7. Hypothesis testing

Now that we’ve covered descriptive statistics and are familiar with our statistical software, it’s time to turn to inferential statistics. Remember, we conduct inferential statistics because we often cannot collect data from an entire population. Therefore, we collect a sample to draw inferences about the population of interest.

One of the ways we make inferences is using hypothesis testing. We are going to be learning about Null Hypothesis Significance Testing (NHST), which means we test and make inferences about the null hypothesis (which we’ll learn about in more detail soon).

I have created a 4-page “cheat sheet” of the hypothesis testing 4-step process detailed here, with brief details on how it differs for the various inferential statistics we’ll learn in this textbook.

I highly recommend you save this and refer to it often. If possible, print it out and have it by your side as you go through all of your inferential statistics so you know how to go through each step!

Statistics with jamovi - Overarching handout

Note: It will not preview the file in GitHub. When you click the link, click the download button to download the raw file.

Regardless of the inferential statistic we are performing, hypothesis testing goes through the same basic set of procedures:

  1. Look at the data by examining the descriptive statistics and describing your hypotheses.
  2. Check assumptions to ensure your data is satisfactory for performing the inferential statistic (or choosing the correct statistic depending on which assumptions are met). Note that this is covered in Chapter 6 so we won’t discuss it in this chapter.
  3. Perform the test by running the inferential statistic by hand or in jamovi.
  4. Interpret the results and make a decision about whether you reject or fail to reject the null hypothesis, write-up the results in APA format, and provide a visualization of the results.

Let’s go through each of these in turn, using a hypothetical example.