Do you use census data? We'd like your feedback.

Last updated

14 May 2021

2001 census: Methodology

The 2001 census was the first census in Scotland to represent the entire population.

We achieved this through:

  • running an independent follow-up survey
  • using estimation and imputation processes on census results

Previous censuses provided estimates for Scotland’s population.

Our processes for 2001 meant we could state, with a high level of accuracy, what Scotland’s population was on census night.

Census population process

  • the census questionnaire was distributed and completed across Scotland in April 2001
  • in May and June we ran a follow-up survey called the Census Coverage Survey is across Scotland
  • we matched records from the CCS to those from the 2001 Census
  • we estimated how many people were missed by both the census and Census Coverage Survey from the results of the matching and using dual system estimation techniques
  • we combined the 'missing' population with the Census population to get an estimate of the true population of the sample areas
  • we estimated populations for each local authority by age and sex were using a combination of standard regression and small area estimation techniques
  • we imputed households and persons estimated to have been missed by the Census to produce a fully adjusted Census database
  • we assured the quality of our population estimates using demographic analysis and comparison with aggregate level administrative data

Census Coverage Survey

The Census Coverage Survey involved face-to-face interviews with people from 40,000 households across Scotland.

It was designed to help us identify where the census had under-counted areas, and adjust our population numbers.

We ran the Census Coverage Survey as independently from the 2001 census as possible. This ‘dual system’ approach meant we could estimate:

  • people missed by the census
  • people missed by the Census Coverage Survey
  • people missed by both

We identified areas that were hard to count in both a census and in the survey, and adjusted our estimates. This meant 40,600 people were added to our population estimates.

Edit and imputation

Editing is the process of ironing out inconsistencies in census data. Imputation is filling in gaps in census responses.

We used an Edit and Donor Imputation System developed by the Office for National Statistics on behalf of all UK census offices.

Using the Census Coverage Survey to help us, we were able to impute complete records, households and individuals who were missed from the census.

Read more about edit and imputation for the 2022 census.

Population base

We conducted the 2001 census on a resident basis. That means it collected information for where people usually live, not where they were on census night. People who were away from home on census night were included at their usual address.

In 2001, unlike previous censuses, households who were all away on census night were legally required to complete the census on their return.

Students and schoolchildren

Students and schoolchildren studying away from home were included as residents at their term-time addresses. We only collected basic information, such as name, sex, age, marital status and relationships, for these people at their ‘home’ addresses.

Students and children studying away from home will not be included in the families, household size, or household composition outputs for their home address.

Communal establishments

A communal establishment is a place that provides residential accommodation and is supervised either part-time or full-time.

Communal establishments can include prisons, larger hospitals and hotels.

Small hotels and guesthouses are counted as communal establishments if they can have 10 or more guests, not including the manager and their family.

Sheltered housing is counted as a communal establishment if less than half the residents have their own facilities for cooking. Otherwise, the whole establishment will be treated as separate households.

Communal establishment residents

The rules for whether a person is resident in a household also apply to communal establishments.

A person is a resident of a communal establishment if they have been living there, or intend to live there, for 6 months or more. Visitors on census day who do not have a usual address elsewhere are also counted as a resident.

Usual residents of communal establishments who were away on census day were required to complete a census questionnaire on their return.

In some tables, the term ‘residents’ excludes members of staff and their families.

People sleeping rough

People sleeping rough are those identified as 'absolutely homeless': people sleeping or bedded down in the open air or buildings or other places not designed or suitable for habitation.

They are included in standard tables as residents in communal establishments.