Linear Regression on TI-84 — LinReg Step by Step
Enter your data, run LinReg, and get the equation ŷ = a + bx — plus r and r² — in under two minutes.
Before You Start: Turn DiagnosticOn
By default, the TI-84 hides r and r² after running LinReg. You need to enable diagnostics once, and it stays on permanently.
- Press
2ND→0to open the CATALOG - Press
Dto jump to entries starting with D - Scroll down to DiagnosticOn
- Press
ENTERtwice
You'll see Done on the screen. Now r and r² will always appear in your LinReg output.
This is the single most common reason students can't find r on their TI-84. Do this step first.
Step 1: Enter Your Data into Lists
- Press
STAT→ 1: Edit - Enter your x-values in L1 (press
ENTERafter each) - Move to L2 and enter your y-values in the same order
- Press
2ND→MODEto quit when done
Example data (hours studied vs. exam score):
| L1 (hours) | L2 (score) |
|---|---|
| 1 | 58 |
| 2 | 64 |
| 3 | 71 |
| 4 | 75 |
| 5 | 82 |
| 6 | 90 |
Step 2: Run LinReg
- Press
STAT - Arrow right to CALC
- Select 4: LinReg(ax+b)
- The command appears on the home screen:
LinReg(ax+b) - Press
ENTER
Reading the LinReg Output
y = ax + b
a = 6.4
b = 51.2
r² = 0.994
r = 0.997
| Output | What it means | How to use it |
|---|---|---|
a = 6.4 | Slope — for each 1-unit increase in x, y increases by 6.4 | "For each additional hour studied, the predicted score increases by 6.4 points" |
b = 51.2 | y-intercept — predicted y when x = 0 | Often meaningless in context (a student can't study 0 hours) — acknowledge this |
r² = 0.994 | Coefficient of determination — proportion of variation in y explained by x | "99.4% of the variation in exam scores is explained by the linear relationship with hours studied" |
r = 0.997 | Correlation coefficient — strength and direction of linear relationship | Close to 1 = strong positive linear association |
Your regression equation: ŷ = 6.4x + 51.2
Writing the AP Statistics Regression Sentence
For the slope (most commonly tested):
"For each additional [x-unit], the predicted [y-variable] increases/decreases by [|a|] [y-units]."
For r²:
"Approximately [r² × 100]% of the variation in [y-variable] is explained by the linear relationship with [x-variable]."
For our example:
"For each additional hour studied, the predicted exam score increases by 6.4 points." "Approximately 99.4% of the variation in exam scores is explained by the linear relationship with hours studied."
Step 3: Draw the Scatter Plot and Regression Line
Set up the scatter plot:
- Press
2ND→Y=to open STAT PLOTS - Select Plot1 → press
ENTER - Set On, Type: scatter (first icon), Xlist:
L1, Ylist:L2 - Press
ZOOM→ 9: ZoomStat — this auto-fits the window to your data
Add the regression line:
- Press
Y= - In Y1, press
VARS→ 5: Statistics → arrow right to EQ → 1: RegEQ - Press
GRAPH
The regression line now appears over your scatter plot.
Making Predictions with the Equation
To predict y for a given x value:
Method 1 — Use the equation directly: ŷ = 6.4(4.5) + 51.2 = 28.8 + 51.2 = 80
Method 2 — Use the graph:
- Press
TRACE - Arrow right/left to move along the regression line
- The x and y values appear at the bottom of the screen
Extrapolation warning: Only predict within the range of your x data. Predicting outside this range (extrapolation) is unreliable — mention this on AP free response.
Common Mistakes
1. r and r² not showing up
You haven't run DiagnosticOn. See the first step above — it only needs to be done once.
2. Using LinReg(a+bx) instead of LinReg(ax+b)
Both calculate the same line, but the letters are swapped. LinReg(ax+b) uses a for slope and b for intercept — the AP Statistics convention. Be consistent with whichever you use.
3. Confusing r and r²
r = correlation coefficient (ranges from −1 to 1, measures direction and strength).
r² = coefficient of determination (ranges from 0 to 1, interpreted as a percentage of variation explained). They are different numbers — don't swap them in your conclusion.
4. Forgetting context in the conclusion "The slope is 6.4" gets partial credit. "For each additional hour studied, the predicted score increases by 6.4 points" gets full credit. Always include context.
Practice running LinReg on the free TI-84 calculator with your own data.
Related Guides