Growing Up in England workshop day 1 slides

GUiE Training
Workshop
Dr Polly Vizard
6th
/ 7th
March 2025
Centre for Analysis of Social Exclusion, LSE

This work contains statistical data from ONS which is Crown Copyright. The work is based on the analysis of Wave 1 of the
Growing Up in England (GUIE) dataset [Office for National Statistics] (2021): Secure Research Service Access. The full citation for
this dataset is: Office for National Statistics, released 14 June 2021, ONS SRS Metadata Catalogue, dataset,
Growing Up in England Wave 1, https://doi.org/10.57906/ct4b-k289. The work was produced using statistical data from ONS
which is Crown Copyright. The views expressed are solely those of the authors and do not reflect the views of the Office for
National Statistics or the wider UK Government. The use of the ONS statistical data in this work does not imply the endorsement
of the ONS in relation to the interpretation or analysis of the statistical data. The work uses research datasets which may not
exactly reproduce National Statistics aggregates. The analysis was carried out in the Secure Research Service, part of the Office
for National Statistics.
Data acknowledgement and disclaimer

Introductions, agenda
and approach

The GUiE dataset:
background, rationale and
underlying data sources

Background and rationale
• GUiE was created as a collaboration between the ONS, Department for Education
and ADR UK
• Background included the Children’s Commissioner for England workstream on child
vulnerability which highlighted substantial data gaps for vulnerable groups
• The ‘Data for Children Partnership’ was established to address this concern
• The GUiE linkage was proposed in this context and progressed as a response to the
identified need for a new dataset focusing on vulnerable and highly disadvantaged
children and young people
• Rationale also relates to the objectives of the ONS Inclusive Data Taskforce and the
post-pandemic ‘inclusive data’ agenda

2011 Census
(information
recorded at a single
time point snapshot
- 27th
March 2011)
Waves 1-3
Links the records of
children and young
people from
administrative education
and children’s social care
systems to their 2011
Census records
Administrative
data from the
education
system
(quarterly,
annual)
Administrati
ve data from
the children’s
social care
system
(annual)
Population census data
(systematic recording of whole
population characteristics – 10
yearly)
Administrative data created by
government and public bodies in
the context of routine operations
and service delivery
GROWING UP
IN ENGLAND
(GUiE)

GUiE - ‘value added’ in a nutshell
• Value added of administrative data
 Valuable information recorded in the context of the everyday operation of
government (national, regional, local) and public services
 Large analysis samples
 Relevant vulnerability flags
• Value added of administrative data linkages
 Brings fragmented information together enabling new learning and insights
(e.g. information from education, children’s social care and health systems)
 Helps address missing information (e.g. ethnicity information pooling)
• Value added of population census – administrative data linkages
 Administrative data may not include all relevant information on individual
characteristics and / or relevant ‘vulnerability flags’
 Administrative data often does not record detailed information about household
and family circumstances
 Linking population census data to administrative data can help to address
these challenges

OVERVIEW OF
UNDERLYING DATA
SOURCES

2011 Census for
England and Wales
(information
recorded at a single
time point snapshot
on 27th
March 2011
about the ‘usually
resident’ population
in England and
Wales)
The 2011 Census for England and Wales records detailed
information on individual characteristics and household
circumstances for adults and children resident in England and
Wales.
Gathers
information on:
Age, gender,
postal address,
marital status,
housing,
ethnicity,
country of birth,
religion, health
and disability,
education,
religion,
language,
unpaid care etc.
Building block 1

Building block 2
Administrative data
from the education
system (collected
by educational
providers and held
by DfE or partners)
GUiE Wave 1:
provides a link to
educational
attainment data
and data on
individual
characteristics
(age, gender,
ethnic group)
GUiE Waves 2 and 3:
provides additional
information on
exclusions,
absences, SEND,
FSM, IDACI
[Annual /
quarterly
/ /
monthly –
collected
by
education
providers
and held
by DfE
and
partners]

