Regression finds a model that fits paired data. On a TI-84-style calculator, the usual workflow is to enter x-values and y-values into lists, choose a regression type, calculate the model, and optionally graph it.
Prepare L1 and L2
Enter x-values in L1 and y-values in L2. The lists must have the same number of entries.
A single missing value creates a dimension mismatch. Always check list lengths before running regression.
Choose a model type
Use linear regression when the scatter plot looks roughly straight. Use quadratic regression for curved data with one turning point. Use exponential regression when the data grows or decays by a multiplicative pattern.
Do not choose a model only because it gives a high value. The model should make sense for the situation.
Run the regression
Open STAT → CALC and select the regression type. Common choices include LinReg, QuadReg and ExpReg.
The calculator returns coefficients such as a, b and c depending on the model. Write the equation carefully and include enough decimal places for the assignment.
Interpret r and r²
The correlation coefficient r is used for linear relationships. The value r² measures how much variation is explained by the model, but it does not prove causation.
Some calculators hide r and r² until diagnostics are turned on. If your class requires them, check calculator settings or teacher instructions.
Graph the model
After finding a regression equation, graph it with the scatter plot to see whether it fits well.
A model can calculate successfully but still be a poor choice if the graph does not match the data pattern.
Quick reference table
| Regression type | Model form |
|---|---|
| Linear | y = ax + b |
| Quadratic | y = ax² + bx + c |
| Exponential | y = ab^x |
| Power | y = ax^b |
| Logarithmic | y = a + b ln(x) |