This three-part mini-series of webinars covers the essential topics of reliability and validity in diagnostic scales. The first section sets out why reliability and validity are important - particularly in making a diagnosis - and what the differences between the two concepts are. Then, reliability is further defined and examined in detail, including how to calculate and interpret key reliability measures such as Kappa and Cronbach’s statistics, as well as inter-class correlation. Next, validity is examined in detail, including the concepts of sensitivity, specificity, likelihood ratios and predictive values, in addition to ROC curves. Part three of the webinar series covers the calculation of reliability and validity in the Stata software package.
Click on the thumbnail images below to stream the recordings of the webinars in three parts (links to YouTube). Parts 1 and 2 cover the webinar slide content (below) while part 3 focusses on the use of Stata software for validation. Updated slides to accompany the talks are available below.
Part 1: Introduction and reliability
Part 2: Validity
Part 3: Using Stata for validation
Course materials (click on each item to open or download a PDF):