Building block 3
Administrative
data from the
children’s
social care
system
GUiE Waves 2 and 3:
provides additional
information from the
Children in Need and
Children Looked
After Censuses
[Annual –
collected
by (local
authority)
children’s
social care
services,
returned to
DfE]

STARTING POINT FOR
UNDERSTANDING THE
GUiE WAVE 1 LINKAGE

Feasibility All Education
Dataset for England
(AEDE)

• Pre-processed longitudinal dataset looking across administrative data sources created by DfE and
transferred to ONS in 2017
• Original reason for AEDE being transferred to ONS was to undertake a comparison of highest
educational qualification data and population census data (i.e. it was not specifically designed to
create GUiE)
• AEDE provides information on individual characteristics (age, gender, ethnic group) and
longitudinal information on educational attainment (by level)
Underlying data sources as identified and described in ONS (2023)
• National Pupil Database (Administrative education datastore held by DfE - England). Includes:
 Information on individual characteristics sourced from the School Census, Pupil Referral Unit
Census and Alternative Provision
 Information on attainment originally sourced from awarding bodies (NB see further information on
YPMAD)
• Individual Learner Records (ILR) (data returned by providers of further education (FE) and work-based
learning (WBL) - England). Includes:
• Information on individual characteristics
• Information on learning aims
• Information on attainment
• Higher Education Statistics Authority HESA records (covering students at government funded
What is the feasibility AEDE?

The feasibility AEDE attainment data draws on information from the Young Person’s Matched
Administrative Dataset (YPMAD)
• YPMAD is itself a linked longitudinal dataset pre-processed and matched within DfE. (Conell 2022) states
it includes:
 School census learner and aims data: information on post-16 participation and personal
characteristics of pupils in state schools at age 15.
 Awarding organisation data: used for attainment and participation for learners not included in the
school census or ILR participation data e.g. independent schools).
 Individualised Learner record (ILR): learners, participation and learning providers in the
government funded education and skills sector.
Time windows : the feasibility AEDE includes:
• NPD data for academic years 2001/2-2014/2015
• Individualised Learner Record (ILR) and Higher Education Statistics Agency (HESA) data for academic
years 2002/2003- 2014/15
Age restriction: the feasibility AEDE is restricted to children and young people aged between 14 and 29
years on 31 August 2015 (that is, at the end of academic year 2014/15 - where age refers to age as
recorded in feasibility AEDE records).
What is the feasibility AEDE? (cont.)

Age range of the cohort contained in the feasibility
AEDE (ONS 2022)
• Academic years 2000/2001 to 2014/2015
• Academic year runs 1st
September-31st
August
following year
• Age is observed on 31st
August each year (ie at the
end of each academic year)
• Shows age range of the cohort included within the
feasibility All Education Dataset for England
COHORT AGED 14-29 ON 31ST
AUGUST
2014/15
The feasibility AEDE age restriction is
fundamental to the GUiE linkage
Source: ONS 2022

Source: ONS 2022
The feasibility AEDE age restriction is
fundamental to the GUiE linkage
COHORT AGED 14-29 ON 31ST
AUGUST
2014/15
(EQUIVALENTLY, 10-25 ON 31ST
AUGUST 2011
– age observation closest to 2011 Census
day)
Age range of the cohort contained in the feasibility
AEDE (ONS 2022)
• Academic years 2000/2001 to 2014/2015
• School academic year runs 1st
September-31st
August following year (age is observed on 31st
August each year (i.e. at the end of each academic
year)
• Shows age range of the cohort included within the
feasibility All Education Dataset for England

Relevant previous linkages and differences to be
aware of ….
• Linkage of 2011 Census and HESA data (ONS 2013)
 Relevant linkage methodology
• Initial linkage of 2011 Census and feasibility AEDE (ONS 2019 - Haines, Taylor
and Tietz)
 Experimental linkage of Census 2011 data to feasibility AEDE data including
HESA data
 Analysis compares data on data highest educational qualification for persons
aged 16-24 sourced from the 2011 Census, administrative education records
and the Annual Population Survey
• GUiE Proof of Concept dataset (ONS 2020)
 Proof of concept experimental linkage of Census 2011 data to a bespoke extract
of the feasibility AEDE
 Excludes HESA data and covers 2010/11 to 2014/15 only
 Includes the individual records of all individuals within a child data subject’s
household
 ONS (2020) provides experimental exemplar breakdowns / analysis


