How to Find a T Score on a Calculator

A t score tells you how far a sample mean is from a reference mean in standard-error units, or helps you find a cutoff for a t distribution. Enter the relevant sample statistics or degrees of freedom and probability, depending on which result you need. The calculator below finds a one-sample t statistic and its degrees of freedom; a scientific or statistics calculator can also do the arithmetic.

Formula

For a one-sample t test, calculate the t score using t = (x̄ − μ₀) / (s / √n). Here, x̄ is the sample mean, μ₀ is the hypothesized population mean, s is the sample standard deviation, and n is the number of observations. The denominator, s / √n, is the estimated standard error of the sample mean. The degrees of freedom are df = n − 1.

This is the one-sample t statistic, not a t-table lookup or an inverse-t calculation. For a paired test, first calculate the differences within each pair, then use this formula on those differences. Two independent samples require a two-sample t formula, whose standard error depends on the sample sizes and standard deviations.

Worked Example

Using the defaults in the calculator:

• Sample mean, x̄ = 12
• Hypothesized mean, μ₀ = 10
• Sample standard deviation, s = 4
• Sample size, n = 16

First find the standard error: s / √n = 4 / √16 = 4 / 4 = 1. Then find the difference between the sample mean and hypothesized mean: 12 − 10 = 2. Divide that difference by the standard error: t = 2 / 1 = 2.000. Finally, df = 16 − 1 = 15. These are the calculator’s results when rounded as specified.

What You’ll Need

• The sample mean and hypothesized mean, in the same units.
• The sample standard deviation, not the standard error.
• The sample size, entered as a positive whole number. A one-sample t test needs at least two observations.
• A calculator with parentheses and a square-root key, or a statistics calculator that supports one-sample t tests.

Step-by-Step

1. Choose the kind of t score you need. The steps below calculate a one-sample test statistic from summary statistics. If you instead need a critical t value for a confidence interval or a cutoff for a test, use the calculator’s inverse-t or distribution function and provide the degrees of freedom and tail probability.
2. Calculate the sample mean. If you have raw observations, add them and divide by the number of observations. A statistics calculator may calculate the mean and sample standard deviation together. Make sure the observations are the sample being tested.
3. Find the sample standard deviation. Use the sample standard deviation, commonly shown as Sx on statistics calculators. Do not use the population standard deviation, often shown as σx, unless the problem specifically gives a known population standard deviation—in that case the usual procedure is a z test, not this t formula.
4. Record the comparison mean and sample size. The hypothesized mean is the value stated in the question or null hypothesis. Count the observations to get n. Keep means and standard deviation in matching units, such as all centimeters or all inches.
5. Find the standard error. Divide the sample standard deviation by the square root of the sample size: s ÷ √n. On a basic calculator, enter the sample size, take its square root, then divide s by that result.
6. Find the difference in means. Subtract the hypothesized mean from the sample mean: x̄ − μ₀. Keep the subtraction order; reversing it changes the sign of t.
7. Divide to get the t score. Divide the difference by the standard error. With a basic calculator, enter (sample mean − hypothesized mean) ÷ (sample standard deviation ÷ √sample size). Use parentheses to keep the denominator together.
8. Calculate degrees of freedom and interpret the sign. For this one-sample test, df = n − 1. A positive t means the sample mean is above the hypothesized mean; a negative t means it is below. The magnitude tells how many estimated standard errors separate the sample mean from the hypothesized mean.
9. Use the right distribution function if you need a probability. A t score alone is not a p-value. Enter t and df in a t-distribution function to get a tail probability. Specify whether the alternative is lower-tailed, upper-tailed, or two-tailed; for a two-tailed p-value, account for both tails.

Common Mistakes

• Using s instead of the standard error. The formula divides by s / √n, not by s alone.
• Entering the wrong standard deviation. For sample data, choose the sample standard deviation. Statistics calculators may label it Sx; check the calculator’s manual if the labels differ.
• Mixing units. The sample mean, comparison mean, and standard deviation must use compatible units. Converting all three to a different common unit does not change t.
• Reversing subtraction. Use sample mean minus hypothesized mean. A reversed order gives the same magnitude but the opposite sign.
• Confusing a t statistic with a critical value. The statistic comes from data. A critical value comes from a specified confidence level or significance level and degrees of freedom.
• Rounding too early. Keep extra calculator digits through the standard-error calculation, then round the final t score. Rounding can slightly change a reported result.

Tips

Write down the formula before entering numbers, particularly on a calculator without a built-in t-test function. Use parentheses around both the difference and the standard-error denominator. Check that the standard error is smaller than s when n is greater than one; if it is not, inspect the square-root and division steps.

For a confidence interval, the margin of error is a critical t value multiplied by the standard error. The critical value depends on the confidence level and df, so it is not automatically the t statistic from the sample. For a p-value, check the alternative hypothesis before selecting one tail or two. Report t, degrees of freedom, and the p-value or confidence interval as appropriate, using a consistent rounding convention.

Troubleshooting

• The result is unexpectedly huge. Check that you divided by s / √n rather than multiplying by √n or dividing by s / n. Verify that n is a count, not a percentage.
• The calculator gives an error. Confirm that n is at least 2, s is positive, and all entries are numeric. A zero standard deviation makes the t formula undefined unless the difference is also zero, which still does not produce a usable statistic.
• Your result has the wrong sign. Check the subtraction order. A negative t is not automatically an error; it indicates a sample mean below the hypothesized mean.
• You cannot find a t-test menu. Use the formula with summary statistics, or look for a one-sample t test under the calculator’s statistics or test menus. Menu names and key sequences vary by calculator.
• Your answer differs from a textbook or software output. Check whether it uses a one-sample, paired, or independent-samples test; whether the sample or population standard deviation was selected; and whether reported rounding differs.

Frequently Asked Questions

What does a t score tell you?
It measures the distance between a sample mean and a hypothesized mean in estimated standard-error units. Its sign shows the direction of the difference.

How do I find a t score on a basic calculator?
Calculate the square root of n, divide s by that result, subtract μ₀ from x̄, then divide the difference by the standard error. Use parentheses where possible.

What degrees of freedom do I use?
For a one-sample t test, use n − 1. Paired tests also commonly use the number of paired differences minus one; other test designs have different rules.

Is the t score the same as the p-value?
No. The t score is a standardized test statistic. The p-value is a tail probability calculated from the t score and its degrees of freedom under a specified alternative.

Can I calculate a t score from raw data?
Yes. First calculate the sample mean and sample standard deviation from the observations, count n, and then apply the one-sample formula. A built-in one-sample t test can do these steps together.

How many decimal places should I use?
Keep full precision during calculation and usually report t to two or three decimal places, alongside df. Follow any rounding instructions in your assignment or reporting standard.

Related

• Counting Months Calculator: Months Between Two Dates
• Free AVI Recorder: How to Record and Save AVI Video
• How to Get e on a Calculator

in xbar | Sample mean | - | 12 | 
in mu | Hypothesized mean | - | 10 | 
in s | Sample standard deviation | - | 4 | 0
in n | Sample size | - | 16 | 1
out t | T score | - | (xbar - mu) / (s / sqrt(n)) | 3
out df | Degrees of freedom | - | n - 1 | 0