Rule of Thumb for CLT Applicability
You'll learn practical heuristics to decide when the Central Limit Theorem gives reliable normal approximations.
The Core Idea
The Central Limit Theorem (CLT) guarantees that sample means approach a normal distribution as sample size grows—but how large is "large enough"? You need rules of thumb to decide when to trust the CLT in practice.
The most common heuristic: n ≥ 30. If your sample size is at least 30, the CLT usually provides a good normal approximation, even if the underlying population isn't normal.
Why This Matters
Think of the CLT as a smoothing process. A larger sample averages out the quirks of the original distribution. But heavily skewed or wildly irregular distributions need more averaging (larger n) to look normal. A symmetric, well-behaved distribution might work fine even with n = 15.
Practical Heuristics
- n ≥ 30: Safe default for most distributions
- Skewness check: If the population is strongly skewed or has extreme outliers, you may need n ≥ 50 or even n ≥ 100
- Already normal: If the population itself is approximately normal, the CLT works perfectly even for small n (like n = 5)
- Continuity correction: For discrete distributions (like binomial), apply the continuity correction and verify np ≥ 10 and n(1−p) ≥ 10
Example Thinking
If you're sampling from a distribution you know is heavily right-skewed (like income data), don't trust n = 30 blindly. Check a histogram or summary statistics. If skewness is extreme, aim for n ≥ 50.
Key Takeaway: Use n ≥ 30 as a starting guideline, but increase the sample size requirement when dealing with skewed, heavy-tailed, or irregular distributions; always consider the population's shape before applying the CLT.