- 1300 – 13:05 (5 Mins) – Introduction – Sir Bernard Silverman
- 13:05 – 13:10 (5 mins) – Actions Update – Owen Abbott
- 13:10 – 13:30 (20 Mins) – COVID-19 Update – Jon Wroth-Smith
- 13:30 – 14:30 (60 ins) – Impact of Pandemic on LFS – Salah Merad
- 14:30 – 14:40 (10 mins) – Break
- 14:40 – 15:40 (60 mins) – EAP155 – Coverage Estimation for Small Communals – Abu Hossain
- 15:40 – 16:00 (10 Mins) – Any other business – Sir Bernard Silverman
- Sir Bernard Silverman (Chair)
- Dr Oliver Duke-Williams
- Dr Nik Lomax
- Prof David Martin
- Prof Natalie Shlomo
Office for National Statistics
- Owen Abbott (ONS Deputy Lead)
- Cal Ghee (ONS Attendee)
- Gareth Powell (ONS Attendee)
- Jon Wroth-Smith (ONS Attendee & Presenter)
- Selah Merad (Presenter)
- Rich Pereira (Presenter)
- David Freeman (Presenter)
- Abu Hossain (Presenter)
- Rob Turner (Secretariat)
- Owen Abbott – Action 80 – ONS to bring paper on required panel expertise for 2023 recommendation.
1 – Introduction
- The chair welcomed attendees.
2 – Actions Update
- There were no actions to update.
3 – COVID-19 Update
- ONS updated panel on ongoing covid work, in particular looking at how potential easing of restrictions may affect collection and Census Coverage Survey.
4 – EAP156 – Impact of Pandemic on Labour Force Survey
This paper is presented to the panel as it links to, and has implications for, the Population and Migration Statistics transformation project.
- ONS presented work currently underway to account for the effect of the pandemic on the Labour Force Survey (LFS) responses. To compensate for the drop of response rate observed following the introduction of Covid restrictions, weighting adjustments were introduced to the LFS which improved national estimates but introduced issues remain for certain characteristics, in particular country of birth.
- We need to know the actual size of the population and its structure by country of birth since the start of the pandemic to address the quality of estimates by country of birth. The International Passenger Survey (IPS) used to be the main source for migration data but it was suspended in March 2020 because of the pandemic. As part of the transformation of population statistics, we will be using administrative data, including Home Office data, to estimate migration. However, estimates since the start of the pandemic are not yet available.
- This work provides a strategy to correct for this using HMRC Real Time Information (RTI) data on employees to estimate change in the non-UK born population.
- The panel questioned why age and region were not also taken into account in the work. ONS explained that the demand for this work is to better understand the differential growth of UK and non-UK born populations, to bridge the gap until new data becomes available and therefore this was not required. ONS also explained that the RTI data provided so far does not have information on employee age. We have requested a breakdown by age but it’s unlikely that it would be provided because of the resource required.
- The panel questioned a key assumption of the work, namely that the non-UK born populations in and out of work trend in the same ways. Students in particular were raised as a group that would likely have a differing change in the non-UK born population to those in employment. The panel suggested using previous LFS data to determine whether this is the case.
- The panel raised concern with using HMRC data as a comparator for those living in the UK, due to the potential for people to be working abroad for a UK business, particularly following increased home working and travel restrictions.
- The panel discussed the use of country of birth as a proxy for nationality, in particular the effect that acquiring British citizenship would have on nationality figures without a corresponding change in where the population were born.
Please note that EAP156 contains sensitive data from HMRC and a methodological paper has therefore been published instead.
5 – EAP155 – Coverage Estimation for Small Communal Establishments
- ONS presented on the proposed strategy for estimating coverage of small communal establishments (CES), defined as addresses that provide managed residential accommodation with 7- 49 bed spaces. The work focussed on the choice of estimation method, in particular whether to use Dual System Estimation approach (DSE), Logistic Regression (LR) or Mixed Effect Logistic Regression (MLER).
- The panel approved of the approach, agreeing with the ONS conclusion that the MELR and LR strategies performed sufficiently similarly to postpone the decision until live data is available.
- The panel discussed whether over-coverage treatment is also needed. ONS responded that they believe CEs are taken into consideration when accounting for over-coverage across CEs and households but will seek confirmation from team.
6 – Any other business
- The panel discussed the need for a piece of work covering expertise the panel may require for work on the 2023 recommendation.
Action 80 – ONS to bring paper on required panel expertise for 2023 recommendation