In overseas departments, the sample is a one stage stratified sample of dwellings.

Dwellings are uniformly distributed over reference weeks of the year.

For overseas departments, the base is the French annual population census.

With the new sample, LFS sectors and primary units of the Master Sample used for other soial surveys are coordinated: LFS sectors are selected in the neighbourhood of the sampled PU, in order to avoid the travel costs to reach isolated LFS sectors located far from the interviewer network.

No PSU in overseas departments

The balanced sampling for old sample uses the following variables: age, incomes, type of dwelling, type of urban/rural areas,resident status (owner/tenant).

The balanced sampling for new sample uses the following variables: age/sexe, type of household, QPV, income (total,unemployment benefits, wages). The new sample is spatially balanced in order to limit spatial autocorrelation.

For overseas departments, the sample is composed of dwellings selected through a stratified systematic sampling (systematic sampling with equal inclusion probabilities, within geographic strata which form a partition of the territory).

For Metropolitan France, the sectors are cut into 6 clusters of nearby dwellings, in such a way that there be around 20 main residences in each cluster. Inside the sectors. Each cluster is randomly assigned to a number between 1 and 6; this number determines when the cluster enters the sample, each cluster is interviewed 6 consecutive quarters and then replaced by another cluster of the same sector.

For overseas departments, dwellings are directly selected within strata through a systematic sampling with equal  inclusion probabilities. Strata sample sizes are proportional to the total numbers of main residences in the strata.

For overseas departments : Stratification by small geographic areas (infra-NUTS3, infraemployment zone).

18 strata in overseas departments

Each quarter, the sample contains 6 subsamples: 1/6 of the sample is surveyed for the first time, 1/6 is surveyed for the second time, …, 1/6 is surveyed for the 6th (and last) time.

Since Q1 2019, Corsica is treated separately in the weighting. Before Q1 2019, Corsica was aggregated with Provence-Alpes-Côte-d'Azur.

Since Q1 2019, Corsica is treated separately in the weighting. Before Q1 2019, Corsica was aggregated with Provence-Alpes-Côte-d'Azur.

For overseas departments, weights are first adjusted to correct for non-response by using an estimation of response probabilities (obtained by a logit model) ; In a second step, they are calibrated in each department on external margins (population margins crossed with waves (subsamples), number and type of housings, micro-region (infra-NUTS3), diploma declared in the Population census, place of birth (3 modalities))

When calibrating (both in Metropolitan France and in overseas departments), the sampling weights are corrected so that the number of respondent households correspond to the total number of households in the field, according to different variables. When the variables are individual (for instance age), they are summed at the household level (for instance proportion of 15-20 years-old in the household).

For overseas departments :  type of housing, microregion (infra-NUTS3), being respondent or not during the previous quarter

The number of unemployed persons by NUTS3 is obtained by breaking down national figures for unemployment (computed with the LFS) using the geographical structure of the registered unemployed persons (more precisely, only those who have not worked during the previous month ("DEFM en catégorie A")).

The number of employed persons by NUTS3 is mainly derivated from employment estimations (synthesis of administrative sources and company surveys on employment).

The unemployment rate is calculated using these 2 agregates.

Was the questionnaire updated for the 2020 LFS operation? (Y/N)

New questions about Covid.

Methods used for adjustments for statistical unit non-response

For overseas departments (through an estimation of response probabilities with a logit model) : type of housing, microregion (infra-NUTS3), being respondent or not during the previous quarter

For overseas departments, weights are first adjusted to correct for non-response by using an estimation of response probabilities (obtained by a logit model) ; In a second step, they are calibrated in each department on external margins (population margins crossed with waves (subsamples), number and type of housings, micro-region (infra-NUTS3), diploma declared in the Population census, place of birth (3 modalities))

When calibrating (both in Metropolitan France and in overseas departments), the sampling weights are corrected so that the number of respondent households correspond to the total number of households in the field, according to different variables. When the variables are individual (for instance age), they are summed at the household level (for instance proportion of 15-20 years-old in the household).

of which people who migrated (or moved) internally or abroad (%)

Variable status

Compulsory / optional

Divergence of national concepts from European concepts

(European concept or National proxy concept used) List all concepts where any divergences can be found

« Emploi quelle source pour quel usage » (PDF, in French)

The registered unemployment includes 5 categories. Category A (“DEFM A”) corresponds to job-seekers who have been unemployed over the month and are required to actively seek employment. This category is the most conceptually similar to that of the ILO concept. The other categories encompass either jobseekers who have worked for a short period (less than 78 hours during the month, category B) or for a longer period (over 78 hours, category C), or individuals who are not required to seek employment, because of subsidised employment contracts, training, illness or for other reasons (categories D and E).

