Lancashire and East Cumbria

Opportunities and Challenges of Using Surveys in Health Research

Date: Thursday 11 April 2024, 2.30PM

Location: University of Central Lancashire

University of Central Lancashire
Brook Building BB007

Local Group Meeting

Share this event

UCLan and the Lancashire Clinical Trials Unit are hosting a Royal Statistical Society (Lancashire and Cumbria local group) seminar on the topic of “Opportunities and Challenges of Using Surveys in Health Research”. The event is scheduled for Thursday, 11th April, from 2.30 to 5 pm, and will feature distinguished speakers Michael Baxter, Director of Statistics at Kantar, and Professor Natalie Shlomo from the University of Manchester.

We are pleased to announce that UCLan and Lancashire Clinical Trials Unit will be hosting a Royal Statistical Society seminar (Lancashire and Cumbria local group) on Thursday 11th April 2.30 – 5 pm, on the topic “Opportunities and Challenges of Using Surveys in Health Research”. We welcome Michael Baxter (Director of Statistics, Kantar) and Professor Natalie Shlomo (the University of Manchester). Details of the talks are below.

Michael Baxter (Director of Statistics, Kantar)
Title: "An improved weighting technique for sample surveys"
Abstract: Rim weighting or raking is a widely used method for weighting sample surveys to reduce bias caused by unrepresentative sampling and non-response. I propose a modification of this method, asymmetric rim weighting. It has the advantage that it reduces both the number of very high weights and the sum of the squares of the weights compared with traditional rim weighting. This means that the variance of the sampling error is reduced and there is less need to have a cap on the weights. There is little or no increase in the time taken to perform the weighting compared to the existing method.

Professor Natalie Shlomo (The University of Manchester)
Title: "Theoretical Sampling Design Options for a New Birth Cohort: An Accelerated Longitudinal Design Perspective"
Abstract: The 2017 Longitudinal Studies Strategic Review funded by the ESRC made a series of recommendations on innovative ways to enhance and to invest in opportunities for longitudinal research in the UK. One of the key recommendations in the Review was to consider a new birth cohort using an alternative approach based on an accelerated longitudinal design. The key difference between an accelerated longitudinal design and the usual single cohort study is that an accelerated design incorporates multiple cohort groups at the outset of the study, enabling the duration of the study to be shortened. In this presentation, we investigate theoretical and quantitative aspects of an accelerated longitudinal design versus a traditional single cohort. We compare designs with respect to the precision of different kinds of statistical analyses, sample sizes and a simulation to study the impact on growth curve estimation. We conclude with other relevant survey methodology considerations that are needed for an accelerated longitudinal design.