What is Z-test?

S3-SA3-0206

Grade Level:

Class 9

AI/ML, Data Science, Physics, Economics, Cryptography, Computer Science, Engineering

Definition

What is it?

The Z-test is a statistical tool used to compare the average (mean) of a sample to a known average of a larger group, or to compare the averages of two different samples. It helps us decide if the difference we see is real or just due to chance, especially when we have a large amount of data.

Simple Example

Quick Example

Imagine your school says the average height of Class 9 students in your city is 150 cm. You measure 50 students in your class and find their average height is 153 cm. A Z-test can help you figure out if your class is actually taller than the city average, or if this small difference is just a coincidence.

Worked Example

Step-by-Step

Let's say the average daily mobile data usage for students in your town is 1.5 GB with a standard deviation of 0.5 GB. You survey 100 students from your school and find their average usage is 1.6 GB. Is your school's data usage significantly higher?

Step 1: Write down the known values.
Population Mean (μ) = 1.5 GB
Population Standard Deviation (σ) = 0.5 GB
Sample Mean (x̄) = 1.6 GB
Sample Size (n) = 100

---
Step 2: Calculate the Standard Error (SE) of the mean. SE = σ / sqrt(n)
SE = 0.5 / sqrt(100) = 0.5 / 10 = 0.05

---
Step 3: Calculate the Z-score. Z = (x̄ - μ) / SE
Z = (1.6 - 1.5) / 0.05 = 0.1 / 0.05 = 2

---
Step 4: Interpret the Z-score. A Z-score of 2 means your school's average usage is 2 standard errors above the town's average. This is usually considered a significant difference. So, yes, your school's data usage is likely significantly higher.

Answer: The Z-score is 2, suggesting a significant difference.

Why It Matters

The Z-test is crucial for making informed decisions from data. In AI/ML, it helps evaluate if a new model performs better than an old one. Data Scientists use it to compare product performance, while economists might use it to compare income levels between regions. It's a foundational skill for anyone working with data.

Common Mistakes

MISTAKE: Using Z-test when the sample size is very small (e.g., less than 30) or when the population standard deviation is unknown. | CORRECTION: For small samples or unknown population standard deviation, the t-test is generally more appropriate.

MISTAKE: Not understanding what a 'significant' Z-score means. | CORRECTION: A Z-score alone isn't enough; you need to compare it to a critical value (from a Z-table) or a p-value to decide if the difference is statistically significant.

MISTAKE: Assuming the data must be perfectly normal. | CORRECTION: While Z-test assumes data is normally distributed, it's quite robust for large sample sizes (n > 30) even if the data isn't perfectly normal, thanks to the Central Limit Theorem.

Practice Questions

Try It Yourself

QUESTION: The average score on a national math test is 70 with a standard deviation of 10. A coaching center's 40 students have an average score of 75. Calculate the Z-score. | ANSWER: Z = (75 - 70) / (10 / sqrt(40)) = 5 / (10 / 6.32) = 5 / 1.58 = 3.16

QUESTION: A tea stall claims its average daily sales are 500 cups. You observe 64 days and find the average sales are 520 cups. If the population standard deviation is 80 cups, what is the Z-score? Is this difference significant at a Z-score of 2 or higher? | ANSWER: Z = (520 - 500) / (80 / sqrt(64)) = 20 / (80 / 8) = 20 / 10 = 2. Yes, it is significant.

QUESTION: A smartphone company believes the average battery life of its new model is 24 hours (population standard deviation = 3 hours). A tech reviewer tests 100 phones and finds the average battery life to be 23.5 hours. What is the Z-score? Does this suggest the company's claim is overstated if a Z-score less than -1.96 is considered significant? | ANSWER: Z = (23.5 - 24) / (3 / sqrt(100)) = -0.5 / (3 / 10) = -0.5 / 0.3 = -1.67. Since -1.67 is not less than -1.96, it does not significantly suggest the claim is overstated.

MCQ

Quick Quiz

Which of the following is a primary condition for using a Z-test?

The sample size is very small (less than 10).

The population standard deviation is unknown.

The population standard deviation is known.

The data is always qualitative (like colors or names).

The Correct Answer Is:

The Z-test requires knowing the population standard deviation. If it's unknown, especially with a small sample, a t-test is usually used. Z-test works with quantitative data (numbers).

Real World Connection

In the Real World

When you see news reports comparing average salaries in different Indian states or comparing the effectiveness of two different fertilizers on crop yield, a Z-test or similar statistical method is often behind those conclusions. For instance, ISRO scientists might use it to compare the performance of a new rocket engine design against previous ones based on test data, helping them make critical engineering decisions.

Key Vocabulary

Key Terms

MEAN: The average value of a set of numbers. | STANDARD DEVIATION: A measure of how spread out the numbers in a data set are from the mean. | POPULATION: The entire group of individuals or items that we are interested in studying. | SAMPLE: A smaller, representative group taken from the population. | STANDARD ERROR: The standard deviation of the sample mean, indicating how much the sample mean is likely to vary from the population mean.

What's Next

What to Learn Next

Next, you should explore the 't-test'. It's very similar to the Z-test but is used when you don't know the population's standard deviation or when your sample size is small, which happens a lot in real-world studies. Understanding both will give you powerful tools for data analysis!