indicating the quality rating as either reasonably precise [marked by (!)] or acceptable [marked by (!!)]. Where such a cellnote exists, particularly the latter, or in cases where the variability exceeds acceptable limits and is suppressed, it is recommended that the user moves up to the next level in the hierarchical aggregation of areas if using the figure in any analysis, particularly if considering change over time.
Frequency of publication
AnnualData reference periods
1997 to 2016Revisions information
Data for the latest year are provisional and are revised on the release on the next years data.Statistical quality
The figures are taken from the Annual Survey of Hours and Earnings (ASHE) run by the Office for National Statistics (ONS). In 2004, the ASHE replaced the New Earnings Survey (NES) by introducing a new methodology into the calculation of earnings data. This new methodology applies weights to the results to take account of the structure of the population in terms of age, gender and occupation and area of workplace (the latter being in London and the South East or elsewhere in the UK). The NES data for 1997 to 2003 have been reworked to provide a backseries of earnings data using the new methodology.There were further changes to the ASHE methodology in 2005 as a result of the introduction of a new questionnaire. 2004 data have been reworked to be comparable with this new methodology, but it has not been possible to do this for earlier years. Thus there is a discontinuity in the data that must be taken account of when making comparisons in earnings data over time. To help with this, there are two estimates for 2004, one on the previous basis (referred to as excluding supplementary survey information) and one on the new basis (referred to as including supplementary survey information).
In 2007 and 2008, there was a sample reduction of around 20 per cent. The sample reduction has been designed to be biggest in those industries where earnings exhibit lower levels of variation.
A new automatic coding system for occupations was introduced in 2007. The main impact of this has been to move a number of jobs away from the top occupational groups to other occupational groups. This has tended to lower the average earnings in the top occupational groups and to lower earnings overall. Partly in response to the change to the sample design, an additional weighting stratum has been introduced for those large enterprises which submit electronic returns to the survey (special arrangements). There has been no reduction in the sample amongst these enterprises. The ONS has produced a version of the 2006 ASHE results which includes the automatic occupational coding change and the special arrangements weighting stratum. This enables yearonyear comparisons to be made which take account of these two changes. These are referred to as 2006 (consistent with 2005) and 2006 (consistent with 2007).
In 2009 the original sample size was reinstated. Please note that due to small sample sizes it is also not advisable to make yearonyear comparisons at a local authority level. To get around this, a set of aggregations of local authorities has been created, for which comparisons over time can be considered more robust. These aggregations group similar local authorities together and are wholly enclosed within the NUTS2 boundaries in Wales that are used to determine European Funding allocations. Figures for the two NUTS2 areas (West Wales and the Valleys and East Wales) are also included.
As the results come from a survey, the results are samplebased estimates and therefore subject to differing degrees of sampling variability, i.e. the true value for any measure lies in a differing range about the estimated value. This range or sampling variability increases as the detail in the data increases, for example regional data are subject to higher variability than the Great Britain or United Kingdom data.
Where the estimate for any cell in this dataset is not considered to be precise (due to excessive sampling variability), a cellnote is included against the estimate indicating the quality rating as either reasonably precise [marked by (!)] or acceptable [marked by (!!)]. Where such a cellnote exists, particularly the latter, or in cases where the variability exceeds acceptable limits and is suppressed, it is recommended that the user moves up to the next level in the hierarchical aggregation of areas if using the figure in any analysis, particularly if considering change over time.