It is important to ensure that new forms are built to the same content and statistical specifications as the base form. But building statistically parallel forms can be challenging. The notion that test information should be maximized at the cut score for may not always be the case. Whether there are practical differences between maximizing information at the cut score and maintaining the psychometric properties of the base form under the Rasch model remains unclear. This session will provide attendees with important considerations for assembling forms. Results from simulated examples will illustrate the impact of test length, form difficulty, and candidate ability on pass rates, conditional reliability and decision quality indices, and test level functions at the cut score. Whether you oversee a credentialing program, assemble test forms, or just want to learn form assembly, this session will provide general guidelines for form assembly and address problems that may be encountered.