Innovative approaches to methodological challenges facing ageing cohort studies
Innovative solutions to methodological challenges associated with ageing cohort studies
Day 1: Wednesday 25 July 2012
MANDEC, 3rd Floor, University Dental Hospital, Higher Cambridge Street, Manchester, M15 6FH
Day 2: Thursday 26 July 2012
Room: Room 2.88, Humanities Bridgeford Street Building
This conference brings together key themes explored in the previous NCRM workshops on the methodological challenges to ageing cohort studies http://www.methods.manchester.ac.uk/ageingcohort/). The focus of this conference will be innovative solutions that have been developed by researchers to deal with problems associated with ageing cohort studies. In particular, some solutions to challenges related to attrition, missingness, comparability across cohorts and over time, and data collection will be highlighted over two days. The first day will feature presentations and where appropriate, a demonstration of syntax programming for statistical packages. The second day will feature practical sessions in a computer lab using statistical software packages.
Day 1: Presentations: 11am- 4pm
James Banks (IFS and University of Manchester): Attrition and Health in Ageing Studies: Evidence from ELSA and HRS
This presents an investigation into observable characteristics associated with attrition in the English Longitudinal Study of Ageing (ELSA) and the Health and Retirement Study (HRS), with a particular focus on whether attrition is systematically related to health outcomes and socioeconomic status.
Ian Plewis (University of Manchester): Using auxiliary variables to adjust for longitudinal missingness
Different sorts of auxiliary variables – variables measured at previous waves, frame variables and paradata - can be used to improve the accuracy of response propensity models, and to enhance adjustments for missing longitudinal data. These variables can be used when constructing iterative probability weights, carrying out multiple imputations, and specifying models that jointly model a substantive process and the missingness mechanism. Data from the first two waves of the UK Millennium Cohort Study are used to illustrate the potential value of auxiliary variables.
Rebecca Hardy (NSHD and UCL): Statistical issues in meta-analysing multiple ageing cohort studies
This presentation examines recent publications from the HALCYON study team that have overcome methodological challenges to meta-analysis of observational cohort studies. In particular, challenges on data comparability across cohorts and over time will be highlighted.
Natasha Wood & Carli Lessof (Natcen): Overcoming problems associated with retrospective data collection in cohort studies
This presentation looks at the development of the lifegrid for retrospective data collection and its CAPI implementation in the ELSA life history interview.
Day 2: Computer based practical workshops: 11am- 4pm
Rebecca Landy (QMW): A simulation of 1946 birth cohort data under three missing data mechanisms (MCAR, MAR and MNAR), using three missing data methods - complete case, multiple imputation and Heckman selection. This session takes the student through the STATA syntax used when applying these missing data methods.
Gopalakrishnan Netuveli and Morten Wahrendorf (ICLS and Imperial College)
The lifegrid data structure for ELSA and the Survey for Health and Retirement in Europe (SHARE) will be described, highlighting similarities and differences. This will be followed by a practical demonstration of handling lifegrid data based on ELSA data using STATA.