Table 2.

Final psychosis risk calculator using ridge regression and the top six predictors

Predictor	Coef.
Intercept	2.590
Children’s Global Assessment Scale (CGAS)	−0.036
SOPS N4 Experience of emotions and self	0.242
PS-R 2: I think that I might be able to predict the future	0.149
PS-R 3: could possibly be something controlling my thoughts or actions	0.130
PS-R 12: I have been concerned that I might be ‘going crazy’	0.169
Percent married in neighborhood	−1.882

Note. SOPS = Scale of Prodromal Symptoms; PS-R (aka ‘PRIME’) = Prevention through Risk Identification, Management, and Education Screen-Revised; Coef = coefficient; SCR = screen; final result will be in log-odds units, which can be converted to probability by exponentiating (to convert from log-odds to odds) and then using the equation probability = odds/(odds + 1); p values and standard errors are not given because they are not meaningful for this type of model (Goeman, Meijer, & Chaturvedi, 2018) due to downwardly biased coefficients (typical rules of general linear model, where the equation is the best linear unbiased estimator, do not apply).