**Unit 3 Overview and Outcomes**

The ability to estimate population parameters or to test hypotheses about population parameters using sample statistics is one of the main applications of statistics in improving decision making in business. Whether estimating parameters or testing hypotheses about parameters, the inferential process consists of taking a random sample from a group or body (the population), analyzing data from the sample, and reaching conclusions about the population using the sample data. For example, what is the mean dollar amount spent by families per month at the movies, including concession expenditures, or what proportion of workers telecommute at least 1 day per week?

**After completing this unit, you should be able to:**

- Write the null and alternative hypotheses.
- Calculate the results of the hypothesis test.

**Course outcome practiced in this unit:**

**GB513-2: **Apply hypothesis testing and probability analysis to solve business problems.

examine practical applications of hypothesis testing. Make sure to read the chapter on hypothesis testing before posting. Without understanding the methodology and how to write the null hypothesis and the alternative hypothesis, .

Successful management is about making decisions to bring about desired change. A manager who has a specific objective will execute an action, hoping that the result will be the achievement of that objective. However, due to the multitude of factors that are in play in any real-life situation, it is difficult to tell if the desired outcome has been achieved or if the new numbers are only due to randomness.

For example, a sales manager implements a bonus system and sees the sales figures go up by 3% in the next month. She claims success, but was this really due to the bonus system or could the increase be the result of random fluctuations, something that would have happened anyway? Can the manager confidently claim that there has been a significant change, something that could not be the result of random fluctuations?

Another example may be where HR is worried there may be a difference between the salaries of men versus women that may need to be corrected. Is the difference between average salaries of the two genders significant, or could it be due to coincidence? Can someone claim there is discrimination at work?

Hypothesis testing settles such questions by analyzing statistical data and is key to managing intelligently. Without these answers, you cannot be sure if the actions you are taking are working out or whether things really are the way you think they are. If you do not know these, how can you know what to do?

For this Discussion, describe a key question or claim that may drive important actions from your work, as in the examples above. Follow the template below and answer all questions:

**Describe the key question:**What is the key question/claim that needs to be settled? What are the actions being considered based on the possible answers? (For example, the question for the sales manager is whether the bonus system is effective, and the actions are whether to continue the bonus system or not. HR's question is whether there is discrimination and the action will be adjusting salaries or not.) Without hypothesis testing, how is management deciding on the answer to the question?**How would you set up the hypothesis test?**State what the null and alternative hypothesis would be and what kind of a test you would use.**Describe the data you would use:**What do you need and where would it come from? Does it need to be collected or does it already exist?**How would you explain to management the significance of this test?**How would you convince them that using this is necessary? What would be drawback of not using a hypothesis test in this situation?