Contains methods for observed-score linking and equating under the single-group, equivalent-groups, and nonequivalent-groups with anchor test(s) designs. Equating types include identity, mean, linear, general linear, equipercentile, circle-arc, and composites of these. Equating methods include synthetic, nominal weights, Tucker, Levine observed score, Levine true score, Braun/Holland, frequency estimation, and chained equating. Plotting and summary methods, and methods for multivariate presmoothing and bootstrap error estimation are also provided.
bootstrap function can now extract presmoothing arguments, if present, to run presmoothing once for all functions requesting it. Note that, when requested, presmoothed score distributions will be used for all methods.
New fit statistics for loglinear presmoothing model comparison: likelihood ratio chi-square, Freeman-Tukey chi-square, consistent AIC, Cressie-Read, and Goodman.
Fix plot axes in plot.freqtab where scales do not match observed score range.
Legend text fixed for single-group and equivalent-groups designs in plot.equate and plot.bootstrap.
Algorithm for default model comparison in loglinear presmoothing was fixed to address problem with degrees of 10 or larger.
Bug in Levine true score equating, introduced in 2.0.5, was fixed.