Download
Dyslexia Package (10-31-2023)
I have cleared off all the old versions of damon in anticipation of a new version that is in the pipeline. Instead, I'm going to link you to "dys_pack" -- a package demonstrating how I analyze multidimensional dyslexia data with the old decomposition algorithm (damon.psymethods.nous_legacy) wrapped in the first version of the new damon.data utility (very young and fragile). You also get to see how the (tweaked and improved) nous_legacy deals with the real-life problem of measuring dyslexia, though I could only post simulated data. To give you the flavor of our approach, here is the abstract of a paper that I and dyslexia expert Brock Eide, CEO of Neurolearning, recently wrote. The paper is called, Information from Noise: Measuring Dyslexia Risk Using Rasch-like Matrix Factorization with a Procedure for Equating Instruments. Below are links to the software. Just download to your computer, unzip, load the Python dependencies, and run.
Abstract
This study examines the psychometric properties of a screening protocol for dyslexia and demonstrates a special form of matrix factorization called Nous based on the Alternating Least Squares algorithm. Dyslexia presents an intrinsically multidimensional complex of cognitive loads. By building and enforcing a common 6-dimensional space, Nous extracts a multidimensional signal for each person and item from test data that in-creases the Shannon entropy of the dataset while at the same time being constrained to meet the special ob-jectivity requirements of the Rasch model. The resulting Dyslexia Risk Scale (DRS) yields linear equal-interval measures that are comparable regardless of the subset of items taken by the examinee. Each measure and cell estimate is accompanied by an efficiently calculated standard error. By incorporating examinee age into the calibration process, the DRS can be generalized to all age groups to allow the tracking of individual dyslexia risk over time. The methodology was implemented using a 2019 calibration sample of 828 persons age 7 to 82 with varying degrees of dyslexia risk. The analysis yielded high reliability (0.95) and excellent receiver operating characteristics (AUC = 0.96). The analysis is accompanied by a discussion of the information-theoretic proper-ties of matrix factorization.