Statistics for Data Science and Business Analysis - The Null and the Alternative Hypothesis

The video tutorial covers the basics of hypothesis testing in statistics. It begins with prerequisites, emphasizing the importance of understanding confidence intervals and distributions. The tutorial then explains the four steps of hypothesis testing: formulating a hypothesis, selecting the right test, executing the test, and making a decision. It defines a hypothesis as a testable idea and distinguishes between null and alternative hypotheses. Examples are provided to illustrate two-tailed and one-tailed tests. The tutorial concludes by encouraging viewers to complete exercises to reinforce their understanding.

What is the first step in data-driven decision making?

• Make a decision

• Execute the test

• Find the right tests

• Formulate a hypothesis

Which of the following is an example of a testable hypothesis?

• The mean data scientist salary is $113,000

• The USA would do better under a Clinton administration

• The weather is nicer in California than in New York

• Apples in New York are expensive

What is the null hypothesis in the example of data scientist salaries?

• The mean salary is less than $113,000

• The mean salary is not $113,000

• The mean salary is more than $113,000

• The mean salary is $113,000

In hypothesis testing, what does the null hypothesis represent?

• A new theory

• The status quo

• A personal opinion

• An untestable idea

What is the main goal of a researcher in hypothesis testing?

• To ignore the null hypothesis

• To prove the null hypothesis

• To reject the null hypothesis

• To accept the null hypothesis

Which type of test is used when comparing if a mean is greater than a specific value?

• Double-sided test

• One-tailed test

• Two-tailed test

• Single-sided test

What does a two-tailed test examine?

• If a parameter is less than a value

• If a parameter is equal to a value

• If a parameter is greater than a value

• If a parameter is different from a value