# What is Hypothesis Testing? - Definition, Steps & Examples

Coming up next: Conducting Hypothesis Testing for a Mean: Process & Examples

### You're on a roll. Keep up the good work!

Replay
Your next lesson will play in 10 seconds
• 1:16 Step 1: Hypothesis
• 2:38 Step 2: Analysis Plan
• 4:03 Step 3: Data Analysis
• 4:33 Step 4: Interpretation
• 5:27 Example
• 7:05 Lesson Summary
Save Save

Want to watch this again later?

Timeline
Autoplay
Autoplay
Speed

#### Recommended Lessons and Courses for You

Lesson Transcript
Instructor: Yuanxin (Amy) Yang Alcocer

Amy has a master's degree in secondary education and has taught math at a public charter high school.

A proper hypothesis test consists of four steps. After watching this video lesson, you'll understand how to create a hypothesis test to help you confirm or disprove an assumption.

## Definition

In this lesson, we will talk about what it takes to create a proper hypothesis test. We define hypothesis test as the formal procedures that statisticians use to test whether a hypothesis can be accepted or not. A hypothesis is an assumption about something. For example, a hypothesis about family pets could be something like the average number of dogs per American household is two.

Hypothesis testing is about testing to see whether the stated hypothesis is acceptable or not. During our hypothesis testing, we want to gather as much data as we can so that we can prove our hypothesis one way or another. There is a proper four-step method in performing a proper hypothesis test:

1. Write the hypothesis
2. Create an analysis plan
3. Analyze the data
4. Interpret the results

Let's take a look. But first, let's meet Sam. Sam has a hypothesis that he wants to test. Sam works as a researcher with the National Food Administration. He is the one that goes out and tests the food that we eat to make sure that it is safe. Let's see how he follows the four-step method.

## Step One: Hypothesis

The first step is that of writing the hypothesis. You actually have two hypotheses to write. One is called the null hypothesis. This is the hypothesis based on chance. Think of this as the hypothesis that states how you would expect things to work without any external factors to change it. The other hypothesis is called the alternative hypothesis. This is the hypothesis that shows a change from the null hypothesis that is caused by something.

In hypothesis testing, we just test to see if our data fits our alternative hypothesis or if it fits the null hypothesis. We don't worry about what is causing our data to shift from the null hypothesis if it does. Keep in mind, when writing your null hypothesis and alternative hypothesis, they must be written in such a way so that if the null hypothesis is false, then the alternative hypothesis is true and vice versa.

What does Sam do here? Sam's null hypothesis is that all meat that is sold to supermarkets is less than 48 hours old. Sam's alternative hypothesis is that all meat that is sold to supermarkets is more than 48 hours old. As you can see, if the null hypothesis is false, then the alternative hypothesis is true.

## Step Two: Analysis Plan

The second step is to create an analysis plan. This involves deciding how to read your results to know whether your null hypothesis is true or your alternative hypothesis is true. Usually, this involves analyzing just one single test statistic.

There are two ways to read your results: P-value method and the region of acceptance method. The P-value is the probability of observing the desired statistic. If this P-value is less than the significance level, then the null hypothesis is not valid. The significance level is the probability of making the mistake of saying that the null hypothesis is not valid when it actually is true. The region of acceptance is a chosen range of values that results in the null hypothesis being stated as valid.

For this step, Sam decides to analyze his data using the region of acceptance. The statistic that Sam decides to use is the number of hours the meat is at that is being sold to supermarkets. Sam goes to various meat providers and checks to see the age of the meat that is being sold. He then analyzes this statistic to see how many meat providers are shipping meat out under 48 hours. The region of acceptance is 99% or higher. This means that if 99% or more of the meat producers ships out their meat in time, then the null hypothesis is valid.

## Step Three: Data Analysis

The third step is that of analyzing the data. It is the putting step two into action. It is in this step that the data is analyzed and either a P-value is found, or the data's region is found.

It is in this step that Sam checks his data to see how many of his meat producers are shipping out their meats within 48 hours. Sam looks at his data and sees that 99.9% of the meat producers are shipping out their meats within 48 hours.

## Step Four: Interpretation

The fourth step involves interpreting the results. It is in this step that the data is compared to the region of acceptance or the significance level. If the P-value is less than the significance level, then the null hypothesis is not valid. If the data is within the region of acceptance, then the null hypothesis is valid.

Sam looks at this data. His data shows that the data's region is at 99.9%. He compares it to his acceptable 99%. Is 99.9% higher than 99%? It is. This means that his data is within the region of acceptance. This tells Sam that he can say that the null hypothesis is valid. Now, he has the data to prove his null hypothesis statement. This is what he wanted to happen. He wanted to be able to tell people that his meat producers are shipping out fresh meat that is less than 48 hours old.

To unlock this lesson you must be a Study.com Member.

### Register to view this lesson

Are you a student or a teacher?

#### See for yourself why 30 million people use Study.com

##### Become a Study.com member and start learning now.
Back
What teachers are saying about Study.com

### Earning College Credit

Did you know… We have over 200 college courses that prepare you to earn credit by exam that is accepted by over 1,500 colleges and universities. You can test out of the first two years of college and save thousands off your degree. Anyone can earn credit-by-exam regardless of age or education level.