/ UiT The Arctic University of Norway

Description of the Population and Housing Census 1990

1. Purpose

2. Survey design

3. Sources of error and uncertainty of results

4. Units and variables

5. Comments on the tables

Information for the public

Personal form

Housing form

Population and Housing Census 1990

1. Purpose

The main purpose of the Population and Housing Census 1990 (Census 90) is to present a picture of the population and living conditions in the Norwegian society. Modern censuses attach importance to acquiring information about housing and housing conditions, employment and occupation, education, income, work journeys, etc, in context.

The purpose of this publication is to provide a description of the whole country and to gather some central national figures into one volume.

2. Survey design

2.1. Scope of the statistics

Census 90 includes all persons (including foreign citizens) registered as resident in Norway according to the Central Population Register per census day 3 November 1990, pursuant to the Population Registration Act of 16 January 1970 and appurtenant provisions.

The housing census includes private dwellings (flats/apartments) where at least one person was registered as resident on 3 November 1990, according to the residential address in the Central Population Register and information from the population census forms.

The census does not include unoccupied (empty) dwellings and dwellings occupied by persons registered as resident elsewhere (e.g. unmarried students or school pupils living away from their parents' home). Dwelling information is not gathered for occupants of old people's home, nursing homes, orphanages and other joint institutions.

House is not a statistical unit in the census, however some information is given about the house in which the dwelling is.

2.2. Statistical basis

In municipalities of 6 000 or more inhabitants excluding 7 municipalities referred to below - the statistics is based on information from various statistical and administrative registers and questionnaires sent to a sample of the population aged 16 and over (born in 1974 or earlier). These municipalities, 166 in all, will henceforth be referred to as sample municipalities. Full census was done in municipalities with less than 6 000 inhabitants, with questionnaires sent to all persons aged 16 and over (full-census municipalities).

Seven municipalities with 6 000 or more inhabitants, requested Statistics Norway to prepare statistics for them on the basis of questionnaire information from all residents of the municipality. These municipalities; Stryn, Sula, Vestnes, Oppdal, Stjørdal, Brønnøy and Vestvågøy had to pay the extra costs involved. Hence there was a total of 282 full census municipalities.

Using samples in 166 municipalities means that data that could not be taken from registers had to be estimated on the basis of a representative sample. On average (for the country) this sample represented over 28 percent of the population.

The most important variables taken directly from the registers are sex, age, other demographic variables, education and income. See annex 8.

Several publications and reports have been published to describe subjects like background, method, content and processing routines, see annex 13.

The Ministry of Finance resolved that Statistics Norway should carry out a population and housing census in 1990. The resolution is based on the Act relating to Official Statistics Act Statistics Norway, no. 54 of 16 June 1989 (the Statistics Act). The Norwegian Storting (the Norwegian Parliament) allocated funds for carrying out the census.

The costs of carrying out the census in this way, was NOK 60 million. This is NOK 40 million less than if the entire population 16 years and over were to respond to the questionnaires.

2.3. Collection of data

The census questionnaires were sent by post directly to each individual respondent. Names and addresses were preprinted on the forms, based on information from the Central Population Register.

In full-census municipalities, the oldest member of each family received a two-sided personal/housing form (annex 1), guidance (annex 2), information (annex 3) and a reply envelope. Other persons aged 16 years and over only received a personal form (blank on the back) with guidance and information. Completed forms from all respondents in the dwelling were sent back together in the same reply envelope. Also where several families lived in the same dwelling, only one reply envelope was used and only one housing form was completed. Hence the collection method provided a basis for grouping families and persons into households, and was the same as that used in Census 80.

In the sample municipalities, all persons selected received a two-sided personal/housing form (annex 4), guidance (annex 5), information (annex 6) as well as a reply envelope. Respondents answered questions about themselves and the dwelling/ household in which they lived. The housing form also contained a preprinted list of persons aged 16 and over belonging to the same family as the respondent. The respondent supplemented the list of names with any other persons living in the dwelling/flat (i.e. registered as residents at that address) and responded to questions about the economic activity of the persons on the list. This information is used to determine the composition of households (see section 4.1 below) and to assign various characteristics to the households such as number of economically active.

Persons who did not return their personal form in time were sent two addressed reminders.

2.4. Sample

The population sample was drawn at the individual level and in such a way as to make the sample representative with respect to sex and age.

