We designed Concerto so that users could build state-of-the-art computer-adaptive tests without writing a single line of code. That is possible using our in-built flowchart test builder, which lets you string together nodes with different functions to visually script your test logic. Everything you might need can be achieved with the flowchart, from dichotomous and polytomous CATs to user authentication to feedback report generation. For those already familiar with the R language, it is easy to integrate custom R code alongside flowchart elements to create easy-to-maintain assessment applications.

Customise the functionality of your CAT using simple drop-downs that allow you to choose from the most popular item-response theory (IRT) models. The Rasch Model, Rating Scale Model (RSM), Graded Response Model (GRM), Modified Graded Response Model (MGRM), Partial Credit Model (PCM), Generalised Partial Credit Model (GPCM) or Nominal Response Model (NRM) are all ready to use when building Concerto tests, which should meet the needs of researchers and practitioners alike.

Your CAT can also use any next item criterion you decide upon - Maximum Fisher Information, Urry’s Rule or even random selection. During the CAT, the test-taker’s ability is automatically estimated after completing an item and you can define which estimator to use, including maximum likelihood (ML), bayes modal (BM), weighted likelihood (WL) or expected a posterior (EAP). It's also easy to implement content balancing and randomesque features to manage item exposure in larger assessments.

Lastly, we understand that different applications and assessment contexts require different stopping rules. Choose to limit the total test time, the time allowed per item, the number of items shown or the minimum accuracy of measurement.

R is one of the most popular languages for statistics and machine-learning, and users that are already familiar with R will find they can do even more with Concerto than the in-built nodes already enable. Concerto can be used to put your R-based tools into production, breaking the boundaries of traditional test development platforms and enabling you to do just about anything you can imagine. For psychometricians, just install the R packages you need via the admin panel (e.g. MIRT or MCAT) and build your own CAT from scratch. For data scientists, you can implement your R-based algorithms into a web app and deploy it in the real world without needing external IT support.