GUiE Waves 1-3:
coverage, content and
linkage

• GUiE Wave 1 links a bespoke extract of the feasibility AEDE (excluding the HESA data) to the 2011
Census
• Bespoke extract of the feasibility AEDE
 Includes AEDE covering the period 2001/2-2014/15
 Excludes the higher education (HESA) data
• Individuals are included in the Wave 1 GUiE linkage if:
 They have a record in the bespoke extract of the feasibility AEDE;
 They have a record in the 2011 Census;
 Their record in the bespoke extract of the feasibility AEDE has been successfully matched to their
record in the 2011 Census.
• GUiE Wave 1 cohort and content summary
• The young people in Wave 1 GUiE were aged 10-25 years old on 31st
August 2011 (due to the age
restrictions built into the feasibility AEDE)
• GUiE Wave 1 census data provides a snapshot of individual and household characteristics on
Census day 2011
• GUiE Wave 1 education data enables us to follow educational journeys longitudinally (main
pathways and attainment)
ONS (2022) population description “Individuals aged between 10 and 25 years on 31 August 2011 are
included within the GUiE data if they were enrolled in government funded education, or non-government
funded further education, in England between the academic years 2001/2002 to 2014/2015. However, some
attainment data from independent schools are also present within the National Pupil Database (NPD),
where students have taken regulated qualifications”.
GUiE Wave 1 linkage description

• Information derived from bespoke extract of the AEDE originally sourced from:
 Attributes data: School Census, Pupil Referral Unit Census, Alternative Provision Census
 Attainment data (“AKA YPMAD” - ONS 2020: pre-linked)
 ILR data
• Years covered
 NPD data 2001/2-2014/15
 ILR data 2002/3-2014/15
• Attributes data includes information on age, gender, ethnic group
• Attainment data
 KS4 and KS5
 Typically supports tracking of ages 14-16 and 16-18 pathways and attainment
 Attainment data looks across FE & WBL equivalent qualifications (not restricted to school
qualifications)
 Especially highest level of educational qualifications by age 19 attained by any route (schools,
FE or skills sector)
Education side of the GUiE Wave 1 dataset
GUiE Wave 1 education data: provides insights into 14-16 and 16-18 pathways and attainment,
especially highest level of educational qualification by age 19 by any route

• What is not included:
 Detailed key stage information typically accessed by NPD users
 Information on attainment at younger ages
 School identifiers (NB cannot control for school level effects)
 Coverage of non-state funded students (e.g. private schools) unclear but likely to be limited
(cont.)

MDX Brows
er > Growin
g Up in Engl
and Wave 1
@ 12169
GUiE Wave 1 AEDE NPD attainment file

MDX Brows
er > Growin
g Up in Engl
and Wave 1
@ 12169
GUiE Wave 1 AEDE NPD attainment file
L1OVERALL
L2OVERALL
L3OVERALL

(cont.)
Highest educational attainment by age 19
• This indicator evaluates qualifications attained through any route (school, college or training) and of
any type (GCSE, A level or equivalent vocational or work-based qualifications – FE colleges,
apprenticeships etc.)
• Working definitions
 Level 1 qualifications (one to four GCSEs at any grade) or equivalent vocational or work-based
training equivalent qualifications
 Level 2 qualifications (five or more GCSES grades A* to C) or equivalent vocational or work-based
training equivalent qualifications
 Level 3 qualifications (two or more A levels or equivalent AS) or equivalent vocational or work-
based training equivalent qualifications
• Examples of vocational and work-based qualifications: VRQ, NVQ (VRQ 2, NVQ 2, VRQ 3, NVQ 3 etc),
BTEC or apprenticeships (by level)
• BUT be aware that Level classifications are time sensitive
 Changes over time to GCSEs/AS/A level qualifications (includes A* to C changing to grade 4 or
above, status of AS etc.)
 Equivalent vocational and skills-based qualifications have also evolved over time

