1 / 25

Family Resources Survey

Family Resources Survey. Data Linking Jo Cockerham. Overview. Background Uses of linked data Development of consent question Methodology Match rates Results from linked 2006/07 data Future projects Questions?. The Family Resources Survey. Launched in 1992 by DWP

fagan
Download Presentation

Family Resources Survey

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Family Resources Survey Data Linking Jo Cockerham

  2. Overview • Background • Uses of linked data • Development of consent question • Methodology • Match rates • Results from linked 2006/07 data • Future projects • Questions?

  3. The Family Resources Survey • Launched in 1992 by DWP • 26,000 private households in UK (about 24,000 in GB) • Detailed information on incomes and benefit receipt, tenure and housing costs, savings • Fieldwork carried out by ONS and NatCen

  4. Background to data linking work • 2004 Strategic Review of FRS • Problems with take-up statistics • Improvements to administrative data • New FRS contract from April 2006

  5. Intended uses of linked data • Statistical and research purposes only • Improve the quality of FRS data • Longitudinal analyses – tracking how different groups move in and out of work and how their situation changes over time • Initially to only be made available internally at DWP and to selected HMRC analysts • Will not be used for operational purposes, such as fraud detection

  6. Informed consent • Requires informed consent of respondent (Data Protection Act 1998) • Personal details need to be passed to DWP for linking (name, address, sex, date of birth – and NINO pre 2008) • Pilot study took place in 2006 • Developed consent question which was introduced in November 2006

  7. Features of 2006 consent question • Asked at end of questionnaire • Separate block to collect full name, address, NINO, date of birth • Written consent forms • Detailed wording • Proxy consent packs

  8. Consent in 2006/07 FRS • Consent lower than anticipated • 40 - 45 per cent for personal interviews • Approx 35 per cent including proxies • Known biases: • Consent rate lower among ethnic minorities • Consent falls slightly as age increases • Employees have higher consent than self-employed

  9. Development of new consent question • Question suspended from August 2007 • Resources diverted to development of improved question • Qualitative pilot October 2007 • Quantitative pilot in January 2008

  10. Qualitative pilot • 30 in-depth interviews with respondents: split into 3 samples • Concluded that question needed to be simplified, more informal and required further clarification in the wording • Interviewer focus groups • Findings consistent with respondent interviews

  11. Quantitative pilot Conducted in January 2008 main stage sample (1900 individuals) To test: • Achieved consent rate • Simplified version of the question • Removal of paper consent forms • Improved survey materials • Removal of NINO/collection of personal details as part of main questionnaire

  12. 2008 pilot results • Consent rate rose to 62% • No bias between sub-groups • Leaflet received positive response • No difference between DWP and ONS consent • New question introduced from April 2008

  13. Administrative data held by DWP • Despite low consent rate, useful analyses can be carried out. • The FRS has been linked to the Work and Pensions Longitudinal Study (WPLS). • 500 million lines of data covering: • benefit claims • employment spells • annual earnings • savings • tax credits • pensions • operational data on customers activities (e.g. participation in back to work programmes).

  14. FRS DATA F R S I D Personal Details FRS DATA – Dataset 1 F R S I D F R S I D Personal Details – Dataset 2 Imputation, editing and DV creation on full FRS. F R S I D Personal Details O R C I D FRS DATA - full release to users as in previous years Forward consenting cases to Data Matching team in DWP F R S I D O R C I D FRS DATA – for those giving consent to link F R S I D O R C I D FRS DATA O R C I D FRS DATA O R C I D WPLS

  15. Matching methodology • “Traffic lights” system • Staged approach by NINO, then surname (soundex), initial of forename, DoB, gender and postcode sector

  16. Matching methodology • Where match for NINO is not available, fuzzy matching by surname (soundex), initial of forename, DoB, gender and postcode sector

  17. Matching Rates 2006/07 data

  18. Results from Linked Data: Savings • Savings/assets data on FRS criticised as unreliable i.e. underestimates people’s savings. • This can impact on high profile National Statistics. For example, figures on Pension Credit Take-Up. • Work carried out to assess the level of any under-reporting, compared the FRS to the HMRC data. • Several caveats: • sample size small • HMRC data covers fewer savings products than the FRS • HMRC data only available to 2004/5.

  19. Savings Data • Table below shows the comparison of the FRS measure with the FRS/HMRC measure of total capital • HMRC figure is calculated using combination of FRS assets plus HMRC assets • Where an HMRC account exists, they have higher/larger amounts in them Note: These figures are unpublished and should not be reproduced or quoted

  20. Comparison of benefits • Only compared benefit spells which were live at the time of interview. • 10 key benefits were examined.

  21. Numbersclaiming benefits

  22. Comparison of WPLS with FRS

  23. Comparison of Benefit Amounts Note: These figures are unpublished and should not be reproduced or quoted

  24. Future Project Proposals • Investigating how benefits analysis may improve FRS validation • Rematch the data without using NINO to compare the quality of the match with/without NINO • Investigating how benefit mis-reporting affects total household income • Linking FRS earnings data to investigate how people are living on reported zero or low incomes • Comparison of FRS employment outcomes to data derived from P45 information

  25. Questions ????

More Related