This demonstration contains seven questions illustrating seven question types; MCQ, multiple selection, pairing, rank ordering, inline word selection, multiple choice hotspot and hotspot area selection. As with all QTI v2.1 question types, there is almost unlimited flexibility in how marks can be allocated and how feedback can be provided and some of these questions give an idea of some of the simpler things that can be done in this way.
Demonstration 1 can be found here.
This demonstration contains six questions, illustrating the different graphical interactions supported. Hotspots defined within an image can be selected, ordered, and can each be associated with another hotspot. A selected point can be mapped into one of a number of areas. External images can be inserted into graphical 'gaps' or can be placed with the position being mapped into one of a number of areas.
Demonstration 2 can be found here.
This demonstration contains five questions, illustrating randomisation of both text and numeric values, the ability to check both the accuracy and the precision of a student entered number, and the ability to use a student's response to one part of a question to create an alternative answer to a subsequent part. Whilst rather detached from the real world, students seem to expect to be marked for what they did right even though the overall answer to a problem may be wrong. The penultimate question illustrates the use of alternative number bases, in this case binary, octal and hexadecimal. The final question demonstrates a very simple adaptive style that allows the user to have more than one try at a multiple choice question.
Demonstration 3 can be found here.
These examples show some of the adaptive features that are provided in QTI v2.1. Users can be offered more than one try at a question if they get it wrong, and for particular wrong answers the feedback can be tailored to the error they probably made. The final question shows a style that was developed for teaching the use of a specific algorithm in a financial derivatives course. Alternative routes through the exercise are supported and targeted feedback is used to guide the student. A 'back' button is provided that allows the user to go back one or more steps to help their understanding of the algorithm.
Some of these adaptive examples also show a SToMP feature that is not supported by QTI v2.1, that is the placing of text, including randomised numeric values, into images. This is particularly useful in questions involving electronic circuits, as shown in D4.2, 3, 4 and 6.
Demonstration 4 can be found here.
For more information about the new system, please contact Dick Bacon.
Department of Physics
University of Surrey
09 January 2017