Since the individual was the unit for drawing and because a dwelling/household unit was established for each person, the sample of dwellings in a sample municipality is almost doubled compared to a direct sample of dwellings/households.

The sample percentage varies according to the size of municipality:

Size of municipalitySample percentage
6 000-7 999 inhabitants


8 000-9 999


10 000-49 999


50 000 inhabitants and over


In this way, all sample municipalities have got approximately equal municipal statistics.

Results that are based on a sample of the population are associated with some uncertainty. This particularly applies to small geographical areas or small population groups, such as occupational groups in which few people are active.

Statistics Norway has developed and implemented a new statistical method (also referred to as the "combined" method). This method makes it possible to exploit administrative data in the production of statistics, even though the quality of the registers is not alone satisfactory for this purpose. This method involves comparing information given in the questionnaire with information in the register, at the individual level. Based on this double set of information about the same variable for persons in the sample, adjustment factors are created for use in the production of tables. This method which combines information from questionnaires and registers, has been used in Census 90 for the variables economic activity and industry, hence reducing the uncertainty of such figures by approximately half. Nevertheless, in this publication this method is only used for economic activity, see also chapter 5 Comments on the tables. Uncertainty is discussed below in section 3.1.

3. Sources of error and uncertainty of results

3.1. Uncertainty due to sampling

When statistics are based upon information on only part of the population, there is bound to be some uncertainty associated with the results. The standard deviation is a measure of this uncertainty. The size of the standard deviation depends, among others, on the number of observations in the survey. The more observations, the less the standard deviation and uncertainty therewith.

Table 1. Relative standard deviation for published figures on individuals (national figures) of different size.

Sample percentage Estimated number of persons
50 100 500 1 000 5 000 10 000 25 000
10,0 42,4 30,0 13,3 9,4 4,0 2,7 1,3

Table 2. Relative standard deviation for published dwelling/household figures (national figures) of different sizes. Per cent

Sample percentage Estimated number of dwellings/households
25 50 100 500 1 000 5 000 10 000 25 000
10,0 41,3 29,2 20,6 9,1 6,4 2,6 1,6 0,3

Sometimes, relative standard deviation is also used as a measure of uncertainty. This term means the ratio between the standard deviation and the estimated figure, expressed as a percentage. In table 1 some relative standard deviations are indicated for national figures, by size of the estimated figure when individual is the statistical unit.

The estimated relative standard deviations in table 1 will be too large in some cases. This applies to the variable economic activity in particular, where the relative standard deviation can be estimated by multiplying with a factor of about 0.6. The reduction is due to the use of register data in statistical production (see section 2.4 on the "combined" method).

Table 2 shows corresponding relative standard deviation when dwelling or household is the statistical unit used.

For the variables number of persons in the household and number of children in the household, relative standard deviation will be considerably less. This is because information about family relations in the Central Population Register is used to inflate the figures.

Confidence interval

To illustrate uncertainty, one can use an interval to indicate the level of the true value of an estimated figure, that is the obtained value if one had carried out a full census instead of a sample survey. Such an interval is called a confidence interval, if constructed by a special method. In this context one can use the following method: Let M be the estimated figure and let S be an estimate for the standard deviation of M. The confidence interval will therefore be the interval with the boundaries (M - 2 x S) and (M + 2 x S). This method will, with about 95 per cent probability, give an interval containing the true value. In other words, no more than 5 per cent of the figures will have an error of more than twice the standard deviation.

The following example illustrates how one can use table 2 to find a confidence interval:

A figure of 10 000 dwellings has been estimated in a certain category of the country as a whole. The relative standard deviation is 1.6 percent; hence the standard deviation is


This means that there is 95 percent probability that the right figure lies within the interval with limits (10 000 - 2 x 160) and (10 000 + 2 x 160), i.e. within the interval (9 680, 10 320).

Publishing strategy

Statistics Norway has chosen to use a three-pronged strategy:

1. Figures with a relative standard deviation below 20 percent are published directly.
2. Where there is relative standard deviation between 20 and 30 percent, the figures are put in brackets. This means that Statistics Norway believes the figures published can be justified, but they must be used with caution.
3. Figures with a relative standard deviation of 30 percent and above are suppressed by replacing the figure with a colon.
4. Although it is possible to obtain the suppressed figure by calculation, Statistics Norway cannot recommend such a method because the figure would be too unreliable.

