1) Usernames of each user:
We will collect the student's first name, last initial, and grade.
We will assign each student a six digit ID so we can distinguish between attempts of students who have the same name in the same grade.
We will also save the time stamp for when this student entered their information in the login page. This will also help tie it to our screen recording.
2) Interaction Time:
We will record student interaction time on each page after login to see how long it takes for the student to grasp initial idea and also how long they interact with the trained models.
This will allow us to compare how much time the student chose to spend on the first shown model vs the second shown model.
3) AI Model Features:
Aside from recording the interaction time each student has with the trained AI models, we will also record the how they interact with the AI model.
We will create a function that counts the number of cards were shown to the AI model, for how long.
We will also timestamp when each student decides to show the AI model a card outside of it's trained data.
Record the output of each AI model vs the card that was shown to it. This will allow us to calculate the accuracy of response for each model.
This data along with the interaction time will allow us to see how student's response attitude and time changes based on their exposure.
4) Model Match:
We will record the student's interaction with the Model Match page which allows the student to submit their guess as to which model is the Sahara model and Forest Model.
We will record the students choices and how long they spent to make these choices.
The time will start from the moment student first interacts with this page until they press the check answers button.
This data usage item will be recorded through a function.
5) Application Data:
The application data recorded through functions will be saved in a growing log file or csv file for further data analysis.
We will also screen and audio record each students interaction with the application. The interaction will be recorded through zoom which will provide auto transcription for what the student says out loud. This is also helpful to record their answers to post-assessment questions.