How to Find Regression Line on Calculator

Updated 6 November 2026 · Standard formula, shown below

This guide shows you how to use a graphing calculator (like the TI-84, TI-83, or similar) to find the equation of a regression line from a set of data points. You enter your x and y values into a list, run the linear regression calculation, and the calculator gives you the slope and y-intercept of the best-fit line. You’ll also get the correlation coefficient (r) to see how well the line fits your data. No prior calculator experience is needed beyond basic navigation.

What You’ll Need

A graphing calculator (TI-83, TI-84, TI-84 Plus, or similar model)
Your data set: pairs of x and y values (at least two points, but more is better)
Familiarity with the calculator’s arrow keys and Enter button

Step-by-Step

Turn on the calculator and press the STAT button. This is usually located near the top of the keypad, often in the third column from the left. The STAT menu will appear with several options.
Select the EDIT menu. Use the arrow keys to highlight “EDIT� (which is option 1) and press ENTER. This opens the list editor, where you’ll see columns labeled L1, L2, L3, etc.
Clear any old data. If there are numbers already in L1 or L2, you need to clear them. Use the up arrow to highlight the column header (e.g., “L1�) and press CLEAR then ENTER. This clears the entire column. Do this for both L1 and L2.
Enter your x-values into L1. Move the cursor to the first row under L1. Type the first x-value and press ENTER. Continue entering each x-value in successive rows. The cursor will move down automatically after each ENTER.
Enter your y-values into L2. Use the right arrow key to move to the first row under L2. Type the corresponding y-value for the first x-value and press ENTER. Continue until all y-values are entered. Make sure each pair (x, y) is in the same row.
Press the STAT button again. This time, use the right arrow key to highlight the CALC menu (which is the second tab at the top).
Choose LinReg(ax+b). Scroll down to find “LinReg(ax+b)� (usually option 4 on a TI-84). Highlight it and press ENTER. The calculator will paste “LinReg(ax+b)� onto the home screen.
Specify the lists (if needed). On most calculators, you can just press ENTER again, and it will use L1 and L2 by default. If your data is in different lists (e.g., L3 and L4), type the list names after the command: LinReg(ax+b) L3, L4. To type a list name, press 2nd then the number corresponding to the list (e.g., 2nd + 3 for L3).
Press ENTER to run the calculation. The calculator will display the results: a (slope), b (y-intercept), r², and r (correlation coefficient). If you don’t see r or r², you need to turn on the diagnostics (see Troubleshooting below).
Write down the equation. The regression line is in the form y = ax + b. Replace a with the number shown, and b with the number shown. For example, if a = 2.5 and b = 1.2, the equation is y = 2.5x + 1.2.

Common Mistakes

Mixing up x and y columns: Entering x-values into L2 and y-values into L1 will give you the regression of x on y, which is a different line. Always put the independent variable (x) in L1 and the dependent variable (y) in L2.
Not clearing old data: If you don’t clear previous lists, the calculator will include old numbers, giving you a wrong line. Always clear L1 and L2 before entering new data.
Forgetting to check r: The correlation coefficient r tells you how strong the linear relationship is. If r is close to 0, the line may not be meaningful, even though the calculator still gives you an equation.
Using the wrong regression type: If your data is curved, a linear regression won’t fit well. The calculator offers quadratic, exponential, and other regressions. Use LinReg only for data that looks roughly like a straight line.

Tips

Turn on diagnostics: To see r and r², press 2nd then 0 (CATALOG), scroll down to “DiagnosticOn� and press ENTER twice. Do this once, and it stays on.
Use the arrow keys to review data: After entering data, scroll through L1 and L2 to double-check for typos. A single wrong number can change the line significantly.
Graph the line: After finding the regression, you can graph the line along with your scatterplot. Press 2nd + Y= to turn on a plot, then press Y= and type the equation. This visually confirms the fit.
Round only at the end: Keep full precision in the calculator’s answer. Round to a reasonable number of decimal places when you write the equation, but use the unrounded values for any predictions.

Troubleshooting

r and r² are not showing: This is the most common issue. Go to 2nd + 0 (CATALOG), scroll to “DiagnosticOn�, press ENTER twice. Then re-run the regression.
Error: DIM MISMATCH: This means L1 and L2 have different numbers of entries. Check that each column has the same number of data points. You may have accidentally left an extra number in one list.
Error: INVALID DIM: This usually means a list is empty or has non-numeric entries. Make sure all entries are numbers and that you haven’t left any blank rows.
The line looks wrong on the graph: Check your window settings. Press ZOOM and select “ZoomStat� (option 9) to automatically set the window to fit your data.
Calculator shows “LinReg� but no output: You might have pressed ENTER before typing the command. Press STAT, then CALC, select “LinReg(ax+b)�, and then press ENTER again. If you see the command on the home screen, you’re good.

Frequently Asked Questions

What does the regression line actually tell me?

The regression line is the straight line that best fits your data points, minimizing the vertical distances (errors) between the points and the line. It allows you to predict y for a given x, and the slope tells you how much y changes on average for a one-unit change in x.

What is the difference between r and r²?

r (correlation coefficient) measures the strength and direction of the linear relationship, ranging from -1 to 1. r² (coefficient of determination) tells you the proportion of the variance in y that is explained by x. For example, r² = 0.81 means 81% of the variation in y is explained by the line.

Can I use this for non-linear data?

No, LinReg is only for linear relationships. If your scatterplot shows a curve, you should use a different regression (e.g., QuadReg, ExpReg) from the same CALC menu.

How many data points do I need?

Technically, you can find a line with just two points, but the line will perfectly pass through both, giving no sense of fit. For a meaningful regression, you need at least three points, and more is generally better (at least 10-20 for reliable results).

Why does my calculator give a different line than my friend’s?

Most likely one of you entered the data incorrectly, or one of you has old data still in the lists. Double-check your data entries and clear the lists before entering. Also, ensure both calculators have diagnostics on, though that doesn’t affect the line itself.

Can I find the regression line without a graphing calculator?

Yes, you can use the formulas for slope and intercept, but they involve many calculations and are prone to error. A graphing calculator or statistical software (like Excel) makes it much faster and more accurate.

Method: the standard formula for this calculation, rounded as each result states. See how the calculators work or report an error.