Posted on Friday, December 18, 2020
This session is jointly sponsored by the Statistics department and the Research
Methods, Measurement, and Evaluation program, University of Connecticut
(UCONN), New England Statistical Society (NESS) and Statistical and Applied
Mathematical Institute (SAMSI) as part of online interdisciplinary seminar
series on statistical methodology for social and behavior research.
Speaker: Bengt Muthen
Date and Time: December 18th at 12:00 EST
Topic: Response Accuracy and Response Time in Cognitive Tests
Abstract: It is an old and still unresolved issue how much a cognitive test
score reflects ability and how much it reflects speed. The well-known
speed-accuracy tradeoff does not make an answer to the question easier. In the
presentation I will report the results of my research steps to investigate the
problem. Briefly summarized, the findings are as follows. First, the
correlation of ability and speed across persons depends on the test. Second,
based on different kinds of modeling and different kinds of data, there seem to
be remaining item-wise dependencies (i.e., conditional dependencies) between
response accuracy and response time after controlling for the underlying latent
variables. Third, the remaining dependencies depend on the difficulties of the
test items and the dependencies also are curvilinear. I will present an
explanation for the findings, and a tentative, complex answer to the old
question what is being measured in a cognitive test.
For inquiry, please contact Dr. Xiaojing Wang at xiaojing.wang@uconn.edu