Initial teacher education: methodology and quality report

The primary source of data for most of these statistics is the Higher Education Statistics Agency (HESA) student record. From the 2022 to 2023 academic year, the HESA student record moved from the legacy format to the Data Futures model. This represented a substantial change to the structure and resulted in substantial challenges to the timing and validation of data. Data quality issues were identified in the HESA student record for 2022 to 2023 and 2023 to 2024, some of which required corrections at a higher education provider level.

Due to the nature of the changes and the move from the legacy format to the Data Futures model, the methodology and definitions used in previous ITE releases were also reviewed. Some data quality issues were identified for academic years prior to 2022 to 2023, which were corrected in this release.

Welsh Government continues to work with Jisc, Medr and higher education providers to improve data quality over time. Where comparisons are made, they should be interpreted in the context of the quality and methodology information provided.

See ‘Initial teacher education: definitions’ for more detailed information on how terms are used throughout the bulletin.

Data source

The primary source of data for most of these statistics is the HESA student record. This is provided to Welsh Government by Jisc, who now own and operate HESA.

A summary of the Student data collection process for 2023 to 2024 covering timescales, validation and business rules and checking processes is included on the HESA website.

Calculation

All counts are based on a count of enrolments as opposed to a headcount of the number of individuals. There are three types of enrolments reported throughout the bulletin: entrant, student, and qualifier. The populations are not mutually exclusive - for example, all entrant enrolments would form part of both the student population and entrant population, and many qualifier enrolments would belong to both the student population and qualifier population.

In cases where an individual is enrolled on more than one course that fulfils the criteria for being included in one of these populations, they would be counted as belonging to the population more than once. For example, an individual who is enrolled on two different courses, with both enrolments fulfilling the criteria for being included in the HESA standard registration population, would be counted as two “students”. See ‘Initial teacher education: Definitions’ for more detailed information on what counts as an entrant enrolment, a student enrolment, and a qualifier enrolment.

Secondary school subject specialisms are measured in Full Person Equivalents (FPE). This means that if a course was focussed on a maths subject specialism half of the time, and physics the other half, it would appear as 0.5 enrolments in a maths specialism and 0.5 enrolments in a physics specialism.

Changes in methodology

From the 2022 to 2023 academic year, the HESA student record moved from the legacy format to the Data Futures model. Due to this change, some of the methodology used differs between years. In some cases, published figures may also differ from previously published releases.

Bilingual teacher training

The Open University currently records all ITE courses as designed to enable individuals to be able to teach in Welsh in the HESA student record. To avoid these figures being over-reported, the Welsh Government obtain information regarding bilingual teacher training directly from the Open University and amend figures annually. In previous ITE statistical releases, individuals were only recorded as training to be able to teach in Welsh if they were returned as such by the Open University for the academic year. This release now treats any individuals identified as training to be able to teach in Welsh throughout the course of their studies as training to be able to teach in Welsh.

Level of study/qualification

Previous ITE statistical releases distinguished between PGCE courses that lead to Qualified Teacher Status (QTS) and other courses that lead to QTS. Postgraduate and PGCE are often used interchangeably, despite subtle differences. The level of study/qualification distinction replaces the PGCE/Other distinction. This change has no substantial impact on the figures reported in this release.

Specification of entrants

For academic years prior to 2022 to 2023, entrants (previously known as first years) are defined by a start date of 1 August (year 1) and 31 July (year 2). For example, for academic year 2021 to 2022, entrants would be those with a start date between 1 August 2021 and 31 July 2022. From the 2022 to 2023 academic year onwards, the start dates for capturing entrants has changed slightly to be between 18 July (year 1) and 17 July (year 2). This difference in definition has no substantial impact on the figures reported in this release.

Subject specialisms

Following the move to the Data Futures model, from 2022 to 2023, the Open University does not record subject specialisms in the HESA student record. To enable subject specialisms to be reported in the ITE bulletin, the Welsh Government obtain information regarding secondary school subject specialisms directly from the Open University to enable the figures to be amended annually.