GUiE census data file includes:
• Information on individual characteristics of the child data subject recorded in 2011 Census
• Household characteristics of the child data subject recorded in 2011 Census
• Some additional information about the household reference person (e.g. hrp NSSEC)
• Relationship indicator (could in principle be used to identify siblings though quality not yet
established)
What is not included
• Most individual records of other household members (non child data subjects) are NOT included e.g.
mother of data subject.
 Limits usefulness of the GUiE linkage and substantial departure from (researcher) anticipated
delivery given poc analysis but in line with ONS delivery. We hope coverage will expand over
time.
• Census 2011 data does not include a household income variable.
• Census 2011 only provides a snapshot of a household at a single census day time point, but lots can
change:
 Household characteristics such as household single parent status
 Household deprivation characteristics

Census side of the GUiE Wave 1 dataset

MDX Browse
r > Growing
Up in Engla
nd Wave 1
@ 12169
GUiE Wave 1 Census 2011 file

Growing Up in England workshop day 1 slides

[ARCHIVED CONTENT] UK
Government Web Archive -
The National Archives
Variables and classification
s - Office for National Stati
stics
Part 4 derived variables
Established population census household
deprivation indicators (used in ONS analysis)

Consider implications! As
we don’t have individual
level records for all
household members, we
cannot derive this from
scratch with a different
definition!
Established population census household
deprivation indicators (used in ONS analysis)

• Additional education and children’ social care data
 Additional education data (non-attainment outcomes: exclusions, absences, FSM, SEND and some
other additional information derived from the English school census)
 Children’s social care data (CLA and CIN datasets)
• Time window
 Wave 2 provides data for 2010/11-2014/15
 In principle: Wave 3 plugs some gaps in the Wave 2 data and provides historical run of both the
additional education and children’s social care data (2009/10 going back to 2001/2 where data
availability allows)
• NB Please be aware of inaccurate descriptions of CIN/CLA time windows on both RAP project
approval and metadata systems!
 Metadata states “Wave 3 extends coverage of the CIN and CLA data back to 2001/02, where data
are available. If you request access to any of the Wave-2 datasets, you will also get access to the
corresponding Wave-3 data”
 But under ‘years’: specifies start date 2010 and latest data 2023
 These dates are not accurate and in addition there are issues with Wave 3 year delivery and year
Content of GUiE Waves 2 and 3

MDX Browser > Growing Up in England Wave 2 - Exclus
ions, Absences and English School Census @ 12168

MDX Browser > Growing Up in England Wave 2 - Children in Need, Children Looked After @ 41
7

MDX Browser > Growing Up in England Wave 2 - Children in Need, Children Looked After @ 417
We are trying to get this fixed!

FURTHER DETAILS OF
GUiE LINKAGE METHODS

• No common identifier in the 2011 Census and the AEDE so record linking exercise
undertaken by ONS
• Good practice to provide information on linkage methods in research outputs
 Methods for matching Census 2011 and AEDE records have improved over time
 Wave 1 linkage: matching in improved environment using a combination of deterministic
and probabilistic records
• ONS 2022 description
 ‘The Wave 1 linkage includes the records of 7 million children and young people. In total,
7,432,115 feasibility AEDE records were successfully linked to the 2011 Census with 98%
(7,255,360) linked by deterministic linkage methods and the remaining 2% (176,758)
linked by probabilistic linkage methods’
Linkage methodology

