The SDG uses a basic statistical model to lend realism to the population of students output by the tool. Under this model, when each student is generated, they are internally assigned a performance profile index.
This index is a value between 0.0 and 1.0, and follows a normal distribution (i.e. a bell-shaped curve) where μ = 0.5, σ = 0.16666. In practice, this means that the performance profile index only covers 3 standard deviations, or approximately 99.7% of a normal population. That is to say, we don't model the top and bottom ~ .15% of the population accurately — this level of accuracy is well beyond what we need for the SDG.
Items Controlled by Performance Profile
The performance profile ensures that the individual student records generated don't have jarringly unusual data. For example, it's atypical for an individual student to have poor attendance, many serious discipline events, yet stellar grades. In the real-world, these values are roughly correlated.
The following items are controlled or influenced by a student's performance profile: