دانلود کتاب تست فرضیه های آماری با Microsoft® Office Excel

If you need to interpret the results of statistical tests, want to be able to analyze your own data, or plan a study without having to consult with a statistician, this text is for you. It is written for the researcher who does not want to take a statistics course but needs to understand what the statistician does. It is largely nonmathematical except where absolutely necessary. It does not assume the reader knows much about statistics except what a mean and proportion are. To get the most out of the text, you should have access to Microsoft® Office Excel® but you need not be an expert on its use.
In Chap. 1 you will learn what a P-value is, where it comes from, and how to interpret its value. You will also learn about the type of errors that can be made when drawing conclusions based on the P-value. You will become familiar with two approaches: test-ing hypotheses and their advantages and disadvantages. You will learn how to use these approaches to interpret a collection of several statistical tests in the same study.
Chapter 2 explains how samples can be taken from the population in a way that creates an unbiased and analyzable dataset. The steps in the sampling process are examined. Then, the methods of drawing the sample are explained with their advantages and disadvantages.
In Chap. 3 you will learn how to choose a statistical test that is appropriate for a given set of data. Then, you will learn about the most common statistical tests, how to use Excel to perform them, and how to interpret the results by using them.
Chapter 4 discusses what you can do if you don’t want to wait until a study gets to its planned end before you start looking at the data and draw conclusions about what will happen if the study is allowed to go to completion. It discusses two approaches, both of which can be used to decide to stop a study early if the results look promising. One can also stop a study early if the results do not look promising.
Chapter 5 tells you how to determine the size of a planned sample that allows it to be big enough to see important relationships without being bigger than required. Methods are given for nominal and continuous variables and for datasets with and without independent variables.