coe-staff: EMPL Colloquium - Dr. Thomas Cook, Northwestern University Fri, Jan 13, 9:30-10:50am HEDCO 220

Denise E McKenney mckenney at uoregon.edu
Wed Dec 21 13:36:58 PST 2016


**Please note the time and location have been changed to HEDCO 220 beginning at 9:30am.**
-------------------------------------------------------------------------------------------------------------------------------------------------
Educational Methodology, Policy and Leadership (EMPL) is pleased to announce the following colloquium:


Dr. Thomas Cook, Northwestern University presents, "Causation in the Non-Laboratory Sciences: The Practical and Theoretical Significance of Elaborative Design Elements”



When:  Friday, Jan 13th

Presentation: 9:30 – 10:50am

Location:  HEDCO 220

Breakfast will be provided beginning at 9:00am



This talk takes three basic causal identification strategies – random assignment, interrupted time series and regression-discontinuity – and briefly describes how design elements have come to be added to each in order to improve its causal yield and relevance across a wider range of research settings. This description is then used to contrast two approaches to causal inference in non-experiments. The first one, associated with the work of William Cochran and his student Donald Rubin, treats non-experiments as though they were broken-down random assignment experiments and so emphasizes testing sharp hypotheses about a single treatment and comparison group posttest mean difference. Propensity score analysis epitomizes this. The other approach is attributed to Ronald Fisher and emphasizes designing studies to test multiple implications of the same causal hypothesis -- thus to test whether a theoretically coherent pattern of multiple posttest mean differences is observed. Design elaborations of interrupted time series and regression-discontinuity are predicated on a pattern-matching theory of causal inference rather than on single sharp tests. This is even true of some elaborations of the basic random assignment experiment and of most laboratory research in the natural sciences.

Please forward this information to students who might be interested in attending.

Thank you,
EMPL Colloquium Committee

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists-prod.uoregon.edu/pipermail/coe-staff/attachments/20161221/05e3c7c6/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Thomas Cook Flyer.docx
Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document
Size: 72970 bytes
Desc: Thomas Cook Flyer.docx
URL: <http://lists-prod.uoregon.edu/pipermail/coe-staff/attachments/20161221/05e3c7c6/attachment-0001.docx>


More information about the coe-staff mailing list