*Learning more from the practice of educational trials - some
methodological developments*

Monday 23 September 2019, 11:00am - 01:00pm (followed by lunch)

Royal Statistical Society, 12 Errol St, London EC1Y 8LX
The enthusiasm for developing evidence for educational interventions using
randomised trials has not been matched by specific methodology development.
Education has different practical dimensions - the entire school system
could be characterised as a universal intervention in child development.
Contamination between control and experimental allocations, efficient
clustering and multiple arm design, assessment burden and attrition from
control groups, and difficulty measuring primary outcomes precisely require
attention. This meeting considers some of the few innovations in
methodology to understand the generalisation of results and to set a
research agenda for further work.

The meeting follows up the Social Statistics Section meeting in 2015 which
identified the need to power trials to understand their impact on
disadvantaged groups. It is also linked to the afternoon meeting reviewing
the experience of 100 years of trials in education (separate registration
required). This event will be of a more technical nature, aimed at
researchers interested in education trial development, and in understanding
social inequalities in policy experiments.

Chair: Harvey Goldstein (University of Bristol)


Zhi Min Xiao (University of Exeter) -  "The case against Perfection in the
Mean: Pupil Advantage Index as a step towards Personalised Intervention"

Ben Weidmann (Harvard University & Education Endowment Foundation) -
"Lurking inferential monsters? Quantifying selection bias in
non-experimental evaluations of school programs"

Mike Adkins (University of Nottingham) Title TBC

Hugues Lortie-Forgues (University of York) Title TBC

*Event Fees:*

Concessionary Fellow:£15
CStat/GradStat: £18
Fellows: £20
None of the above: £30

Registration is available here
the meeting will be followed by a free event
from 2pm - 6:30pm in partnership with NFER. Separate registration is
required for each as the first has a fee and the second is nearly at

Please get in touch with any questions
Tom King

Hon. Sec. RSS SIG on Data Ethics:
Chair, RSS Working Group on Statistics and School Governance
RSS Social Statistics Section committee

Personal Blog public statistic <>
My Twitter @acertaintom <>

You may leave the list at any time by sending the command

SIGNOFF allstat

to [log in to unmask], leaving the subject line blank.