When Do I Need to Do a Power Analysis?

You have your dissertation topic, fleshed out your research questions, and have a plan in place for your target study population and how you’re going to collect your data. But how do you know whether your sample size is adequate to yield enough power in your study? A power analysis can help with that. If you’re doing a quantitative study, you should do a power analysis in order to be confident that you’ve met all the conditions in order to correctly reject your null hypothesis.

What is a Power Analysis?

A power analysis helps you figure out how big your sample needs to be in order to see an effect with a certain degree of confidence given your specific analysis choices. In other words, it also helps you determine how likely it is that you’ll be able to detect a significant effect. If the results are low, you might need to revise your experiment.

In experimental design, there are 4 interrelated concepts:

1. Sample size


2. Effect size


3. Significance level, also known as alpha, or the likelihood of committing a Type 1 error, or the probability of finding an effect that isn’t really present


4. Power, also known as beta, or the likelihood of committing a Type 2 error, or the probability that you’ll find a significant relationship if it exists

If you have 3 out of the 4, you can figure out the fourth.

Types of Power Analysis

There are a few kinds of power analyses, but the two main ones that are often used are: a priori and post-hoc. A priori power analyses are done when doing research planning, and helps you find the sample size you will need to achieve a certain level of power in your study. Post-hoc power analyses are done after you’ve already done your study. This can help you explain results that are not significant.

Is a Power Analysis Necessary?

No, it’s not necessary, per se. No one can force you to do one, although more and more advisors are requiring students to do so. Will your dissertation be better received and your results more respected if you do one? Yes. Conducting a power analysis shows a level of competence and thoughtfulness that a well-trained researcher has. It shows that you thought things through and take your work seriously. If you want to publish your research, a power analysis will be necessary in order to be considered by journals and other researchers.

If you don’t know how to go about doing a power analysis, or want to do one but need assistance in choosing which statistical test to run in order to analyze your data or what significance level you want to choose, contact us today! Here at Thesis Editor, our stats team is highly experienced in all things data analysis and can help you with all of your data needs.