Welcome to Θ – b, a site dedicated to improving the field of educational and psychological measurement through the sharing of instructional resources, research findings, and statistical tools.
This site is maintained by Tony Albano, Associate Professor in the School of Education at the University of California, Davis. More here.
Measurement, in the fields of education and psychology, is the process of gathering information and making inferences about human attributes like knowledge, achievement, and motivation.
Measurement is concerned with the interaction between person and item, sometimes expressed as Θ – b. Good measurement optimizes this interaction via:
The ideal result is actionable information that supports decisions regarding things like academic abilities, preparation for college, and qualifications for a career.
If you’re liking what you hear, check out my intro measurement book. It’s free online and as a PDF, and will soon be available for print on demand.
University of California (UC) faculty are hoping to un-cancel standardized testing for undergraduate admissions. Some math and STEM faculty circulated a letter earlier this month (2000+ signatures) and social science and humanities faculty started their own letter soon after (800+ signatures). I signed the second letter. My argument for testing – no surprise – goes …
Continue reading “Grade Inflation and the Failed Experiment of Test-Free College Admissions”
I’m teaching a class on item response theory – we started this week – and I realized on day one that I don’t have the classical test theory proof for reliability as correlation written down anywhere, and I couldn’t quite reproduce it on the whiteboard, to my chagrin. Here it is. The primary goal of …
Continue reading “Classical Test Theory and Reliability as Correlation”
Differential item functioning (DIF) with intersectional groups or interaction effects requires lots more statistical comparisons than traditional groupings do. In large-scale, high-stakes testing programs, we usually look at A grouping variable with $k = 5$ levels produces $k – 1$ DIF comparisons per item (each group level compared with the reference level), so the traditional …
Continue reading “Omnibus tests of differential item functioning”
