Skip to main content
Daily Math Minute

Statistics & Probability

Fitting a Linear Model

Informally fitting a line to a scatter plot and interpreting its slope and intercept.

Intermediate20 min lesson3 min readUpdated August 12, 2026Author not yet attributed

Prerequisites

  • Scatter Plots

Drawing the Line That Best Fits the Data

A scatter plot's points show a clear upward trend, but no single straight line passes through every one of them exactly. Before reading on, think about what would make one candidate line a better fit than another — what should a well-chosen line try to balance?

Definition — Line of Best Fit

A single line drawn through a scatter plot to informally represent its overall trend, chosen so that roughly as many points fall above the line as fall below it, staying as close to the data as reasonably possible.

Once a reasonable line is fit to the data, its slope and y-intercept carry real meaning drawn directly from the context — the slope describes the typical rate at which one quantity changes as the other increases, and the y-intercept describes a starting or baseline value where the input is zero (when that makes sense for the situation).

Worked Example — Interpreting a Fitted Line's Slope and Intercept

A line fit to a scatter plot of study hours versus test score is y = 4x + 62. Interpret the slope and intercept. The slope, 4, means each additional hour of studying is associated with about 4 more points on the test. The y-intercept, 62, suggests a baseline score of about 62 for a student who studied 0 hours.

Worked Example — Using a Fitted Line to Predict

Using the model y = 4x + 62, predict the test score for a student who studies 7 hours. Substitute x = 7: y = 4(7) + 62 = 28 + 62 = 90.

Graph Visualizer

Domain & range
2
Evaluate a point
  • x^2 = 0

Tip

A fitted line is most reliable for predicting within the range of the original data — extending it far beyond where any real data points exist stretches the model past what it was actually built to describe.

Common Mistakes

  • Drawing a line that passes through as many individual points as possible, rather than balancing points above and below across the whole data set.

    A well-fit line doesn't need to touch many points exactly — it should track the overall trend, with points roughly balanced above and below along its whole length.

  • Using a fitted model to predict far outside the range of the original data, treating the trend as if it must continue indefinitely.

    Predictions are most trustworthy within the range the original data actually covered — a real-world relationship often doesn't continue the same trend far beyond that range.

Key Takeaways

  • A line of best fit informally represents a scatter plot's overall trend, balanced between the points above and below it.
  • The fitted line's slope and y-intercept carry real meaning from the data's context.
  • Predictions from a fitted model are most reliable within the range of the original data.

Summary

Fitting a linear model turns a scatter plot's visual trend into a usable predictive equation, tying together slope, functions, and real data. This closes out Grade 8 mathematics and middle school entirely — Algebra I begins with these exact foundations: linear equations, functions, and systems, now developed with full algebraic rigor.

Sign in to track your progress and mark this lesson complete.

Track your progress
← Scatter PlotsBack to Statistics & Probability