1 / 11

Nicola Massarelli – Eurostat (nicola.massarelli@ec.europa.eu)

Workshop on Labour Force Survey Methodology, Paris, 15-16 April 2010 Reviewing the LFS precision requirements. Nicola Massarelli – Eurostat (nicola.massarelli@ec.europa.eu). References and reminder. References:

wardah
Download Presentation

Nicola Massarelli – Eurostat (nicola.massarelli@ec.europa.eu)

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. Workshop on Labour Force Survey Methodology, Paris, 15-16 April 2010Reviewing the LFS precision requirements Nicola Massarelli – Eurostat (nicola.massarelli@ec.europa.eu)

  2. References and reminder • References: • Dedicated space on CIRCA to share documents:http://circa.europa.eu/Members/irc/dsis/employ/library?l=/quality_lfs_follow&vm=detailed&sb=Title • Functional mailbox: ESTAT-LFS-QUALITY@ec-europa.eu • Support group • Reminder: National Action Plans (1st draft) by end of April Reviewing the LFS precision requirements

  3. Reviewing the LFS precision requirements: at the roots of the issue • Legal text for LFS precision requirements to be clarified • Agreement on the assessment of compliance to be reached • Convergence on variance estimation (methods and/or tools) • Increasing demand for flow estimates Reviewing the LFS precision requirements

  4. Recommendations from the Quality TF 1. to 3.: Creation of a group of experts, coordinated by Eurostat (R1) 4.: Development of common methodological tools to produce comparable flow estimates (R7) • Preliminary step: reflect on the desirable precision Reviewing the LFS precision requirements

  5. Eurostat’s approach • Directors of Methodology (DIME) TF • general principles/recommendations for the formulation of precision requirements and for the computation of sampling errors for household surveys in general • Group of experts on LFS precision requirements • specific precision requirements for the core LFS • shared method(s) for variance estimation (if feasible) Reviewing the LFS precision requirements

  6. Main conclusions from the DIME TF (1) • Effective sample size vs. precision thresholds Preference to precision requirements for the quality of the output  precision thresholds to be preferred to effective sample size • Precision measures • Relative precision measures (e.g. CV) recommended for totals, means (other than proportions), ratios • Absolute precision measures (e.g. SE) recommended for proportions/percentages and for estimates of change Reviewing the LFS precision requirements

  7. Main conclusions from the DIME TF (2) • Implications for NSIs If precision thresholds are referred to the output, the sample size should in principle be adapted to the changing value of the estimate To avoid this, it is recommended to follow a conservative approach when determining the sample size Reviewing the LFS precision requirements

  8. Main conclusions from the DIME TF (3) • Compliance monitoring • Should target both, bad designs and bad conduct of the survey (e.g. high non-response) • Should not lead to a continuous redesign of the system • Should focus on the actual precision of the survey results, i.e. each time considering the actual design effect • Methods and tools for variance estimation • No single variance estimation method suitable for all sampling designs and statistics. • Possible alternative methods are: • Analytical methods • Taylor linearization • Replication methods (e.g. Jackknife or Balanced Repeated Replications (BRR)). Reviewing the LFS precision requirements

  9. Main conclusions from the DIME TF (4) • Possible approaches for variance estimation • Decentralized approach i.e. sampling errors are computed by NSIs according to agreed standards and transmitted to Eurostat • Fully centralized approach i.e. Eurostat computes the sampling errors on the basis of an agreed methodology (e.g. Jackknife), by using information on each national sampling design transmitted by NSIs Reviewing the LFS precision requirements

  10. Points for discussion • Precision requirements: • should refer to the output, or • serve as guidelines for design only? • From MS points of view, to compute sampling errors is it preferable: • that each NSI does it on its own, according to common methods and/or tools, or • to let Eurostat do that, by providing it with the needed additional information? Reviewing the LFS precision requirements

  11. Thank you for your comments Reviewing the LFS precision requirements

More Related