1. Quality information

Higher education qualifications and apprenticeships

Comparison with other data sources suggests that the census results over estimate the number of people with higher education qualifications and apprenticeships. This seems partially down to younger respondents reporting that they have already obtained qualifications for which they are still studying. It also seems partially down to holders of older and non-UK qualifications incorrectly identifying their modern UK equivalent qualifications.

Comparability with the 2011 Census

The structure and content of Census 2021 questions on qualifications largely differ from those in the 2011 Census. This means that differences in these estimates from the two censuses may partly represent real change and partly be because of the change in the way we collected the data.

The new question structure and content for Census 2021 improved the estimated accuracy for highest level qualification however this may be because of online response. The structure and content of Census 2021 questions on qualifications largely differ from those in the 2011 Census. This means that differences in these estimates from the two censuses may partly represent real change and partly be because of the change in the way we collected the data. Therefore, the highest qualification variable is not comparable with the 2011 census.

Age groups split over two academic years

Census Day, 21 March 2021, was partway through the academic year. This means that when analysing younger age groups, particularly people aged 16 to 21 years, there will be split age cohorts. For example, census results for people aged 16 years will include some who turned 16 before the start of the 2020 to 2021 academic year, and who will have obtained some qualifications in summer 2020. The results will also include some people who turned 16 after the start of the 2020 to 2021 academic year and who have not yet obtained the corresponding qualifications. Take care when interpreting census estimates of qualifications for these age cohorts.

Quality of responses from older people and international migrants

The accuracy of census responses for the qualifications questions is expected to be lower where the specific qualification is not listed on the census questionnaire. This is likely to be a particular issue for older people who obtained qualifications that no longer exist. Examples of such qualifications are polytechnic qualifications, General National Vocational Qualifications (GNVQs), and many other vocational, technical and professional qualifications. This also applies to qualifications that exist in very different forms, such as apprenticeships. Difficulties in remembering details of qualifications obtained a long time ago is likely to further lower the accuracy.

The issue of qualifications not being specifically listed on the census questionnaire is also likely to affect responses to these questions from those who obtained their qualifications overseas. There is added complication by these questions collecting attainment data, for example "5 or more GCSEs (A* to C, 9 to 4)", as well as qualification level, for example "NVQ level 1". This meant that people with qualifications obtained overseas had to convert their achieved attainment levels into the equivalent listed on the census questionnaire.

It is important to be aware of these issues if you are analysing qualification profiles for groups with very different age profiles. Similarly, take this into account if you are analysing qualification differences in groups with very different immigrant proportions, such as different ethnic groups.

Incompatibility with other standard classifications

The census qualifications classification is largely consistent with the national qualification frameworks for England and Wales but did not capture the type and level of apprenticeship. Neither is it fully consistent with the international standard classification of education (ISCED) or Department for Education standard definitions of highest level of qualification. Take care if you are comparing census estimates with data based on these other classifications.

Highest qualification: Level 4 or higher

While most people in this group are likely to hold a degree, this group also includes other higher education qualifications below degree level, such as Higher National Certificates (HNC) and Higher National Diplomas (HND). The census did not collect information specifically on degree holders.

Individual qualification categories

The census questionnaire asked people to tick all the categories of qualifications they have obtained. We have assessed estimates of people with each individual qualifications as unreliable, as there is evidence that people have selected only their highest qualification. For example, people have not selected "NVQ level 2" if they had already selected "NVQ level 3" even if they hold both qualifications. So, standard census outputs do not include estimates for each individual category or response option. They instead include qualifications data about the highest level of qualification obtained. This meets the main user need for this data.

Topic summary feedback

Tell us what you think about this publication by answering a few questions.

Back to table of contents

3. Cite this methodology

Office for National Statistics (ONS), released 10 January 2023, ONS website, methodology, Education quality information for Census 2021

Back to table of contents

Contact details for this Methodology

Census customer services
Telephone: +44 1392 444972