What is Measurement Decision Theory?
aimeQED uses Measurement Decision Theory (MDT) to estimate the probability that participants are keeping their knowledge current based on their pattern of responses to questions. MDT represents an innovative and transparent way for organizations to continuously evaluate and inform participants on how they are performing.
Using each participant’s pattern of responses, the MDT algorithm calculates the probability that they belong to one of two groups – those who are keeping their knowledge up-to-date or those who are not.
What makes using MDT distinct?
- MDT provides a probability, not a score.
- MDT can be used to reward learning and recall.
- MDT does not grade on the curve. Everyone can remain up to date.
- MDT accounts for the fact that aimeQED's experience is personalized.
- MDT accounts for the difficulty of each question.
An overview of how MDT works.
Specify your initial level of confidence that your participants’ knowledge is current; typically, the starting probability value, or p-value, is high, around 0.97, meaning you are 97% confident that your participant is up-to-date.
With each question answered, the MDT algorithm recalculates this confidence level using the accuracy of your participants’ answers to the questions they received and the difficulty of those questions.
You will also specify the probability threshold below which you would consider a participant not to be keeping their knowledge sufficiently up-to-date and current.