GUiE Wave 1
• ONS state that alongside information on % records linked deterministically & probabilistically it is
good practice to cite overall linkage rate as a basic statistic on linkage quality
• To frame their results appropriately, GUiE linkage information researchers would expect / like to
understand:
• Percentage of Census 2011 records successfully matched to an AEDE record
• Percentage of AEDE records successfully matched to a Census record for each academic year
• This information is published in previous 2011 Census AEDE linkages (ONS 2019) but not in
2022 GUiE methodology document
• The feasibility AEDE and YPMAD were created as linked datasets within DfE - there remain substantial
information gaps around their linkage methods / match rates / limitations etc – very little progress
has been made in addressing this over a five year period
Waves 2&3
• Wave 2 and 3 are a ‘join’
• We require clarification of whether residual information (nonmatched cases) are included in both
Wave 2 and 3
GUiE linkage rates – information gaps

Ethnicity (from feasibility AEDE records –
except for combined GTR variable
specified below)
All feasibility
AEDE subset
for linkage
(number)
Feasibility
AEDE –
Census linked
cohort
(number)
Unlinked
feasibility
AEDE
(number)
%
matched
(matching
rate)
White 6168375 4510895 1657480 73.18
Feasibility AEDE any year - Gypsy/Roma 14181 5628 8553 39.78
Feasibility AEDE any year - Irish
Traveller 3804 1472 2332 38.78788
Feasibility AEDE any year - combined
Gypsy/Roma/Irish Traveller 17839 7033 10806 39.51
Pooled ethnicity data (feasibility AEDE
any year or 2011 Census - any
Vizard et al (2023) Table 2: Results of residuals analysis (differential matching rate by ethnic
group)
Source: ONS
For Wave 1 GUiE, residuals analysis produced by ONS and reproduced in
Vizard et al 2023 Table 2 provides some relevant information

What to expect when
you gain access to the
GUiE Wave 1 data

• GUiE is a new linkage and is very much in its infancy
 You may be the first person using the dataset and delivery may not be as expected
 Triangulate as much as possible with published and / or external sources
 Administrative data is used by government and public bodies and there is often published data which
provides a guide to cohort and population sizes (e.g. no of children in an academic year cohort,
percentage attaining Levels 2/3 by age 19, number of children recognized as falling within statutory
categories ‘child in need’, ‘looked after child”
 Other administrative datasets using the same data files (NPD, LEO, ECHILD etc)
 Growing number of data explainers
 There may be gaps in research readiness and therefore it is critical that there is a line of communication
between GUiE researchers and ONS
 Looking backwards, there been good feedback mechanisms were introduced, but going forward this requires
clarification
Things to be aware of ……

• GUiE data arrives in multiple csv files. E.g. for Wave 1:
 AEDE person spine
 AEDE NPD attributes data
 AEDE NPD attainment data
 AEDE ILR data (multiple data files – learner, AIMS etc)
 2011 Census data
 Data spine
• GUiE data files need manipulation before use
 Import multiple files e.g. into STATA
 May need to append files when a data file is split across two or more csv files
 AEDE person spine includes HESA sourced data (drop as not part of Wave 1)
 Will need to merge files together using ADR identifier
 Sometimes different formats – string / numeric
 Wave 1 data files require merging (e.g. census and AEDE attainment / attributes data)
 Merging is also required across GUiE Waves (1-3)
 Remember Wave 1 age restriction when Wave 1 is linked to Waves 2 and 3
 Substantial data cleaning is required (e.g. out of range age data, dates, errors, missing info…)
 Some data is episodes level not individual level and data can be in different formats
 Many cases where there is more than one ADR_ID record per person per academic year
Things to be aware of ……

• There are duplicate cases in the Wave 1 AEDE person spine
ONS undertook a quality assurance exercise using a GUiE sample to explore
whether cases are ‘true’ duplicates or not
 (i.e. to identify “Instances of true duplication, whereby multiple records on the
same unique ID were deemed to relate to different people”)
 Probability based analysis (ADR-IDs compared with names, postcodes, gender
and other basic information to establish likelihood relate to same person)
 ONS re-issued a duplicate spine dataset which provides a flag to identify a case is
a ‘true’ duplicate or not (checks duplicate ADR-IDs against names, postcodes,
gender and other basic information – likelihood test)
 Researchers can use flag to drop true duplicate cases
 But for retained cases where there is more than one ADR_ID record per person
per academic year you will need to decide how to deal with this on a study-by-
study basis
 More on this tomorrow!
Things to be aware of (cont.) ……

