Setting and Referring to Technical Constraints

Technical constraints are constraints on statistical item properties that take real number values. Psychometricians can set constraints in the Technical Constraints panel, and the Content Development team can refer to these constraints in order to make forms meet them using the instructions in Adding and Removing Items.

To get started, click to expand the Technical Constraints panel.

The panel contains a table of the constraints present, followed by some graphs and sometimes other tables that display technical information about the current form.

Constraints can be on single values, theta statistics, or proficiency levels. Table 6 lists the types of theta statistics that appear in the graphs and can be constrained. You can view and constrain both difference (diff) and non-difference theta statistics. The non-diff statistics represent statistical calculations derived from the current form, and the diff statistics compare current form items to reference form items.

Table 6: Types of Theta Statistics

Type	Description
Test Characteristic Curve (TCC)	A Test Characteristic Curve shows the relationship between the ability score estimated by the model (IRT3pl, IRT3pln, IRTPCL, IRTGPC, IRTGRL, IRTComposite) and the expected score, which is calculated differently for each model.
Standard Error of Measurement (SE or SEM)	Standard error of measurement is the amount of uncertainty expected for a given set of items and theta level.
Information	Test Information Function (TIF) is the sum of the Item Information Function (IIF) for all items. The IIF describes how well or precisely an item measures at each level of the trait that is being measured by a given test (theta). In item response theory (IRT), the term “information” is used to describe reliability or precision of an item or a whole instrument.