“ESS” surveys are annual surveys, constituting the first part of the four-yearly cycle of surveys called “Ecmoss” (“Survey on Labour Costs and Wage Structure”, “Enquête sur le Coût de la Main d’Oeuvre et la Structure des Salaires” in French). Those surveys cover establishments and employees in the private sector, the hospital civil service and the local civil service.

Annual

The Ecmoss surveys are employer surveys. Data is collected from April to October for each year of collection (2026 for the ESS 2025).

Establishments can answer by paper questionnaire, computer file (for large companies) or by internet via the “Platine” portal (for establishments with no more than 5 employees surveyed).

Establishments receive an Establishment questionnaire and Employee questionnaires about one or more of their employees who are specifically named (between 1 and 24). It’s therefore employers who respond to the Employee questionnaires.

The ESS 2025 collection take place from April to October 2026.

Local unit (of an enterprise)

The sample is selected using a two-stage design, stratified at each stage. First, establishments are drawn and then employees within these establishments. The stratification used is designed to optimise the accuracy of the main indicator (hourly wages) according to the main breakdowns required by law (by sector of activity, company size, region in particular).

The sampling rate is approximately 3.2% for the establishment level and 0.9% for employees.

The Ecmoss samples are part of the negative co-ordination process between the surveys undertaken at INSEE.

Around 18,000 establishments are surveyed, representing 155,000 employees

The ESS 2025 questionnaires are provided in the french section.

The results are analysed and compared with other indicators disseminated by INSEE, particularly in the context of the quality report sent to Eurostat.

Enrichment with administrative data

At the end of the collection process, the respondent file is enriched with information from the ‘All Employees’ database (BTS), derived from nominative social declarations (DSN). This enrichment of information from administrative sources is central to the survey process, both to supplement the survey data and to verify the consistency between administrative information and information from the survey and decide whether to make adjustments if necessary.

Clearing and correction of non-response

For Employee questionnaires:

Dares is responsible for adjusting the remuneration and working time variables in the Employee questionnaires. They eliminate Employee questionnaires for which no response has been received or which cannot be used, and carries out checks and adjustments using individual data from the BTS and other survey variables in an individual-level consistency analysis.

INSEE then performs a second set of adjustments on the employee data to meet the requirements imposed by Eurostat. In particular, special processing is carried out for employees on a fixed salary, for whom Eurostat requires the number of hours paid to be provided.

For Establishment questionnaires:

The adjustment of Establishment questionnaires only concerns the Ecmo format, for which the establishment section is essential for responding to Eurostat.

The first stage of the adjustment consists of distinguishing respondents according to their level of response: they may respond for the establishment (expected level of response) or for the enterprise, when the information at establishment level is not known. The Establishment database is then cleaned up by switching to non-response those questionnaires for which the key variables (payroll and employer contributions) are not filled in and which are therefore considered as partial non-response. Finally, missing or incorrectly completed responses are adjusted by deterministic imputation based on consistency between administrative data and survey data, or by random imputation when the source of the error cannot be identified.

Treatment of influential units

A treatment of influential units is performed on each adjusted annual Employee base, which makes it possible to control the ‘influence’ of individuals who, due to their responses and their high weighting – and without their responses being incorrect – lead to measurements that are certainly still unbiased but potentially much less accurate for statistics of interest in the areas to which they belong. A winsorisation technique (Kokic and Bell method) is thus applied, which reduces the weight of the influential individual without losing the information in their response.

Calibration to margins

The Employee and Establishment bases are calibrated for each survey year. The calibration method consists of calibrating the variables from survey N to the margins of the total population in paid employment from BTS N in the Ecmoss field, according to numerous criteria (social category, gender, geographical location, etc.).

Marginal calibrations are now performed using the Icarus package from R