Standard Errors of Equipercentile Equating for the Common Item Nonequivalent Populations Design

An equating design in which two groups of examinees from slightly different populations are administered different test forms that have a subset of items in common is widely used. A procedure for equipercentile equating under this design has been previously outlined, but standard errors for this rather complex procedure have not been provided. This paper provides these standard errors and a simulation that verifies the equations for large samples. A real data example is provided for considering issues involved in using these procedures.