You may be wondering, how does Firecracker pick my clinical vignette for the new Daily CV feature?
The answer is that Firecracker uses a predictive algorithm for your chance to answer any given CV correctly. Using this algorithm, we choose questions that are appropriately challenging - the CV shouldn't be too easy and it shouldn't be impossible. Here's how the algorithm works:
First: Each Firecracker CV has a primary topic association. Your Daily CV will only come from the pool of CV's that you haven't seen yet and that have a primary topic association with one of the topics you've marked.
Second: For each CV, we retrieve your data on the recall questions that are a part of the CV's primary topic. Higher recall scores (1-5) increase your predicted chance of answering the related CV correctly. However, as more time elapses since your most recent review of those recall questions, the predicted chance of answering the CV correctly decreases.
Third: We estimate a skill parameter for you. This is calculated using your performance on the CV's you've already answered. This parameter is updated as you answer more questions.
Fourth: We estimate a difficulty parameter for each CV using the results of how other students like you performed on that CV. Each CV's difficulty parameter updates as more students answer the CV.
Fifth: Now that we have all the inputs to the algorithm (topic recall data, user skill, and question difficulty) we calculate your chances of answering each of the available CV's correctly. Then, for your Daily CV, we pick the CV with a predicted chance of a correct answer closest to 50%.
This algorithm adapts to your performance by matching your prior Daily CV performance and incorporating your data on related recall-style questions. We're looking forward to your feedback!
Data Scientist, Firecracker