Census without questionnaire

Statistics Norway collects information around every 10 years on the entire population of Norway and their housing conditions for the Population and housing census. This is the largest statistical survey that has been undertaken in Norway, as is the case for most other countries.

Until 1970, the method for collecting census data remained more or less the same; enumerators went from door to door asking the necessary questions. In 1980, Statistics Norway sent questionnaires in the mail for the first time ever, and in 2001, respondents were able to submit the data via the Internet. Regardless of the data collection method used, data collection on this scale is very expensive. Statistics Norway therefore began examining other methods for collecting data in the 1970s.

Using data from many different sources

In 2011, Statistics Norway can for the first time compile all the statistics that are required in the Population and housing census without sending out questionnaires to the population. As from 2011, the census will solely be based on administrative registers. It will therefore be referred to as a register - based census .

The most important register that will be used in the census is the Central population register. Since the Register’s inception in 1964, it has not actually been necessary to count the population in the true sense of the word. All births, deaths and migrations are registered, and the population register can therefore tell us who lives in Norway at any given time, as well as where they are domiciled.

Another important register is the Cadastre, which has provided data on dwellings in Norway since 2001. Data is also retrieved from a number of other administrative sources, with the registers in the Norwegian Tax Administration and the Labour and Welfare Service being the most important. These registers hold data on who is employed and unemployed, and who are pensioners. Statistics Norway also has its own education register, with data collected from educational institutions on who are pupils and students, as well as the education level of the population.

In 1980, the Population and housing census began using administrative data to a far greater extent than previously, and the scope of this has increased in every subsequent census. This development does not only apply to the censuses; more and more of the statistics compiled by Statistics Norway are based on such data.

Major savings…

By basing its statistics on administrative data, Statistics Norway is saving society substantial amounts. The Population and housing census 2011 will only cost NOK 3 per capita. This is around 10 per cent of the cost of the 2001 census, even although this was also largely based on administrative data. By way of comparison, the USA, which conducted a traditional census in 2010, spends almost 80 times as much per capita.

The population also saves time by not having to fill in a questionnaire. Completing a questionnaire every 10 years is perhaps not a huge burden, but many no doubt consider it unnecessary to provide information that they know is already held by the authorities.

Data from administrative sources needs to be edited in order to be used as a basis for statistics. However, this is work that Statistics Norway carries out on an ongoing basis anyway. The census can use this data, which has already been edited for statistics that are published regularly within areas such as population, housing and the labour market. This work also incurs costs but is already a standard part of Statistics Norway’s activity.

…but it’s not completely free

Notwithstanding the foregoing, some costs are still incurred in connection with a census. For instance, as this is the first time we will conduct a census based solely on register data, we need to safeguard the consistency between statistics on persons, households and dwellings. However, this type of development work, which is carried out in several different areas, will benefit other statistics and subsequent censuses.

In 2008, the EU ratified legislation on population and housing censuses that is also applicable to Norway as a member of the EEA. The aim was to enable comparable statistics for the entire EEA area. In order to achieve this, the countries, including Norway, must make certain adaptations to their censuses. A large part of the cost of the 2011 census is a result of the requirements set by the EU legislation.

What does the 2011 census contain?

One of the aims of censuses is to throw light on developments over time, particularly in relation to long-term trends. The content must therefore be as similar as possible in every census. Nevertheless, the 2011 census includes some new content.

In the vast majority of housing censuses in Norway, only occupied dwellings have been included, but in 2011, statistics will also be compiled on unoccupied dwellings. This will provide a clearer picture of the total housing stock in Norway.

Another useful addition to the latest census is that Statistics Norway will retrieve students’ addresses from sources other than the population register. In the population register, many students are classed as domiciled at their parents’ address and not at their place of study - where they spend most of their time in reality. This change will enable population figures and household statistics to be compiled based on students’ actual addresses. This information will be very useful in a number of different contexts.

International comparisons have always been an important goal of population censuses, but the countries involved have been free to choose whether they want to adhere to the international recommendations on census content. However, the EU legislation now stipulates specific obligations with regard to the content, which has made it necessary to adapt and extend the content of the 2011 census. For example, some extra statistics on families and households now need to be created.

What does it not contain?

Now that questionnaires are no longer sent out, it will not be possible to obtain information on aspects that are not covered by administrative sources since a register-based census can only be based on the data that is available. In some cases, the registers do not provide information that is fully compatible with the requirements of the statistics. By combining data from various sources, and by applying statistical methods, Statistics Norway is normally able to create statistics that meet the users’ requirements.

However, there will always be limitations to a certain extent. For instance, some of the information on dwellings that was collected using a questionnaire in the last census will no longer be included, e.g. information on the dwelling’s accessibility for wheelchair users, which energy sources are used for heating and whether the dwelling has access to a garden, garage or parking space. Neither will the register-based census provide information on housing conditions for residents who do not live in conventional dwellings, e.g. those who live on boats or in caravans.

Census and other statistics

The Population and housing census now uses the same data basis as other statistics in Statistics Norway. The annual statistics on the population, households, dwellings, labour and education can therefore provide the same information as the census to a large degree. However, Statistics Norway has the opportunity to provide extra details in the census years, particularly in relation to three areas:

• First, the censuses provide long time series. In order to obtain comparable figures over time, a different method is sometimes needed to create the tables than that used in the annual statistics.
• Second, the censuses publish more statistics for municipalities, urban districts and basic statistical units than are found in the annual statistics.
• Third, the census statistics assemble data from various areas, e.g. on who is employed, who is undertaking education, who is a pensioner and the main source of income in households. In the census we will also combine data on ownership of cars with household data, thus enabling us to produce figures for how many households have a car at their disposal in different parts of the country.

First results in summer 2012

The census results will be published on ssb.no as they become available. The first results, including key figures for persons, households and dwellings, are planned for publication at the end of June 2012. Statistics on employment, education and income will follow, as well as more detailed information on households and dwellings. Most of the results will be published in autumn 2012, but some will also be published in spring 2013.

Statistics for municipalities, urban districts and basic statistical units will be accessible via StatBank. In previous censuses, separate statistics publications were compiled for all municipalities and counties. However, guidelines will now be published on ssb.no in order to simplify the retrieval of corresponding statistics in StatBank.

The EU legislation also requires all of the relevant countries to create standardised tables, which will be accessible in a dedicated publishing system in the EU’s statistics office. This will be a flexible system that gives users the opportunity to create detailed tables as required. The main aim of this is to enable international comparisons. However, these tables will not contain statistics for small geographic areas or time series.

Statistics Norway’s annual statistics also provide figures on the population, households, dwellings, labour and education, and in some cases these will be available before the corresponding figures are published as part of the census. Indeed, the census provides figures for the situation on 19 November, while the annual statistics on households and dwellings relate to the situation at the end of the year. It will nevertheless be helpful for users to see the results from the census and the annual statistics in conjunction with each other.