The publishing strategy for county figures is referred to in more detail in annex 11.

3.2. Non-response and sample bias

Altogether 968 030 persons were obliged to respond to census questionnaires. Of these, 12 584 persons did not return the questionnaires. This means that the census response was 98.7 for the country as a whole. For individual municipalities, response varied between 100 and 95. Furthermore, there is some non-response to individual questions. This nonresponse may lead to bias since the responses of others were used to calculate values where data was lacking. The size of the bias will however be fairly acceptable, as the response rate is so high.

3.3. Errors in collection and processing

During preparation and processing of a statistical survey, errors always occur and may be reflected in the results. This may occur both in full censuses and sample-based results.

Most errors usually occur during collection of data and are due to wrong responses or inadequately completed questionnaires. During processing, errors may occur when information is coded. This applies especially to information about industry and occupation. Beside these two variables working hours and place (municipality) of reporting for work the only variables that were coded manually. The coding routines are described in Rapporter 92/19 Folke- og boligtelling 1990 - Dokumentasjon av kodeopplegget i Folke- og boligtelling 1990 (in Norwegian only).

Statistics Norway did not contact the respondent if errors were found or forms were inadequately completed. However, the material underwent comprehensive automatic control and correction routines. For variables with no response, a most probable response was determined on the basis of the other information the individual provided and from responses of other respondents.

Control and correction routines are described in more detail in Rapporter 92/18 Folke- og boligtelling 1990 - Dokumentasjon av kontroll- og opprettingsreglerfor skjemakjennemerker (in Norwegian only).

Statistics Norway did a quality survey in connection with the Census. One of the goals of the survey was to check the quality of household composition in the census. The results from the survey have been published in Rapporter 96/10 Kvalitetsundersokelsen for Folke- og boligtelling 1990 (English translation available: The Post Enumeration Survey for the 1990 Norwegian Population and Housing Census, by Arnfinn Schjalm).

4. Units and variables

Unless otherwise specified, data in the population and housing census describes the situation per 3 November 1990. A systematic review of the content of the census and the sources of the different variables is presented in annex 8.

This publication does not provide figures for geographic breakdowns of the country, urban settlements or type of residence area (densely or sparsely populated areas). Figures for counties, municipalities, statistical tracts, basic (statistical) areas and urban settlements are respectively published in the national summary, county summaries and municipal reports from Census 90.

Nevertheless, the most central geographic concepts are described in annex 12.

4.1. Units

Person The census includes all persons registered as resident in Norway in the Central Population Register, on census day 3 November 1990. The Population Registration Act with provisions determine who shall be considered as resident and where (at which address they reside).


In this publication, "resident" refers to persons residing in private households. The total number of residents will therefore be less than the total number of persons (population), which also includes persons registered as residents of joint households.


In tables 1.16, 1.23, 1.24, 3.3, 3.4, 3.7 and 3.9, the term children includes unmarried children under 18 years registered as living with the family of at least one of their parents or other guardian. Older children living at home are not considered children by this definition.

Couples (married couples and cohabiting couples) In this publication, a married couple is a couple where both spouses is registered as residing at the same address (private dwelling, old people's home, nursing home or such).

A cohabiting couple consists of two persons of the opposite sex, who responded that they were cohabiting (without being formally married to each other) and were registered as residing at the same address.

Thus, in order to be counted as a couple, both persons had to be registered as residing at the same address, which is not always the case for people who are married to each other.


A family includes: Married couples with or without unmarried children living at home, irrespective of the child's age mother/father with unmarried children living at home irrespective of the child's age every individual person (single/lone) who does not belong to the groups above.

In the Central Population Register every person has a family number. This family number is used to group individuals into families.

In the census, all persons in one family must be registered as resident at the same address (same municipality and postal area/post code), and they may only belong to one household. Persons, for e.g. spouses registered at different addresses, are not grouped into the same family in the census, even though they might have the same family number in the Central Population Register.

Children who no longer belong to the family (statistically) because they have their own children, are officially registered at a new address and/or have married, are henceforth never to be counted with the parents' family again. For e.g. when a married couple live with a divorced child, the spouses and the child constitute separate families.

A family may never consist of more than two generations.

Unmarried siblings living together in one dwelling where neither of the parents resides are each considered as one family with one person.

Persons with unknown residence or family number are considered lone (one-person family).

For a cohabiting couple, each of the cohabitants constitutes their own family with their own children. Children they have together are included in one of their parent's family.

