The Test Information Function (TIF) is related to the reliability of a test as well as the precision of the scores. While reliability is a central concept in Classical Test Theory (CTT), it does not fit well in the realm of Item Response Theory (IRT). The reason is that the uncertainty of estimates is conceptualized differently in IRT compared to CTT.

In IRT, the main interest is estimating the value of one or more latent traits, using observed responses to a set of questions. The question parameters can be used to calculate the total information of the set of items in determining the latent trait.

Table 1 shows the average TIF in the range [1, 10] on the Standard scale used in Alva's personality test, along with the Standard Error of Measurement (SEM). TIF is calculated using Fisher information, and SEM is calculated from TIF using the following formula:

SEM = 1 / sqrt(TIF)

According to EFPA Standards, a TIF above 10 is deemed 'Excellent' and between 5 and 10 'Good'. All facets in Alva's personality test are 'Good'.


Did this answer your question?