This article discusses visual techniques for detecting test items that would be optimal to be selected to the final—compilation on the one hand and, on the other hand, to out-select those items that would lower the quality of the compilation. Some classic visual tools are discussed, first, in a practical manner in diagnosing the logical, illogical, and anomalous patterns in item responses, and, second, a new way of illustrating the behavior of the items, cut-off curve, is introduced. All the illustrations in the article are accessible even without specific software packages; they are made by using basic spreadsheet software.
Metsämuuronen, Dr. Jari
"Essentials of Visual Diagnosis of Test Items. Logical, Illogical, and Anomalous Patterns in Tests Items to be Detected,"
Practical Assessment, Research, and Evaluation: Vol. 27, Article 5.
Available at: https://scholarworks.umass.edu/pare/vol27/iss1/5