Family nucleus

In grouping households composed of more than one family (tables 2.3, 3.1 and 3.2) family nucleus is used as a joint term for all family types excluding single (lone) persons (cf. preceding section).


Persons registered as residing at the same address and who responded that they lived in the same dwelling, are considered to belong to the same household. Such a household is referred to as a dwelling household. A household may consist of one or more families. The composition of dwelling households was mainly based on information derived from the housing form (the "name list" in sample municipalities) and the way persons from the same household returned questionnaires in a common reply envelope. Even though the Central Population Register does not provide information about who lives together in each individual dwelling, the addresses recorded in the register are used as important supplementary information in the composition of households.

Census 90 does not give any information about households where household members have joint meals.

The Census distinguishes between private households and joint households (residences for communities). The number of joint households is not available. However, the number of occupants in such households will appear indirectly.

A private household includes persons registered as living at the same address who also responded in the questionnaire that they lived in the same private dwelling/apartment. Since this is the only type of household for which the census has further information, the total of households is equal to the number of private households.

A joint household includes occupants (persons being cared for) that are registered as resident (at the address of) an old people's home, nursing home, orphanage and such. Persons employed in/in connection with joint households are always considered as residing in a private household. The same applies to military personnel who are registered as living at the camp's address. In sample municipalities, only persons 67 years and over are considered as residing in joint households.

For joint households, Census 90 does not provide other statistics than the number of residents in such households. This is mainly because persons staying at institutions but still registered as living at home (married persons with spouse living outside the institution), statistically belong to this private household. Statistics on such a basis would therefore have given a misleading picture of joint households.

Dwelling (apartment/dwelling unit) A dwelling is one or more rooms that are built or converted into all-year private dwellings for one or more persons. There must be a means of access to the room(s) without having to go through another dwelling. For simplicity's sake we use the term dwelling instead of private dwelling in the tables.

A dwelling is for e.g. detached/terraced house (dwelling unit), dwelling unit in a semi-detached house, an apartment rented from a private person, company, society or co-operative, a flat in a block or a bachelor flat. A detached house is also a dwelling unit.

A bachelor flat in a private house is a dwelling unit if it has a separate entrance. In a block of bachelor flats, each flat is considered as a separate dwelling, including places where rooms like kitchen and bathroom are shared. Apartments and bachelor flats used by private households at hospitals, institutions, military camps and such are always considered as separate dwellings.

Because the term dwelling household is used, the number of dwellings is equal to the number of private households in the census.

Data has not been obtained for dwellings where no person is registered as resident (see section 2.1).

House is not a separate unit in the census, however dwelling variables include some information about the house in which the dwelling is located.

4.2. Person variables

Age In age distribution, the individual is classified by age on 31 December 1990. (Age = 1990 minus year of birth).

Earned income

Earned income is equal to the sum of pensionable income. Pensionable income includes income from work inside and outside ordinary jobs and income from self-employment. Sub-groups of pensionable income are respectively income from employment and income from self-employment. Pensionable income is further classified as income from farming/ forestry/fishing and income from other industries. Only income that can be ascribed to the taxpayer's work contribution may be considered as pensionable. Pure returns on capital such as interest received or share dividends as well as pensions, endowments and similar payments are not considered as pensionable income.

Journey to work - general

Data on journey to work in this publication covers the place of reporting for work, number of work journeys and traveling time to work (see these). The variables apply to persons 16 years and over who performed paid employment in the census week 27 October - 2 November 1990, and had a regular place of reporting for work. Data on journey to work applies to the journey from home (registered residence) to the place of work.

Special emphasis was laid on the design of questions pertaining to journey to work. It was explained in the questions and guidance that domicile is the residential address at which the person in question was registered as resident on 3 November, according to the Population Register.

Work journeys - number of

The number of work journeys is the number of times the respondent travelled from his/her domicile to the place of work (one way) in the census week. Domicile always refers to the residential address at which the person in question is registered as resident. For persons who were temporarily living for e.g. in a bachelor flat/barracks that week, travel between bachelor flat/barracks is not included. Persons working at home (e.g. farmers), are grouped in did not travel. The same applies to persons that were not at their domiciles during the census week.

Working hours

The census has information about working hours in two time periods: in the year: (the 12-month period 3 November 1989 -2 November 1990) and in the census week (27 October - 2 November 1990). For that week, a distinction was also made between normal/agreed and actual working hours.

