Calculating the Correlation Between Two Variables on a TI-84
Use this interactive calculator to find Pearson correlation, the regression line, and a scatter plot with trend line. It is designed to mirror the same statistical workflow students often follow on a TI-84 while adding instant visual feedback and a full expert guide below.
Correlation Calculator
Enter paired values for your two variables. Separate values with commas, spaces, or new lines. The calculator will compute the Pearson correlation coefficient r, coefficient of determination r², and the least squares regression line.
How to Calculate the Correlation Between Two Variables on a TI-84
Calculating the correlation between two variables on a TI-84 is one of the most practical statistics skills students learn in algebra, AP Statistics, business math, economics, psychology, and lab science. Correlation tells you whether two quantitative variables tend to move together, whether they move in opposite directions, and how strong that relationship appears to be. On the TI-84, this is usually done by entering values into lists, turning diagnostics on, and running a linear regression. The calculator then displays the regression equation plus the correlation coefficient r and the coefficient of determination r².
If you want the short version, correlation on a TI-84 is usually found by typing your paired values into L1 and L2, enabling diagnostics, selecting LinReg(ax+b), and reading the output. This page expands that process so you understand not only which buttons to press, but also what the numbers mean and when they can mislead you. That matters because a high correlation can look impressive while hiding outliers, non-linear patterns, or data entry mistakes.
What Correlation Means
The Pearson correlation coefficient, written as r, ranges from -1 to +1. A value close to +1 means that as X increases, Y also tends to increase in a strong linear way. A value close to -1 means that as X increases, Y tends to decrease in a strong linear way. A value near 0 means there is little or no linear relationship.
Basic interpretation of r
- r = +1.00: perfect positive linear relationship
- r = -1.00: perfect negative linear relationship
- r around +0.70 to +0.90: strong positive relationship
- r around -0.70 to -0.90: strong negative relationship
- r around +0.30 to +0.69: moderate positive relationship
- r around -0.30 to -0.69: moderate negative relationship
- r near 0: weak or no linear relationship
The TI-84 also shows r², called the coefficient of determination. This value is interpreted as the proportion of variation in Y explained by the linear relationship with X. For example, if r = 0.80, then r² = 0.64, meaning roughly 64% of the variation in Y is explained by the fitted linear model.
| Correlation value | Strength | Direction | Typical practical reading |
|---|---|---|---|
| +0.90 to +1.00 | Very strong | Positive | Variables rise together in a very consistent linear pattern |
| +0.70 to +0.89 | Strong | Positive | Clear upward trend with limited scatter |
| +0.30 to +0.69 | Moderate | Positive | Upward trend exists but points are more spread out |
| -0.29 to +0.29 | Weak | Mixed or none | Little linear association is visible |
| -0.30 to -0.69 | Moderate | Negative | As X rises, Y tends to fall with noticeable spread |
| -0.70 to -1.00 | Strong to very strong | Negative | Clear downward linear trend |
Step by Step: TI-84 Instructions
Here is the standard process for calculating the correlation between two variables on a TI-84 or TI-84 Plus.
- Clear old data if needed. Press STAT, choose 1:Edit, move to the list names, and clear the contents of the lists you plan to use.
- Enter X values. Type the first variable into L1. For example, study hours.
- Enter Y values. Type the second variable into L2. For example, test scores.
- Turn diagnostics on. Press 2nd, then 0 to open the catalog. Scroll to DiagnosticOn, press ENTER, then press ENTER again. This is essential because otherwise the TI-84 may not display r and r².
- Run linear regression. Press STAT, move right to CALC, choose 4:LinReg(ax+b) or a similar linear regression option depending on your model preference.
- Specify the lists. Enter L1, L2. On many TI-84 models, these can be inserted with 2nd 1 for L1 and 2nd 2 for L2.
- Press ENTER. The calculator will display a and b for the regression line, plus r² and r if diagnostics are on.
After that, interpret the output carefully. The sign of r tells the direction, the absolute value of r suggests the strength, and the scatter plot tells you whether the relationship is actually linear enough for correlation to be meaningful.
Worked Example
Suppose a teacher wants to examine whether the number of practice sessions is related to quiz performance. The paired data are:
X: 1, 2, 3, 4, 5, 6, 7, 8
Y: 2, 4, 5, 4, 5, 7, 8, 9
Using either the TI-84 or the calculator above, you would get a positive correlation close to 0.93. That means there is a strong positive linear relationship. As practice sessions increase, quiz performance tends to increase too. The corresponding r² is around 0.87, so the line explains about 87% of the variation in scores.
Why this example matters
- The relationship is not perfectly linear, but it is clearly upward.