• Information in Wave 1 data dictionary is overly basic – description information often
inadequate, lacks binary and categorical response variable information and lack of variable
labels in some of the data
 Census ethnicity information as an example
 GUiE users are hoping the documentation will continue to improve!
MDX Browser > Growing Up in England Wa
ve 1 @ 12169

• Importance of precise cohort selection in study design
 Always important BUT as census data is a snapshot conceptualizing time sequence is
essential
 E.g. using GUiE we can observe household deprivation characteristics on 27th
March 2011
and then observe what happened ‘next’ using education / social care data through careful
cohort selection
 Think through how GUiE Wave 1 design affects the cohorts you select
 Think through implications of GUiE data limitations for cohort selection (e.g. use of
census derived household deprivation variables (education) / lack of individual level
records might be important)
• Terminology and status of GUiE – refer to GUiE as an experimental population census –
administrative data linkage

Examples of early
analysis using GUiE

ONS analysis
• ONS breakdowns using PoC GUiE
ONS (2020) Available at
Educational attainment and household composition, feasibility research and methodology - Office for Nati
onal Statistics
• ONS breakdowns using GUiE Wave 1
ONS (2022) Growing Up in England Data: personal characteristics, general breakdowns, individualised
learner records, household characteristics and educational attainment, vulnerable groups and educational
attainment. Available at: Growing Up in England (GUiE) - Office for National Statistics (ons.gov.uk)
• ONS analysis using GUiE Children’s Social Care data
ONS (2022) Who are the children entering care in England? - Office for National Statistics (ons.gov.uk)
ONS (2022) Methodology: Who are the children entering care in England? - Office for National Statistics
Researcher analysis
• Vizard, Obolenskaya and Dolling (2023)
Inclusive data: better data and evidence to understand the lives, circumstances and needs of children an
d young people from the Gypsy, Traveller and Roma Communities - Casereport149.pdf
• Zhang (forthcoming) time spent in care and educational outcomes
Analysis using GUiE is in it’s infancy ……

Vizard et al (2023) Inclusive data: better data and e
vidence to understand the lives, circumstances and
needs of children and young people from the Gyps
y, Traveller and Roma Communities
• Trialling of ethnicity pooling methodology (pooled GTR identifier derived from
across population census and the various Wave 1 education data files)
• Careful selection of age cohort (taking account of):
 availability of attainment data
 census information being a ‘snapshot’
 census education deprivation specification
• Focus on 14-16 and 16-18 pathways and educational attainment by Level (1-3)
by any route and age 19
• Initial descriptive analysis by detailed individual and household
characteristics (including household deprivation indicators)

Gypsy, Traveller and Roma
young people aged 14 or
15
Other young people 14 or 15
Sex
Male 46.60 50.20
Female 53.40 49.80
Family type
Child in couple family 49.30 70.10
Child in single parent family 46.20 27.60
Other 4.50 2.40
Number of families in the household
One 96.50 98.00
two or more families 3.50 2.00
Country of birth
Born within the UK 77.20 92.00
Born outside the UK 22.80 8.00
Household tenure
Owned(incl.shared ownership) 26.20 66.10
Private rent 45.90 21.00
Social rent 26.00 12.10
Rent free 1.90 0.80
Type of accommodation
House/Flat 89.50 99.90
Caravan/mobile or temp structure 10.50 0.10
Household overcrowded (bedroom standard)
Table 9: Prevalence of multidimensional child disadvantage (children
and young people in Wave 1 GUiE who were age 14 or 15 at the time
of 2011 Census)

