Comparisons among Designs for Equating Mixed‐Format Tests in Large‐Scale Assessments