:

Martinique:

1.Title of the survey:

Employment Survey, 1986 (Enquête sur l'emploi de 1986).

2.Organization responsible for the survey:

National Institute of Statistics and Economic Studies, Regional Service and Inter-regional Service for Antilles and French Guiana (Institut national de la statistique et des études économiques - INSEE, Service régional et Service interrégional Antilles-Guyane - SIRAG).

3.Coverage of the survey:

(a) Geographical:

The whole territory.

(b) Persons covered:

All ordinary households (as defined by the general population censuses), including members temporarily absent.

Excluded are members of collective households (religious communities, residents of old people's homes, etc.) and populations counted separately (pupils at boarding schools, military personnel in barracks, etc.). Persons absent from ordinary households for a long period or intermittently absent, if they have another principal residence, are not subject to the individual questionnaire on employment; nor was any person born after 1971.

4.Periodicity of the survey:

The survey is carried out every seven years, between population censuses.

5.Reference period:

The week preceding the survey.

6.Topics covered:

The survey provides information on employment, unemployment, hours of work, wages, duration of employment, duration of unemployment, discouraged workers and occasional workers, industry, occupation, status in employment, level of education/qualifications and usual activity.

7.Concepts and definitions:

(a) Employment:

Employed persons comprise all persons aged 15 years and over who state that they have had a gainful activity for one hour or more during the reference week (even when such activity is not a permanent job), and persons who, although employed, are temporarily absent from employment during the reference week and maintain a formal attachment with that employment, by continued payment of wages and assured return to work.

The following persons are therefore included:

  1. persons with a job but temporarily absent from work (for less than six months) because of illness or injury; vacation or annual leave; maternity or paternity leave; educational leave; absence without leave; bad weather or mechanical breakdown; labour-management dispute or other reduction in economic activity;
  2. full- and part-time workers seeking other work during the reference week;
  3. full- and part-time students working full- or part-time;
  4. persons who performed some work for pay or profit during the reference week, while being subject to compulsory schooling; or retired and receiving a pension; or registered as jobseekers at an employment office or receiving unemployment benefit;
  5. paid and unpaid apprentices and trainees;
  6. participants in employment promotion schemes;
  7. paid and unpaid family workers, including those temporarily absent from work;
  8. private domestic servants;
  9. volunteer and career members of the armed forces (excluding conscripts, who are not covered by the survey).

Employed persons are divided into 'modern' members of the economically active population, whose principal activity brings them an income exceeding the Civil Service Minimum (MFP), 'intermediate' members of the economically active population, whose principal activity brings them an income below the MFP but above the statutory minimum wage (SMIC), and active members of the economically active population who are visibly or invisibly underemployed (see Underemployment below).

Excluded from the employed are persons whose only activity is work around the house; conscripts and persons doing civilian service equivalent to military service; and those doing unpaid community or social work.

(b) Underemployment:

It is measured by two yardsticks: weekly hours of work, of which the threshold is fixed at 35 hours (legal hours of work are 39 hours and the hours actually worked are on average 37 hours), and the net monthly minimum wage on the basis of legal hours of work (SMIC). The yardstick of income takes precedence over that of hours of work, and the latter makes it possible to distinguish between visible and invisible underemployment.

The visibly underemployed population is accordingly defined as all persons who are paid less than the monthly SMIC for their principal activity during the reference week and whose weekly hours of work are fewer than 35.

The invisibly underemployed population comprises persons whose principal activity brings them a monthly income which is less than the SMIC although they work more than 35 hours weekly.

(c) Unemployment:

Unemployed persons are persons aged 15 years and over who, during the reference week, did not work for reasons other than illness, leave, bad weather, etc., but who are available for work and actively looked for work during the month preceding the interview.

"Actively looking for work" includes the following steps: visits to employers, replying to advertisements or sending applications, seeking assistance through friends, relatives, etc., awaiting offers from the National Employment Agency (ANPE), etc.

Unemployed persons include:

  1. persons laid off temporarily or for a period of indefinite duration without pay;
  2. persons without a job who are immediately available for work, and who have made arrangements to work in a new job at a date subsequent to the reference period (no time limit is fixed for starting in the new job).

Full-time and part-time students seeking full-time or part-time work; and seasonal workers awaiting seasonal (agricultural or other) work, are not considered as unemployed but as inactive.

Unemployed persons are divided into those unemployed for longer than one year, those unemployed for less than one year, those who are first-time jobseekers, and marginal unemployed (inactive persons who at an advanced stage of the interview say that they are seeking work).

(d) Hours of work:

These are the number of hours actually worked during the reference week in the principal activity.

The survey also covers usual hours of work.

(e) Informal sector:

This topic is not covered by the survey.

(f) Usual activity:

This is measured on the basis of all activities in the year preceding the interview. The information is collected for each month and the number of weeks in each month.