- The correlation is strong enough to support using a linear model.
- The scatter plot would confirm that no single extreme outlier is dominating the result.
Comparison Table: Example Relationships and Reported Correlations
The table below shows real and commonly cited examples from educational and public data contexts where correlation is used to describe relationships. These are useful benchmarks when you are learning what different values of r look like in practice.
| Dataset example | Variables compared | Reported or representative r | Interpretation |
|---|---|---|---|
| Height and weight in many adult samples | Height vs body weight | About +0.60 to +0.80 | Moderate to strong positive relationship, but far from perfect because body composition varies widely |
| Outdoor temperature and home heating use | Temperature vs heating demand | Often negative and strong in cold season data | As temperature rises, heating usage tends to drop |
| Study time and exam scores in classroom examples | Hours studied vs score | Commonly +0.50 to +0.80 | Positive relationship, but motivation, prior knowledge, and test difficulty add noise |
| Age of a car and resale value | Vehicle age vs price | Often around -0.70 or lower | Older cars generally sell for less, though mileage and condition matter too |
Why You Should Always Look at a Scatter Plot
One of the biggest mistakes students make is relying only on the value of r. A scatter plot can reveal problems that a single number cannot. For example, you might have a curved pattern, several clusters, or a single outlier pulling the line upward. On the TI-84, it is good practice to make a scatter plot before interpreting the regression output. In the calculator on this page, the chart is generated automatically after you compute the result.
Look for these issues
- Outliers: One unusual point can dramatically change correlation.
- Nonlinearity: Correlation measures linear association, not curved association.
- Restricted range: If all X values are packed into a narrow band, the correlation can appear weaker than it truly is in the broader population.
- Clusters: Two separate groups can produce a misleading overall correlation.
Common TI-84 Problems and Fixes
Problem 1: The calculator does not show r or r²
This almost always happens because diagnostics are off. Use the DiagnosticOn command from the catalog and run the regression again.
Problem 2: Error because lists do not match
Your X list and Y list must contain the same number of observations. If L1 has 10 values and L2 has 9 values, the TI-84 cannot pair them correctly.
Problem 3: Correlation seems wrong
Check for data entry mistakes such as a missing decimal, transposed value, or one Y value entered into the wrong row. Also inspect a scatter plot to see whether an outlier is distorting the result.
Problem 4: Strong relationship but low r
This can happen if the relationship is curved. Correlation only measures how well points follow a straight line. A parabola can show a strong pattern while still producing an r close to zero.
Manual Formula Behind the TI-84 Output
Although the TI-84 handles the arithmetic automatically, understanding the formula helps build confidence. Pearson correlation is based on standardized covariance between X and Y. In conceptual terms, it asks whether values that are above average in X also tend to be above average in Y, and whether values below average in X tend to pair with values below average in Y.
In a classroom setting, the formula is commonly written as:
r = sum[(x – mean of x)(y – mean of y)] / sqrt(sum[(x – mean of x)^2] × sum[(y – mean of y)^2])
The important point is not memorizing every symbol. The important point is realizing that correlation compares paired deviations from each variable’s mean and then scales the result so it always falls between -1 and +1.
When to Use Correlation
- Both variables are quantitative
- You want to measure the direction and strength of a linear relationship
- The data are paired, meaning each X value belongs to one specific Y value
- A scatter plot suggests the pattern is approximately linear
When Not to Use Correlation Alone
- When variables are categorical rather than numeric
- When the relationship is clearly curved
- When extreme outliers dominate the pattern
- When you need to establish cause and effect rather than association
Best Practices for Students and Analysts
- Plot the data first.
- Check that every X has a matching Y.
- Turn diagnostics on before regression.
- Interpret both r and r².
- Write your conclusion in context, not just as a number.
For instance, a better conclusion is: There is a strong positive linear relationship between study time and exam score, with r = 0.81. That is far more useful than writing only r = 0.81.
Authoritative Resources for Further Study
If you want to verify concepts or explore statistical interpretation more deeply, these authoritative sources are excellent references:
- NIST Engineering Statistics Handbook
- Penn State STAT 200 Online Notes
- Centers for Disease Control and Prevention data resources
Final Takeaway
Calculating the correlation between two variables on a TI-84 is straightforward once you know the sequence: enter data into lists, turn diagnostics on, run linear regression, and interpret the output with a scatter plot. The TI-84 gives you speed, but understanding gives you accuracy. Correlation is powerful because it summarizes direction and strength in a single number, yet it should never be used blindly. Always combine the numerical result with graph inspection, context, and common sense. If you do that, the TI-84 becomes more than a button pushing tool. It becomes a reliable statistical assistant.