Gypsy, Traveller and Roma
young people aged 14 or
15
Other young people 14 or 15
No dependent children N/A N/A
One or two 52.20 71.20
Three or more 47.80 28.80
Deprivation in multiple dimensions simultaneously
Not deprived in any dimension 13.50 50.10
deprived in 1 dimension 28.80 30.30
deprived in 2 dimensions 33.40 14.00
deprived in at least 1 dimension 86.50 49.90
Source: All statistical results remain Crown Copyright. Note: Experimental statistics. The analysis was undertaken in the Secure Research
Service (SRS) by Obolenskaya using Wave 1 GUIE (merged cases only). The full citation for this dataset is: Office for National Statistics,
released 14 June 2021, ONS SRS Metadata Catalogue, dataset, Growing Up in England Wave 1, https://doi.org/10.57906/ct4b-k289. The
individual and household characteristics in this table are derived from information recorded in the 2011 Census. Identification of Gypsy,
Traveller or Roma ethnicity uses combined information from the 2011 Census and the feasibility AEDE.

Childhood deprivation characteristics (household level
deprivation, recorded in the 2011 Census when young people in
the cohort were aged 14 or 15)
Highest educational qualifications at age 19
(recorded in administrative education data for
academic year 2014/15)
% who had not
attained Level 1
% who had not
attained
Level 2
% who had not
attained
Level 3
Gypsy,
Traveller,
Roma
Other
Gypsy,
Traveller,
Roma
Other
Gypsy,
Travelle
r, Roma
Other
All 37.04 6.06 52.57 13.08 86.86 39.88
Male 39.55 7.49 56.39 15.93 89.05 44.7
Female 34.96 4.61 49.41 10.19 85.04 35.01
Household education deprivation 42.03 12.02 59.31 23.62 88.87 59.1
Household employment deprivation 41.75 11.93 60.19 22.84 90.05 56.16
Household health and disability deprivation 38.68 11.31 54.71 20.4 89.58 50.08
Household housing deprivation 42.25 10.13 55.35 20.02 89.84 51.49
Not deprived in any dimension 23.13 2.33 36.88 6.97 75 29.51
Deprived in at least 1 dimension 39.43 9.78 55.23 19.21 89 50.43
Number of deprivation dimensions = 1 33.54 7.3 48.78 15.44 86.89 45.08
Number of deprivation dimensions = 2 42.35 12.34 56.47 23.32 90.59 56.31
Number of deprivation dimensions = 3-4 43.2 16.84 62 29.52 89.6 64.86
Source: All statistical results remain Crown Copyright. Note: Experimental statistics. The analysis was undertaken in the Secure Research Service (SRS) by Obolenskaya using Wave
1 GUIE (merged cases only). The full citation for this dataset is: Office for National Statistics, released 14 June 2021, ONS SRS Metadata Catalogue, dataset,
Growing Up in England Wave 1, https://doi.org/10.57906/ct4b-k289.
Vizard et al (2023) Table 11: The relationship between educational outcomes
by age 19 and multidimensional child disadvantage at ages 14 or 15 (cohort of
young people who turned 15 in academic year 2010/11)

• Become an accredited researcher
 Accredited_Researcher_Application_Guidance_v1.2.docx
 Apply for accreditation through the Research Accreditation Service (RAS) or its replacement (Spring
2025?)
 Attend Safe Researcher Training (SRT) course and pass test (and ID check)
(registering for an account with the Research Accreditation Service (RAS) or
Training and events — UK Data Service)
• Apply for an accredited research project
 To access the Secure Research Service (SRS) or IDS, researchers must submit a research project
application for accreditation in the Research Accreditation Service (RAS) or its replacement (Spring
2025?)
 Assessment of project aims, methodology and data request
 Also assessment of legal basis and whether in ethical and public interest
(Research Project Application Form (DOCX 256KB))
Secure data application process

• Ethics and public good
 Ethics Self-Assessment Tool – UK Statistics Authority
 HEI Research Ethics Committee review
 Some projects are referred for review by the NSDEC (National Statisticians Data Ethics Advisory
Committee) Minutes and Meetings – UK Statistics Authority
 Might make formal recommendations
 Some projects are referred for retrospective audit case studies to check compliance with
NSDEC recommendations
 Guidelines on using the ethics self-assessment process – UK Statistics Authority
 Considering Public Good in Research and Statistics: Ethics Guidance – UK Statistics Authority
• IT arrangements for secure access to data
• Safe room / pod
• Secure organisational connectivity (SOC)
 Organisational / remote (home) access
Secure data application process (cont.)

