Math · Problem-Solving and Data Analysis · Q.D.

Two-variable data: Models and scatterplots

Scatterplots and models are slope-in-context plus 'the line is a model, not the truth.' Residuals and interpolation sit at the top of the band.

What Bluebook is testing

Q.D. is two-variable data: association, line of best fit, interpreting slope and intercept of a model, and predictions inside the data range.

How it shows up

A scatterplot with a sketched line, or an equation ŷ = … Hard items ask what the slope means in units, or which point is farthest from the model.

A method that survives Module 2

Write the slope as a sentence with units before you look at the choices. For a prediction, plug into the model — do not read a nearby point off the scatter and stop. Desmos can fit a line if they give the table. On a picture, estimate two points on the line, not on the scatter.

Traps that look like knowledge

  • Reading correlation as causation
  • Extrapolating past the data and treating it as fact
  • Slope units backwards (x per y instead of y per x)
  • Mixing a residual (actual − predicted) with the predicted value

Questions

What is "Two-variable data: Models and scatterplots" on the Digital SAT?
It is an official College Board skill in Digital SAT Math. The wording on Bluebook matches the domain taxonomy Drilled uses in drills, analytics, and the notebook.
How does Drilled train Two-variable data: Models and scatterplots?
Filter the bank to the skill, run a timed drill, autopsy every miss, and let the notebook bring the same pattern back. Domain drills and full section sims include the skill at the official mix.
What is the most common trap on Two-variable data: Models and scatterplots?
Reading correlation as causation

Related

Hard practice. $19 a month.

7 days free. Adaptive sims, skill drills, error autopsy. Not affiliated with College Board.