Using Spreadsheets to Teach Signal Detection Theory
Year: 2014
Type: article
Abstract: Signal detection theory (SDT) is a mathematical framework for evaluating the performance of a detection system, which is broadly applicable across many academic and practical domains, including medicine, psychology, statistics and engineering. In this article, I present an interactive spreadsheet that allows students to systematically explore the function of the two key parameters of SDT: the decision criterion and discriminability. Stimulus distributions and ROC curves update in real time according to user input, thereby providing an intuitive visualization of the SDT parameters. The spreadsheet can be used to find an optimal decision criterion for a given level of discriminability, prior probabilities and payoff matrix. The article includes six problems and suggested answers that instructors may use in the classroom. The article concludes with a discussion of simple adaptions of the spreadsheet for teaching frequentist statistics and testing psychologically based models of detection.
Author: Christopher R Fisher
Institution: Miami University
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W3379699