Department of Statistical Science
Duke University
presents:
A. Philip Dawid
dawid@stats.ucl.ac.uk
University College London, UK
"Who Needs Counter-Factuals?"
Abstract: A popular approach to the framing and answering of causal questions relies on the idea of counter-factuals: outcomes that would have been observed had the world developed differently---e.g., if the patient had received a different treatment. By definition, we can never observe such quantities, nor can we assess empirically the validity of any modelling assumptions we may make about them, even though our conclusions may be sensitive to such assumptions. This seems an unsatisfactory state of affairs. I shall argue that, for making inference about the likely effects of applied causes, any sensible approach can manage without counter-factuals. However, properties of counter-factuals are relevant for inference about the likely causes of observed effects, and then close attention to what can and cannot be supported empirically is needed to qualify the conclusions drawn.
March 25, 1997
4:00 pm - 5:00 pm
116 Old Chem Building Any questions concerning the seminar may be addressed to Cheryl McGhee @ [919] 684-8029 or e-mail cheryl@stat.duke.edu. Please contact the author(s) directly for reprints etc.