Th The Integrated Household I t t d H h ld Survey: The next instalment Alessio Fiacco Office for National Statistics (ONS) UK Office for National Statistics (ONS), UK
Outline • Background • Implementation plan • Survey instrument strategy • Future developments
Introduction • ONS Integrated Household Survey (IHS) IHS components IHS components IHS Topics IHS Topics Labour Force Survey Work (WRK) • • (LFS) (LFS) General Household • Lifestyles (GLF) • Survey (GHS) Expenditure and Food • Living Costs and Food • Survey (EFS) (LCF) National Statistics National Statistics • • Opinions (OPN) • Omnibus Survey (OMN) English House Conditions • Housing (HSG) • Survey (EHC) / Survey of English Housing (SEH)
IHS implementation plan WRK ( Wave 2 ) WRK/HSG Sept-Oct. 07 Apr. 08 Unclustered IHS sample Disability Survey GLF/LCF/OPN Oct. 08 of GB January 08 Pilot 2 Apr. 09 June-Jul. 07 2007 2009 2008 We are here
Survey instrument strategy • Pilot 1 (Feb.-Mar. 06): � Single Blaise instrument � Single Blaise instrument • Pilot 2 (Jun.-Jul. 07): � Multiple Blaise instruments � Multiple Blaise instruments • Common code blocks (the ‘Core’) – Census-type socio-demographic variables Census type socio demographic variables – New statistical outputs – Avoiding duplication of effort g – Obtaining consistency of data • Mode Library • Question-by-Question Help files
Issues and constraints • Field management – monitoring response monitoring response • Business organisation • Legacy/structural factors • Legacy/structural factors – primary key fields • Time series • Time series – Continue to produce current outputs • Data processing Data processing
Data processing WRK GLF LCF OPN Survey Continuous Monthly Monthly Monthly cycle cycle (weekly) (weekly) Monthly, Outputs Annual Monthly, Monthly q quarterly, y, quarterly quarterly annual � Northern � � � Ireland Ireland (GB only) (NI) data Derived Blaise and Blaise and Manipula p Blaise and Variables SIR SPSS SPSS (DVs)
Key considerations • Data transfer system / Download times Data transfer system / Download times – Context – One big download vs five smaller ones One big download vs. five smaller ones – Staggered transmission • Handling of errors – Opinion survey stream y
Handling of errors Blaise instrument approach Issue Single IHS Multiple IHS instrument instruments - Error in the Rescatter once Rescatter five times common Core (impact: every (impact: every section ti stream affected) stream affected) - Error in the Rescatter once Rescatter once survey-specific (impact: every (impact: one stream question blocks question blocks stream affected) stream affected) affected) affected)
Future developments • New English Housing survey – April 2008 April 2008 • New Longitudinal Disability Survey of GB – fieldwork starts in April 2009 fieldwork starts in April 2009 • Unclustered IHS sample design • Survey Case Management System • Survey Case Management System • Core IHS team – Version control and managing change – Version control and managing change
Conclusions • The IHS is still evolving � Five datamodels built around a common Core � Five datamodels built around a common Core • End-to-end processing � Constraints and legacy systems � Constraints and legacy systems • Minimising the impact of change � Phased-implementation instead of ‘big bang’ approach � Phased implementation instead of big bang approach � Integrity of key time series • Longer term: Harmonisation of topic-specific o ge te a o sat o o top c spec c questions • New surveys are brought into scope of the IHS y g p
Any questions?
Recommend
More recommend