Working hours in the year (3 Nov. 1989 2 Nov. 1990) In Census 90, annual working hours are given in the questionnaire as the total number of complete months in which the individual had employment, full-time and/or part-time, counting all types of work. Full-time is considered as total working hours of 30 hours or more per week, corresponding to 120 hours or more over a period of four weeks. Absence is not subtracted. The sum of months with full-time and part-time work may not exceed 12.

Working hours in the census week (normal/agreed) The agreed weekly working hours is given in man-hours and indicates the number of hours for which the economically active individual was employed in the census week. For economically active persons who had not entered into agreement on fixed working hours, the data applies to average working hours. The same applies to persons with varying working hours or for those who had shift work combined with off periods. Regular preparatory work outside working hours is included (teachers). Working hours of several jobs should be summed up. Paid leave is not subtracted, on the contrary however, incidental extra work (seasonal variations, overtime) is not added.

Actual hours worked in the census week

The actual number of hours worked are given in man-hours. Overtime and other extra work are included. However, the concept does not include absence due to sickness, holidays, paid leave and the like. Persons that were not at work in the census week for e.g. because of shift or rotation work, did not perform paid employment that week.

The labor force Employed persons and applicants for work without earned income belong to the labor force.

Persons at ages 16 and over who declared they had a job during the reference week, are counted as employed persons. Applicants for work without earned income are persons who are registered as entirely unemployed in the registers of the Labor Marked Authorities. Applicants for work and persons who are entirely unemployed are registered at the Employment Offices and entered into the registers.

Residence - municipality of residence

In the census the individual is considered to have his/her residence in the municipality and at the address at which he/she was registered as resident on 3 November 1990, according to the population register. This is to say that the census provides figures for the resident (de jure) population.

The main rule is that a person shall be registered as resident at the place where he/she takes his/her diurnal/nocturnal rest. The most important exceptions from this main rule apply to:

Unmarried persons staying outside their parents'/guardians' residence because of studies, schooling or compulsory national service. They shall be registered as living with parents/guardians. Married persons who have to live away from home most of the time because of employment, studies, national service (e.g. economically active persons that are weekly commuters). They shall be registered as living at the married couple's joint residence. This also applies to persons at an old people's home or nursing home with a spouse living in their joint residence. Persons admitted into hospital or imprisoned. They shall be registered where they were living before the admittance/imprisonment.

Persons who could not be connected to a specific address are considered as living in the municipality in which they had their last fixed residence. In distributions by type of residence area, statistical tracts and basic areas are included among unknown.

Information has not been gathered on temporary residence. Hence the census does not provide information about the present (de facto) population.

Type of residence area (densely populated sparsely populated) See urban settlement - densely populated/sparsely-populated area in appendix 12.

Marital status

Three groups of marital status are used; unmarried, married, previously married (widows/widowers, separated and divorced). Marital status describes the formal situation.

Type of family In the section on the unit family (section 4.1 above) it is made clear as to whom shall be considered as belonging to the same family. On the basis of which persons are included in the family and the marital status of every individual, families are grouped into the following types:

Married couple without unmarried children: the family consists of a married couple with no unmarried children living at home.

Married couple with unmarried children: the family consists of a married couple and the couple's unmarried children living at home, irrespective of the child's age.

Mother/father with unmarried children: the family consists of father or mother and his/her unmarried child/children living at home, irrespective of the child's age

Single person: Lone person (one-person family)

Family type is based on the formal marital status of the persons in the family. Hence for a cohabiting couple, each cohabitant is counted as a separate family.


A person is classified as having full-time work when the sum of agreed/normal working hours for all paid employment are 30 or more hours per week (120 or more hours in a four-week period).

Income See earned income and disposable income.


Economically active persons aged 16 years and over are classified by industry on the basis of information from the 12-month period. This variable is connected to the employment situation with the longest duration. No specific data on industry was gathered for the census week 27 October - 2 November.

The classification of industries is based on the 1983 edition of Standard for Industrial Classification (Standards for Norwegian Statistics 2), with some minor changes. The standard is a 5-digit pyramidal system in which each digit represents a classification level.

Census 80 was classified by the previous edition of the Standard Industrial Classification (published in 1978). The most important changes from this edition can be found under Retail Trade. The changes in the standard are of no significance for comparability with the previous census for industry at the 1-digit level.