Being registered to Pôle emploi under category A and ILO unemployment represent two close realities, but they do not necessarily overlap. Therefore, it is possible for an individual to be a DEFM A, but not ILO unemployed; this can be the case, for example, if the only action he or she made is to renew his ou her registration: being registered to Pôle emploi is not considered an active search according to the ILO  criteria. On the other hand, an individual who is ILO unemployed is not necessarily registered to Pôle emploi under category A : he or she can be not registered at all (for example if he or she is not eligible to any unemployment benefit) or can be registered under a different category (category D, for example, if he or she is following a training course but considers that he or she is available and has actively sought employment, or category B or C if he or she worked during the month outside the reference week).

Consequently, some events can affect the number of DEFM A without necessarily affecting that of ILO unemployed, or vice versa. In particular, reforms or changes of the rules that dictate the follow-up with jobseekers registered with Pôle emploi, the assistance they are provided, and the compensation they receive can have an impact on the number of DEFM A without affecting, with the same magnitude, the number of ILO unemployed.

The registered unemployment is an administrative information derived from the list management files produced by Pôle emploi; the reference period is the month; collective household are included.

Registers: 3.93 millions (in 2020, yearly average, DEFM A concept (registered jobless jobseekers who have to seek a job))

Registers: 0.29 million (in 2020, yearly average,  DEFM A concept (registered jobless jobseekers who have to seek a job)

Registers: 1.71 million (in 2020, yearly average, DEFM A concept (registered jobless jobseekers who have to seek a job)

Registers: 0.26 million (in 2020, yearly average, DEFM A concept (registered jobless jobseekers who have to seek a job)

Registers: 1.66 million (in 2020, yearly average, DEFM A concept (registered jobless jobseekers who have to seek a job)

First step: calculation of the domestic employment; This calculation is based on the Insee estimation of employment made from the Business statistics data. The domestic employment takes into account every person leaving abroad but working in France; conversely, every person leaving in France but working abroad is not taken into account. The calculation is made for every Economic activity of the local unit.

Second step: calculation of the domestic Employment by Economic activity of the firm. The Insee estimations of Employment used are given for each Economic activity of the local unit with the Business statistics data (UNEDIC data and INSEE data).

Third step: calculation of the domestic employment in “full-time equivalent” for each firm’s economic activity, using the annual business survey.

The NA use a co mponent method, using the collective weekly worked hours (measured by a survey on firms named Acemo) adjusted to take into account the non-worked hours (public holidays, sickleave...) and the overtime hours and the self-employed. This estimation marginally uses some LFS results.

Informations Rapides: publication of some statistics on employment, unemployment, "halo" around unemployment, underemployment (every quarter)

Insee Première (IP): 4-pages publication; one IP on main results from the survey (once a year); IP on an analysis of a specific topic (no periodicity)

Insee Résultats: web pages with detailed results of the survey (once a year)

Insee Références "Emploi, chômage, revenus du travail" : publication with factsheets about the labour market (once a year)

https://www.insee.fr/fr/metadonnees/source/operation/s1459/presentation

Link to the national web page (English):

https://www.insee.fr/en/metadonnees/source/operation/s1459/presentation

A more complete anonymised file (but with non sensible variables only) is available to every one after having signed a convention with Insee (for non researchers or specific institutions, some fees may be due) or through a specific network (Centre Quételet : http://www.progedo-adisp.fr/serie_ee.php)

The complete file (with undirectly identifying variables) is available to the researchers via a Research Data Center (CASD), after the agreement of the Statistical confidentiality committee that verifies the purpose of the research.

A public file is available to every one on the Insee web site for free. The file is anonymised and contains very few detailed informations.

A more complete anonymised file (but with non sensible variables only) is available to every one after having signed a convention with Insee (for non researchers or specific institutions, some fees may be due) or through a specific network (Centre Quételet).

The complete file (with undirectly identifying variables) is available (but charged) to the researchers via a Research Data Center (CASD), only with the agreement of the Statistical confidentiality committee that verifies the purpose of the research.

Convention with Insee

Agreement of the Statistical confidentiality committee for full data

Questionnaire, Explanatory
notes (in french)

Anonimization is processed at different levels:

- directly identifying variables (name, address, etc.) are deleted from files. They are kept for a limited time by the production team only.

- undirectly identifying variables (company identification number, NUTS3, etc.) are made available only in a protected file. For access, researchers need the agreement of the Statistical confidentiality committee.

- identification variables in the non protected file are fully anonymised.