--- title: "IRT linking, DIF, DTF, and invariance evidence" output: rmarkdown::html_vignette vignette: > %\VignetteIndexEntry{IRT linking, DIF, DTF, and invariance evidence} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8} --- ```{r setup, include=FALSE} knitr::opts_chunk$set(collapse = TRUE, comment = "#>") library(eyeprocess) ``` Scale linking and invariance are treated as evidence components rather than binary declarations of validity. ```{r} ref <- data.frame(item_id=paste0("I",1:8),a=seq(.8,1.5,length.out=8),b=seq(-1.5,1.5,length.out=8),c=0,d=1) A <- 1.2; B <- -.3 foc <- ref; foc$a <- ref$a * A; foc$b <- (ref$b - B) / A link <- eyeprocess_irt_mean_sigma_link(ref, foc) link eyeprocess_irt_apply_link(foc, link) ``` Native utilities include mean-sigma, mean-mean, Stocking-Lord, and Haebara objectives, anchor screening/purification, DIF/DTF effect curves, anchor-set stability, session/device drift, and descriptive process-channel concordance. For mature equating workflows, `equateIRT` remains an explicit optional engine rather than being replaced by a simpler estimator. Primary package source: .