Rounding and disclosure control

All figures derived from the HESA student record are rounded in accordance with HESA’s standard rounding methodology as follows:

• All numbers are rounded to the nearest multiple of 5
• Any number lower than 2.5 is rounded to 0
• Halves are always rounded upwards (e.g. 2.5 is rounded to 5)
• Percentages based on fewer than 22.5 individuals are suppressed
• Averages based on 7 or fewer individuals are suppressed
• The above apply to headcounts, full-person equivalent (FPE) and full-time equivalent (FTE) data
• Financial data is not rounded

Proportions based on small figures are displayed and calculated based on rounded figures to avoid disclosure and may therefore be misleading.

ITE phase for students

For 2022 to 2023 and 2023 to 2024, some students at the University of South Wales that were recorded as being on ITE courses that led to QTS did not have an ITE phase accreditation recorded. Upon coordinating with the provider, an ITE phase was derived for the affected students using course titles where appropriate. Students who were on a course with no ITE phase accreditation were treated as not on an ITE course that leads to QTS.

ITE phase for qualifiers

Cardiff Metropolitan University returned all ITE qualifiers in 2022 to 2023 as having qualified in an “Other” ITE phase. This meant that ITE qualifiers from Cardiff Metropolitan University were not correctly assigned an ITE phase that matched their QTS qualification. This was corrected by assigning all ITE qualifiers with an “Other” phase for qualification with the same ITE phase as the ITE phase at which they trained to teach in.

University of South Wales incorrectly returned all qualifiers from ITE courses in 2022 to 2023 and 2023 to 2024 without an ITE phase. This meant that no qualifiers from ITE courses were treated as having obtained QTS, which would have resulted in a large underreporting of ITE qualifiers. This was corrected by assigning all ITE qualifiers at University of South Wales a qualifying phase that was the same as the ITE phase which they trained to teach in, and treating all qualifiers from ITE courses as obtaining QTS.

Welsh speaker status

Following the move to the Data Futures model, from 2022 to 2023, there were substantial changes to the Welsh speaking ability data collected in the HESA student record. The quality of Welsh speaker status data was also impacted following the implementation of the Data Futures model. Data quality issues in Welsh speaker status include individuals who have multiple contradictory records for Welsh speaker status and individuals who are recorded as Welsh speakers with an unknown proficiency. Given these issues, Welsh speaker status is reported as those known to be fluent in speaking Welsh, with all other categories, including where Welsh speaker status is not known, treated as not known to be fluent in speaking Welsh. Where individuals have multiple contradictory Welsh speaker status records, they are treated as not known to be fluent in speaking Welsh, even where one of the records indicate that they are fluent Welsh speakers. This means Welsh speaker status figures show the proportion known to be fluent Welsh speakers, and improvements in data completeness can affect year-on-year changes.

Bilingual teacher training

Data quality issues were identified at multiple higher education providers for bilingual teacher training status, across a number of years. Providers appear to have overreported or underreported the number of enrolments on courses that were designed to enable individuals to be able to teach in Welsh. Where issues have been identified, amendments have been made based on the course title and course codes in the HESA student record. However, some of the issues may not have been identified due to the relationship between course titles and bilingual teacher training.

Given these amendments, as well as a change in methodology for the reporting of bilingual teacher training at the Open University, the number of enrolments associated with being able to teach in Welsh medium differs to what was published in previous ITE statistical releases. The differences between data published for academic years prior to 2022 to 2023 in the current and previous releases can be seen in table 1.

[Note 1] Differences are based on rounded figures

Sex

HESA have identified that higher education providers recorded individual’s sex as ‘Information Refused’ and ‘Not Available’ for students previously returned as ‘other’. For this reason, HESA do not deem it appropriate to show these categories separately in the published data and they have therefore been aggregated with ‘other’.

Ethnicity

For 2022 to 2023, there was a large increase in the amount of missing ethnicity data for students whose permanent address is in the UK. To account for this, ethnicity data from 2021 to 2022 was used to impute ethnicity for continuing students at Welsh higher education providers where ethnicity information was missing or unknown for students within coverage, excluding where it had been returned as 'Prefer not to say'.