The purpose of this study is to demonstrate how test information function (TIF) in item response theory (IRT) can be used to select informative items from a measurement scale. We provided a brief review of TIF and IRT and used genetics knowledge scale as an example. The data was estimated by Samejima's graded response model and test information function is calculated for each item. The results showed the usefulness of IRT and TIF in item selection and scale development