t-statistic Calculator

Calculate t-statistic (t-value) for hypothesis testing and Student's t-test

Calculate t-statistic

Average value of your sample data

Hypothesized or known population mean

Number of observations in your sample

Standard deviation of your sample

t-statistic Formula

Basic Formula

t = (x̄ - μ) / (s / √n)

x̄: Sample mean

μ: Population mean

s: Sample standard deviation

n: Sample size

Standard Error

SE = s / √n
Standard error of the mean

Basketball Example

Scenario

Basketball player's average: 15 points

Population average: 10 points

Games played: 36

Standard deviation: 6 points

Calculation

t = (15 - 10) / (6 / √36)

t = 5 / (6 / 6)

t = 5.0

Strong evidence of above-average performance!

Interpretation Guide

|t| > 2: Strong evidence against null hypothesis

1 < |t| ≤ 2: Moderate evidence against null hypothesis

|t| ≤ 1: Weak evidence against null hypothesis

Positive t: Sample mean > Population mean

Negative t: Sample mean < Population mean

Understanding the t-statistic

What is the t-statistic?

The t-statistic (or t-value) is a measure that describes how many standard errors a sample mean is away from a population mean. It's central to Student's t-test and is used to determine whether to support or reject the null hypothesis.

When to Use t-statistic vs z-score

  • t-statistic: Small sample size (<30) or unknown population σ
  • z-score: Large sample size (≥30) and known population σ
  • t-distribution has heavier tails than normal distribution
  • As sample size increases, t-distribution approaches normal

Hypothesis Testing Steps

Step 1: State Hypotheses

H₀: μ = μ₀ (null hypothesis)
H₁: μ ≠ μ₀ (alternative hypothesis)

Step 2: Calculate t-statistic

Use the formula: t = (x̄ - μ) / (s / √n)

Step 3: Compare with Critical Value

If |t| > critical value, reject null hypothesis

Applications of t-statistic

One-Sample t-test

Compare sample mean to hypothesized population mean. Used when testing if a sample comes from a population with a specific mean.

Two-Sample t-test

Compare means of two independent samples. Tests whether two groups have significantly different means.

Paired t-test

Compare paired observations (before/after, matched pairs). Tests whether the mean difference between pairs is zero.

Historical Background

The t-statistic was developed by William Sealy Gosset in 1908 while working at the Guinness Brewery in Dublin. Due to company policy preventing employees from publishing, he published under the pseudonym "Student." This is why the t-distribution is also known as "Student's t-distribution." Gosset developed this test to solve practical problems related to quality control in brewing with small sample sizes.