8.Classifications used:

(a) Industry:

Employed and unemployed persons are classified by the Nomenclature of Activities and Products (NAP-1973), and are coded in 99 groups. The NAP is convertible to the International Standard Industrial Classification of all Economic Activities (ISIC-1968) at the second level of detail (transition from the most detailed level, NAP600, to the ISIC 2-digit level).

(b) Occupation:

Employed and unemployed persons are classified according to the Nomenclature of Occupations and Vocational Categories (PCS). Coding is by 42 groups for the employed and 8 groups for the unemployed with previous work experience.

No conversion has so far been established between the PCS and the International Standard Classification of Occupations (ISCO-1968).

(c) Status in employment:

Employed persons are classified in three groups: employers, employees and self-employed.

Unemployed persons with previous work experience are divided into employees and self-employed.

(d) Level of education/qualifications:

All persons aged 15 years and over who are covered by the survey are classified by their highest level of studies and highest diploma obtained.

The classification comprises three sectors: general education (primary or secondary); technical or vocational education; and higher (or higher technical) education. Each sector is divided into subcategories (eight or nine according to sector) by level of studies and highest diploma obtained.

9.Sample size and design:

(a) The sample frame:

This is based on all dwellings included in the 1982 Census as principal or secondary residences or vacant dwellings, and updated by a list of dwellings built since 1982.

(b) The sample:

The sample of dwellings to be surveyed was designed in two phases:

In the first phase, the dwellings covered by the 1982 Census were stratified according to their geographical area and the size of the resident household. The geographical areas are those of the 1982 Census; the sizes of households of principal residences were determined by the number of persons and divided into three groups; the sizes of households of secondary residences and vacant dwellings formed a fourth group.

In this way 28 strata were obtained, in which the sample was selected on the basis of one dwelling out of 50. The sample rate was then reduced by removing two out of ten dwellings from the list previously obtained.

In the second phase a supplementary sample was formed, being selected from the list of dwellings made available after 1 January 1982, using the definitive sample rates.

The sample rate is 1/62.5 and the total size of the sample is 1,739 dwellings, i.e. 1,627 dwellings covered by the Census plus 112 new dwellings.

(c) Rotation:

Not applicable.

10.Field work:

(a) Data collection:

Data are collected by a team of INSEE pollsters recruited for the survey, as household surveys are too infrequent to maintain a permanent operational network.

The interviewer's first job is to find the dwelling he/she is to survey. Data are then collected by means of a household survey form filled up by the interviewer at an interview with the head of the household, and a personal questionnaire (only for persons aged 15 years and over) also completed by the interviewer at direct interviews with the persons concerned.

(b) Substitution of ultimate sampling units:

Some dwellings are hard to find or cannot be found, especially in rural areas. In this case the dwelling is replaced in the sample and the interviewer must then find the new dwelling in the sample.

When the interviewer finds a dwelling that is empty (either because it is vacant or is a secondary residence, or has been provisionally "abandoned" by persons absent for a long time and resident elsewhere for longer than the last six months), or if the dwelling is lived in and its occupants refuse to reply to the survey (although it is compulsory), the dwelling is not replaced but adjusted for.

11.Quality controls:

The questionnaires are checked by INSEE officials, who code the occupations and social categories. The data are then extracted in a supervised data capture phase in which the same officials take part.

12.Weighting the sample:

All the results obtained, after adjustment for non-response and other defects in the survey relating either to individuals or households, were enlarged to the whole territory by multiplying them by the inverse of the sample rate.

13.Sampling errors:

There is no information available on sampling errors.

14.Adjustments:

(a) Population not covered:

No adjustment is made.

(b) Under/overcoverage:

New dwellings being an inadequate basis for the survey, its gross results show male undercoverage. This has been adjusted at the global level on the basis of the estimated population aged 15 years and over on 1 January 1987 and by applying a coefficient of 1.11.

(c) Non-response:

The global rate of non-response is about 5 per cent. Non-response was adjusted as part of the adjustment for undercoverage.

15.Seasonal adjustment:

No adjustment is made for seasonal variations.

16.Non-sampling errors:

The interviewers' greatest difficulty when collecting data was to find out what work had been done, and for how long, during the past week and year. Answers relied too much on memory and are certainly incomplete and inaccurate in many respects.

17.History of the survey:

The first Employment Survey in Martinique was conducted in 1971. The second survey, in 1979/80, was in two stages, the first for the period of high-season agricultural activity (October-December 1979) and the second for that of low-season agricultural activity (April-June 1980).

The 1986 survey differs from previous ones in that the concepts of employment, unemployment and underemployment are nearer the present international recommendations. In previous surveys the principle of "actively seeking work" was not strictly applied. Underemployment was measured in the same way, but its duration was measured taking into account legal hours of work.

18.Documentation:

Institut national de la statistique et des études économiques, Service interrégional Antilles-Guyane : "Les Dossiers Antilles-Guyane No. 14, Enquête sur l'Emploi de 1986, Principaux Résultats" (Pointe-à-Pitre, July 1988).

The survey results are also available on microfiches and unpublished results can be made available, on application.