BEGIN:VCALENDAR
PRODID:-//Columba Systems Ltd//NONSGML CPNG/SpringViewer/ICal Output/3.3-
M3//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20220411T093414Z
DTSTART:20220505T123000Z
DTEND:20220505T150000Z
SUMMARY:Social Statistics Research Event: New Developments in Combining P
robability and Non-probability sampling
UID:{http://www.columbasystems.com/customers/uom/gpp/eventid/}o1jj-l1uipv
xl-gpds9g
DESCRIPTION:Social Statistics Research Event\, University of Manchester\n
\nMay 5th 2022\n\n\nNew Developments in Combining Probability and Non-pr
obability sampling\n\n\nNon-probability sampling is receiving increasing
attention in social research due to the escalating costs of data collec
tion and decrease in response rates. While this has often been presented
in opposition to traditional probability sampling new research is inves
tigating the potential of combining the two approaches to collecting dat
a. In this half day event we will be exploring how combining probability
and non-probability data can help improve causal inference\, population
estimates and reduce costs. \n\n\nSchedule\n\n\nDate: 5th of May 13:30-
16:00 BST (UTC+1 time)\n\nLocation: Whitworth Building Council Chamber\,
University of Manchester and ONLINE\n\nRegistration Form Here: https://
forms.office.com/r/RRbUvRUrpJ\n\n13:30 – 14:00 Coffee\n\n14:00 – 14:40 U
sing Synergies Between Survey Statistics and Causal Inference to Improve
Transportability of Clinical Trials\nSpeaker: Michael R. Elliott\, Univ
ersity of Michigan\n\n14:45 – 15:15 Implementing and Adjusting a Non-pro
bability Web Survey: Experiences of EVENs \nSpeaker: Natalie Shlomo\, Un
iversity of Manchester\n\n15:20 – 16:00 Supplementing Small-Sample Proba
bility Surveys with Nonprobability Surveys to Improve Estimation and Red
uce Costs\nSpeaker: Joseph W. Sakshaug\, Institute for Employment Resear
ch (IAB) & LMU-Munich\, Germany\n\n16:00 – 16:30 Wine reception\n\n\nAbs
tracts\n\n\nUsing Synergies Between Survey Statistics and Causal Inferen
ce to Improve Transportability of Clinical Trials\n\nMichael R. Elliott\
, University of Michigan\n\nRandomized trials have been the gold standar
d for assessing causal effects since its introduction by Fisher in the 1
920s\, since they eliminate both observed and unobserved confounding. Wh
en randomized trials are conducted in human populations\, estimates of c
ausal effects at the population level can still be biased if there is bo
th effect modification and systematic differences between the trial samp
le and the ultimate population of inference with respect to these modifi
ers. Recent advances in the survey statistics literature to improve infe
rence in nonprobability samples by using information from probability sa
mples can provide an avenue for improving population causal inference in
clinical trials when relevant probability samples of the patient popula
tion are available. We review some recent work in ``transporting'' causa
l effect estimates from trials to populations\, and propose a ``doubly r
obust'' estimator of population causal effects that is consistent if eit
her the odds of being the clinical trial versus the population can be co
rrectly estimated\, or if the effect modification of the treatment can b
e correctly estimated. We explore our proposed approach and compare with
some standard existing methods in a simulation study\, and apply it to
a study of pulmonary artery catheterization in critically ill patients w
here we believe differences between the trial sample and the larger popu
lation might impact overall estimates of treatment effects.\n\n\nImpleme
nting and Adjusting a Non-probability Web Survey: Experiences of EVENs (
Survey on the Impact of COVID19 on Ethnic Minorities in the United Kingd
om)\n\nNatalie Shlomo\, Andrea Aparcio-Castro\, Daniel Ellingworth\, Jam
es Nazroo\, Harry Taylor – University of Manchester \n\nNissa Finny – Un
iversity of St. Andrews \n\nAngelo Moretti – Manchester Metropolitan Uni
versity \n\nWe discuss the challenges of implementing and adjusting a la
rge-scale non-probability web survey. For the application\, we focus on
the 2021 Evidence for Equality National Survey (EVENS) which was led by
the Centre on Dynamics of Ethnicity (CoDE) at the University of Manchest
er in the United Kingdom\, in partnership with Ipsos-MORI. The aim was t
o understand the impact of the COVID19 pandemic on ethnic and religious
minority groups in the UK. Standard probability-based surveys\, even wit
h ethnic minority group boosts\, do not have the sample sizes required t
o obtain reliable estimates for small group statistics. We therefore pla
nned and implemented a non-probability web survey of ethnic minority gro
ups to overcome these limitations. We formed partnerships with community
organizations and used innovative recruitment strategies\, including di
gital and social media. Daily monitoring of the data collection against
desired sample sizes and R-indicator calculations allowed the team to fo
cus attention on the recruitment of specific groups in a responsive data
collection mode. We also supplemented the sample with existing members
in both established non-probability and probability-based panels in the
UK. We describe the measures applied to improve the quality of the colle
cted data and the statistical adjustments to correct for selection bias
based on estimating the probability of participation in the non-probabil
ity sample using a probability-based reference sample and calibration. \
n\n\nSupplementing Small-Sample Probability Surveys with Nonprobability
Surveys to Improve Estimation and Reduce Costs\n\nJoseph W. Sakshaug\, I
nstitute for Employment Research (IAB) & LMU-Munich\, Germany\n\nLarge p
robability-based sample surveys can be prohibitively expensive to carry
out. As such\, many survey institutes have shifted away from fielding ex
pensive probability samples in favor of less expensive\, but possibly le
ss accurate\, nonprobability online samples. Instead of abandoning proba
bility sampling (and all its useful properties) entirely\, we propose a
model-based approach that allows for fielding a small probability sample
survey and borrowing information from a parallel (and potentially biase
d) nonprobability sample survey. Through simulations and real-data appli
cations we show that the method reduces the variance and mean-squared er
ror of regression coefficients and predictions\, relative to a probabili
ty-only sample survey. The potential for cost savings is also evident.\n
STATUS:TENTATIVE
TRANSP:TRANSPARENT
CLASS:PUBLIC
LOCATION:Whitworth Building Council Chamber\, University of Manchester an
d ONLINE\, Whitworth Building\, Manchester
END:VEVENT
END:VCALENDAR