At the 3-digit level the following changes have been made in the new edition:

625 Retail sale of hardware, kitchen equipment, glass, paints and painting articles and sporting equipment. Change in text. Of no consequence for comparability. (A merging of 2 4-digit groups)

626 Retail sale of watches and clocks, optical goods, musical instruments, gold and silverware and film and photo goods. Change in text. Of no consequence for comparability.

631 Restaurant and other catering work. Change in text, sales from hot-dog stands transferred from 622 in 1980 to 631 in 1990. Affects comparability at the 2- and 3-digit level.

632 Hotel and overnight accommodation services. Change in text. Of no significance for comparability.

714 Pipeline Transport. New main group of industry formerly under 711 Land Transport. Affects comparability at the 3-digit level, however only comprises a small part of the whole.

After minor modifications at the level of tables, the industrial classification can be compared with the International Standard Industrial Classification of all Economic Activities (ISIC) of 1968.

Place of reporting for work

The place of reporting is the place where the respondent showed up for work at the beginning of the working day. Information is available at the municipality level if a specific place of reporting is given. Persons who reported for work at different locations, were at sea etc, do not have a specific place of reporting.

Means of travel to work

Figures for means of travel are not given in this publication. See the national summary, county summaries and municipal reports.

Traveling time to work

Traveling time is the duration of time from the domicile (registered residential address) to the place of work (one way). Walking time and waiting time during the journey to work are included. The same applies to regular duties such as taking children to kindergarten. On the contrary, incidental duties and delays are not included.

For commuters who lived temporarily outside home (bachelor flats, barracks), travelling time is calculated from the domicile to the work place.

Cohabiting couples - cohabitants See couples in section 4.1.


Information about citizenship is taken from the Central Population Register. Only one citizenship is registered there. If a person has both Norwegian and foreign citizenship, the Norwegian citizenship is registered.


Data on education in the publication have been taken from Statistics Norway's statistical register on the highest level of the population's education by 1 October 1990, re-coded edition. The re-coding implies that the 9-year primary/lower secondary school is used as a basis for classifying basic vocational education, even when this had been completed before the school year 1972/73. (See below)

The data material on education completed abroad after 1 November 1980 is not complete. Among other things, this applies to education completed by Norwegians who have studied abroad. Educational information for most persons who immigrated to Norway after 1980 has been added to the data material from a separate survey among such persons born abroad.

All education statistics from Statistics Norway are coded according to the Norwegian Standard Classification of Education. The standard is constructed as a pyramidal classification system. Education is first ordered into levels of education according to (total) duration by a normal progression of studies. Within each level, education is subdivided into special fields of study and subject groups.

From the 1987 statistics inclusive, a new edition has been adopted: the Norwegian Standard Classification of Education Revised 1989 (Standarder for norsk statistikk 7). Simultaneously some education courses were placed at a higher level. The largest groups were engineering and health care subjects.

In Census 90, as well as in annual statistics on the population's highest level of education, re-coded education is published. Here the classification by level is based on the 9-year primary/lower secondary school, and accordingly, basic vocational training completed before 1972/73 is placed at grade 10, upper secondary school level. Previous censuses also used the old standard classification of education. Consequently, published figures for education from Census 90 and older censuses are not comparable. Statistics Norway can make Census 90 figures comparable to earlier censuses for the most important education levels by re-coding all basic vocational training completed before 1972/73 to grade 9.


Economically active persons 16 years and over are classified by occupation on the basis of information for the 12-month period. This variable is linked to the employment situation of longest duration. No specific data was collected for occupation for the census week from 27 October to 2 November.

The basis for occupational classification is the Norwegian Standard Classification of Occupations (in Official Norwegian Statistics), published by the Directorate of Labor in 1965. The standard is often referred to as the NYK (Nordisk yrkesklassifisering: Nordic Classification of occupations). The standard was brought up-to-date for Census 80. New occupational terms were rightly placed in the system, a number of occupation groups were combined and some were divided into several groups. Since 1980, Statistics Norway has continued bringing new occupational terms up-to-date.

In the census, occupation is coded by using three of the standard's five digits.

Occupation is coded by the same method and standard both in Census 80 and Census 90, making it possible to compare occupational statistics from the two censuses.

