What is a Normal Distribution?
Learn what a normal (Gaussian) distribution is, the empirical 68-95-99.7 rule, its link to the Central Limit Theorem, and how to check for normality.
Expected Interview Answer
A normal distribution, or Gaussian distribution, is a symmetric, bell-shaped probability distribution fully described by its mean (center) and standard deviation (spread), where about 68% of values fall within one standard deviation of the mean, 95% within two, and 99.7% within three, the empirical rule.
The Central Limit Theorem explains why normal distributions show up so often: the sampling distribution of the mean of many independent, identically distributed variables tends toward normal as sample size grows, regardless of the shape of the underlying population. This is why parametric tests like the t-test and assumptions in linear regression, such as normally distributed residuals, rely on normality. You can check whether data is approximately normal with a histogram, a QQ-plot comparing sample quantiles to theoretical normal quantiles, or a formal test like Shapiro-Wilk. Real-world data is rarely perfectly normal, but many phenomena, from measurement errors to aggregated averages, are close enough for these assumptions to hold reasonably well.
- Fully described by just two parameters: mean and standard deviation
- Underlies many statistical tests and confidence interval formulas
- Explained and justified by the Central Limit Theorem
- Enables simple probability estimates via the empirical rule
- A useful baseline for spotting skewed or unusual distributions
AI Mentor Explanation
A normal distribution is like the spread of a batter's scores across a full season clustering around their average, with most innings close to that average and only a few very high or very low scores far out. About 68% of their innings land within one 'typical swing' of their average, mirroring the bell curve's 68-95-99.7 spread.
Step-by-Step Explanation
Step 1
Identify the shape
Check whether a histogram of the data looks roughly symmetric and bell-shaped.
Step 2
Compute mean and standard deviation
These two parameters fully define a normal distribution.
Step 3
Apply the empirical rule
Estimate that about 68% of data lies within 1 standard deviation, 95% within 2, and 99.7% within 3 of the mean.
Step 4
Verify formally
Use a QQ-plot or a test like Shapiro-Wilk to check normality more rigorously than eyeballing a histogram.
Step 5
Connect to the Central Limit Theorem
Recognize that sample means of many independent variables tend toward normal regardless of the underlying distribution.
Step 6
Use it appropriately
Rely on normality assumptions only where they roughly hold, since many parametric tests depend on them.
What Interviewer Expects
- Describes the shape and defining parameters correctly
- States the empirical (68-95-99.7) rule
- Connects the concept to the Central Limit Theorem
- Names a way to check normality, such as a QQ-plot or Shapiro-Wilk test
- Knows why normality matters for parametric tests
Common Mistakes
- Saying all real-world data is normally distributed
- Forgetting the Central Limit Theorem applies to sample means, not raw individual data
- Confusing 'normal' the statistical term with 'typical' or 'common'
- Not checking normality assumptions before applying a t-test
Best Answer (HR Friendly)
“A normal distribution is the classic bell-shaped curve where most values cluster around an average, and extreme values become rarer the further you get from that average. It shows up naturally in many real-world measurements and underlies a lot of common statistical methods.”
Code Example
import numpy as np
from scipy import stats
data = np.random.normal(loc=100, scale=15, size=10000)
mean, std = data.mean(), data.std()
within_1sd = np.mean(np.abs(data - mean) <= std)
print(f"Within 1 SD: {within_1sd:.2%}") # close to 68%
stat, p = stats.shapiro(data[:500])
print(f"Shapiro-Wilk p-value: {p:.4f}")Follow-up Questions
- What is the Central Limit Theorem and how does it relate to the normal distribution?
- How would you test whether a dataset is normally distributed?
- What statistical tests assume normality, and what happens if that assumption is violated?
- What is the difference between a normal distribution and a skewed distribution?
- What is a z-score and how is it computed from a normal distribution?
MCQ Practice
1. What two parameters fully define a normal distribution?
A normal distribution is completely specified by its mean, which sets the center, and its standard deviation, which sets the spread.
2. Approximately what percentage of data falls within one standard deviation of the mean in a normal distribution?
The empirical rule states roughly 68% of values fall within one standard deviation of the mean in a normal distribution.
3. The Central Limit Theorem states that:
The CLT says that averages of large enough independent samples approximate a normal distribution, even if the original data isn't normal.
Flash Cards
What shape does a normal distribution have? — A symmetric bell curve centered on the mean.
What two parameters define it? — The mean (center) and standard deviation (spread).
What is the empirical rule? — About 68% of data within 1 SD, 95% within 2 SD, and 99.7% within 3 SD of the mean.
What does the Central Limit Theorem say? — Sample means of many independent draws tend toward a normal distribution as sample size increases, regardless of the original distribution's shape.