Defining Constructs with Context and Claims
–
Speakers
RA
Russell Almond, PhD – Associate Professor of Measurement and Statistics, Florida State University
About the Talk
In the human sciences, there is often a need to measure properties of a person which cannot be directly observed. Constructs are artificial variables which measure the ability or tendency of someone to perform certain actions in certain contexts. Using examples from education, psychology and exercise physiology, award-winning researcher and statistician Dr. Russell Almond will illustrate procedures for measuring real-life constructs, including practices that can advance how tomorrow's researchers work to define, measure and improve brain health.
A context consists of a subject attempting a task in a certain environment. Constructs are defined through observations (made during or after the attempt at the task) and claims (predictions about what the subject will do in a future attempt). Both claims and observations can be placed on a concept map. The public or theoretical definition of the construct is based on the claims. The local or operational definition is based on the actual collection of observations. The central question of validity of procedure for measuring a construct is related to the alignment between the observations and claims. This principle will be illustrated with a number of examples from education, psychology and exercise physiology.
Register for virtual to receive updates on talks and online access to our full virtual season.
Except where otherwise noted, you can also join us in person at Center for BrainHealth.
No registration is required at this time. Lunch begins at 11:30am and talks at noon.
For more information, contact brainhealthevents@utdallas.edu or visit our events FAQs.
Dr. Russell Almond is an associate professor of Measurement and Statistics in the Department of Educational Psychology and Learning Systems at Florida State University. He received his undergraduate degree from the California Institute of Technology (Caltech) and his PhD in statistics from Harvard University. Dr. Almond's research interests focus around of the question of how to gather, track and monitor evidence of student growth using both traditional sources (e.g., assessments and homework) and non-traditional sources (e.g., simulations and games), especially in situations where multiple dimensions of student proficiency are considered. Together with R. Mislevy and L. Steinberg, he was a part of the effort to create Evidence Centered Design (ECD) — a process for designing assessments based on explicating the underlying evidentiary arguments that support their claims. This work won the National Council on Educational Measurement Award for Outstanding Scientific or Technical Contribution to Educational Measurement in 2000. He co-authored a monograph on building Bayesian networks using ECD, Bayesian Networks in Educational Assessment (Springer 2015).
We are located at 2200 West Mockingbird Lane Dallas, TX 75235
The entrance to the surface parking lot is on Forest Park Road. The talks will be held in the Brain Performance Institute building, the second building when you enter through the gate.