NYK and the census material are to a limited extent comparable with the latest, the 1988-edition of the International Standard Classification of Occupations (ISCO). On the other hand there is good concordance with the previous ISCO (1958, revised in 1968). This is the version that is used in most European countries per 1990.

Economic activity

In Census 90, data is collected on economic activity both for the last 12 months (3 November 1989-2 November 1990) and for the week prior to census day (27 October-2 November 1990).

Economically active comprises persons aged 16 and over that had paid employment for at least 100 hours in the 12-month period. Paid employment is all work done with remuneration in the form of wages, income from self-employment, commissions, fees, etc. Part-time work, work as substitute, childminding, holiday work etc. are also included. The same applies to sick leave, holidays, paid leave etc. In Census 80, such temporary absence was not to be counted. Working as a family member without fixed pay in a family enterprise (e.g. shop, farm) and national service, are also considered as economic activity.

Those who were economically active in the 12-month period also responded to the question as to whether

they had paid employment of at least one hour's duration in the week 27 October - 2 November 1990. In the Census 80 a person should have performed at least one hour's work in order to be considered economically active in that week.

Occupational status

Economically active persons 16 years and over are classified by occupational status from the work situation they gave for the 12-month period. Specified data on occupational status was not collected for the census week 27 October - 2 November.

Persons (also co-owners) who worked in limited liability companies, co-operative societies or other company forms with limited liability are always considered as employees.

A person is considered self-employed if he/she operated a one-man business or was co-owner of an unlimited liability business(e.g. partnership).

The group family member without fixed pay in family enterprise comprises persons who worked in family enterprises without fixed pay, and who were neither owners nor co-owners of the enterprise. A family member without fixed pay may, however, be remunerated for work in the form of board and lodging, share of profits, etc.

In cases where married couples worked in a joint enterprise (e.g. farm) which was not organized as a limited company or the like, one spouse was considered self-employed and the other, family member without fixed pay. The exception here is if both spouses were registered as owners of the enterprise.

4.3. Housing and building variables

Housing and building variables only apply to private dwellings and persons in private households.

Floor space Floor space of a dwelling is the total area of all rooms that are used for dwelling purposes. The entrance, hallways, laundry room, fitness room etc are also included in the area.

Occupants per dwelling ,Occupants per dwelling is equal to the number of residents in private dwellings divided by the total number of private dwellings.

Period of construction

The period of construction is the year in which at least half of the dwellings in the building were ready for occupation. For buildings that have been rebuilt, the original period of construction is registered.

Tenure status

Persons having the right of occupation by being members of a housing co-operative are also considered as owners.

Tenancy agreements running until an agreed date as well as subleases are considered as lease. On the contrary, dwellings put at one's disposal through a work situation for a limited period of time are included among staff dwellings.

Leasing a dwelling on other terms apply for e.g. to a tenancy agreement according to the Landlord and Tenant Act, verbal leasing agreements or conditional agreement. Dwellings financed through mortgage credit bonds and municipal housing for pensioners are also included here.


This variable indicates in which of the building's floors the dwelling has rooms for dwelling purposes. For dwellings on several levels, all floors are given. See also accessibility in section 4.3.

Elevators See accessibility in section 4.3.

Type of building A detached house is separated from the next building by at least half a metre. It may also contain a bachelor flat, and one or more rooms may be used as a bachelor flat.

A farmhouse includes all types of residential buildings on the farm.

A linked house (also including a detached chain house) is connected to the next house by a storeroom, garage/carport, etc.

In a terraced house, dwellings are built together on a slope where several of the dwellings have terraces covering all or part of the roof of the dwelling below. Terraced blocks are considered as blocks and not as terraced houses.

Other building with less than three storeys include for example, residential buildings with four dwellings, detached house with less than half a metre's distance from the next house, etc.

The dwelling is regarded to be ins a commercial building or in a residential building for communities (e.g. boarding house, old people's home), if less than half of the floor space in the house is used for private dwelling.

Linked houses, terraced houses and other apartment buildings with less than three storeys may all be denoted by the term small houses.


There is no lower limit (area) for a kitchen. Kitchen counters and cupboards are included in the floor space.

Heating source In Census 90 all sources of heating in the dwelling are included, whereas in the previous census only the principal source was registered.

Floor levels in the dwelling

The dwelling is considered to be on one floor if the kitchen, bath/shower, WC and at least one living room are on the same floor. See also accessibility in section 4.3.