• GUiE legal gateway
 Section 6 of the GUiE User Manual (under ‘resources’)
 Section 45A of the Statistics and Registration Service Act 2007 (as inserted by the Digital
Economy Act 2017)
 Section 87 of the Education and Skills Act 2008,
 Section 537A of the Education Act 1996
 Section 83 of the Children Act 1989.
• GUiE applications are for Waves (1-3) (i.e. not a variable-by-variable application)
• RAS dropdowns for you to specify GUiE waves and years
 Wave 1 reasonably straight forward but some complexities around Wave 2 and Wave 3 applications
 There is no RAS drop-down option for Wave 3 (available options are Waves 1 and 2 – if you request
Wave 2 you will receive Wave 3)
 Wave 2 education data coverage is specified as 01/01/2002-31/12/2015
 Wave 2 children’s social care data year coverage is specified as 01/01/2010 - 30/11/2023 (this is
inaccurate, however if you request the data you will receive Wave 2 and Wave 3)
 As noted above year coverage on the ONS metadata catalogue for Wave 2/3 children’s social care
data is similarly misdescribed in year coverage section (although narrative description is more
accurate)
 These are in the process of being corrected
Applying for GUiE (Waves 1-3)

• IDS transition (needs to be clarified?)
• SRS as a working environment and SDC (‘safe outputs’)
• Specific SDC controls for GUiE:
• Sub LA analysis
 Issue of school identifier
 IMD / IDACI and levels
• Naming of LAs
• CIN/LCA data enhanced controls
Working in the SRS

Sources of information
and support

General dataset information and links to core documentation and other supporting resources
 SRS Metadata Catalogue Growing Up in England Wave 1
 ADR flagship dataset information gateway Growing Up in England - ADR UK
• GUiE user guide
 Download here (under “resources” option)
 Data dictionaries are available on the ADR UK Data Catalogue / SRS Metadata Catalogue
GUIE Wave 1 Census 2011, AEDE and ILR (under “resources” option)
GUIE Wave 2 - Children in Need, Children Looked After (under “resources” option)
GUIE Wave 2 - Exclusions, Absences and English School Census (under “resources” option)
• Also useful PoC metadata excel is downloadable here:
Educational attainment and household composition, feasibility research and methodology - Office f
or National Statistics
(under “related downloads” option)
• Some further information is available within SRS (e.g. GUiE data quality manual)
• Support (tbc)
• ONS GUiE team ?
• GUiE community webpage (with ONS facilitation and / or support)?
GUiE: general sources of information and
support

Precursors to the GUiE linkage
ONS (2019) Admin-based qualification statistics, feasibility research: England.
ONS (2019) 2011 Census data to attainment data – FOI
Feasibility AEDE
ONS (2023) Feasibility All Education Dataset for England (AEDE)
ONS (2020) Proof of Concept Study
Educational attainment and household composition, feasibility research and methodology
Example data tables
ONS (2022) GUiE linkage documentation and published data tables
Methodology document
Growing Up in England (GUiE): Linking demographic, geographic and household information to the highest
level of educational attainment and vulnerability characteristics using the GUiE dataset
Published data tables
Growing Up in England (GUiE): personal characteristics, Growing Up in England (GUiE): general breakdowns;
Growing Up in England (GUiE): Individualised Learner Record (ILR);
Growing Up in England (GUiE): household characteristics and educational attainment;
GUiE: further information on content, coverage
and linkage methods

Growing Up in England workshop day 1 slides

More Related Content

Similar to Growing Up in England workshop day 1 slides (20)

More from EleanorCollard (11)

Recently uploaded (20)

Growing Up in England workshop day 1 slides