Need to be able to talk about why/how the methods were aligned with the needs/requirements of the district.
By this point I would assume we could have had several different people cycle through to use the R application (need to restart the application each time in order for it to generate a new user ID) which could help us display and describe how these processes could look for others
This is an attempt at a 50,000 foot view of why these particular methods were brought to the data.
Ideally, we would want to model the autoregressive relationship within each user session (e.g., the probability that the user views X report next conditional on the report they are currently viewing), within daily periods, within persons (e.g., a 3-level HLM)