Sanitary conditions (bath/shower - WC) Only bath/shower and WC in the dwelling are counted. Here bath/shower means bath or shower or both.

Number of rooms

Number of rooms includes dwelling rooms of 6 square metres or more that can be used all year round. Kitchen, bathroom, WC, laundry room, corridors etc, are not included. Rooms that are only used for commercial purposes are not included either.

Rooms per dwelling

Rooms per dwelling is total rooms divided by the number of private dwellings. In Census 90, the number of rooms in dwellings with more than 8 rooms was not further specified. In data on total rooms, the figures for the number of rooms in dwellings with nine or more rooms are therefore estimated on the basis of an average figure. Data from the 1988 Survey of Housing Conditions was used as a basis for calculation.

Accessibility As a measure of accessibility of the dwelling (in addition to floor levels in the dwelling) tables 2.1, 2.2, 3.5, and 3.8 show the storeys of the building in which the dwelling is located and whether the building has an elevator.

4.4. Household variables

Household variables apply to private households and persons in private households.

Waste sorting (sorting at source)

Data on the sorting of waste shows selected types of waste that are sorted out from household waste and are delivered/collected afterwards. Every household could cross several types of sorted waste on the questionnaire. Some households sort several types of waste. Hence the sum of households by type of waste is larger than the total number of households that sort their waste.

Disposal of private car

Having a car at one's disposal means that at least one car can be used privately.

All cars used privately by the residents of the dwelling are taken into account, including estate cars, combined passenger cars and vans for private purposes.

Disposable income

Disposable income is the sum of several different types of income, minus direct taxes. The following method of calculation is used:

Net income (according to the central government taxation rules) or basis for surtax on gross income (the highest amount for every individual in the household is selected) + Student loans and education grants from the State Educational Loan Fund Housing allowance from the Norwegian National Housing Bank + Family allowance + Misc. tax-free social security benefits (including basic benefit, supplementary benefit, maternity benefit, educational grants etc.) + Parent's tax deduction + Other transfers = Total income - Total assessed taxes = Disposable income

Disposable income is only used as an income term for households, not for individuals. This is because several types of income such as housing allowance, are paid to only one member of the household.

Number of families in the households See family in section 4.1 and family type in section 4.2.

Households (number of persons/children in the household) In the tables, households are in some cases divided by the number of families, that is again divided into lone-person families and nuclear families. The nuclear family is used as a joint term for all family types with the exception of single persons (cf. previous section).

In tables 3.3, 3.4, 3.7 and 3.8, the households are divided by the number of persons in the household. In some of the tables the households are further divided by the number of children and/or with/without children.

The term children is referred to in section 4.1. Households in which all children living at home are 18 years and over are therefore included in the group households without children.

Disposable household income See disposable income in section 4.4.

Private/joint household Private household and joint household are defined under households in section 4.1.

5. Comments on the tables

The tables in this publication contain some minor discrepancies with relation to previously published municipal reports and county and national summaries. This is due to several factors:

More complete data (material)

The first municipal reports and county summaries were published before data for the country as a whole had been fully processed. Hence the figures in this publication are more complete.

Updated data on education

The table on persons by education will not correspond fully with previously published figures. This is because the statistical register of the population's highest completed education per 1 October 1990 were not yet ready when the first municipal reports were published. Therefore these figures contain education data per 1 October 1989. In this publication, education per 1 October 1990 is used as a basis for all municipalities.

Use of the "combined" method

During work with the municipal reports for the sample municipalities adjustment factors were applied to tables with the variables economic activity and industry. The method combines information from questionnaires (sample) and registers and provided figures with approximately half the uncertainty of what equivalent sample data alone would have given. In this publication, the "combined" method is only used for economic activity, such that the figures for industry are only based on sample data. Figures in tables on economically active by industry in this publication, are therefore generally somewhat more uncertain than in the previously published national summary, which - along with the county summaries - showed aggregates of figures for municipals. See section 2.4.

Rounding up of figures

When one calculates total numbers on the basis of a sample, one usually ends up with decimal figures as a result. The figures that are published are rounded up to the nearest whole number, thus resulting in deviation whereby a given summed figure may not always be equal to the sum of the individual figures in the table. The deviation is seldom large, most often 1 or 2. Corresponding deviations may also occur between summed figures in the various tables.

In table 3.9 on couples, the deviation could be larger, up to 10.

Information for the public

Personal form

Housing form