Let me tell you a very interesting story of how animals detect earthquakes. You won’t believe it, but a snake’s motion can tell us when and where the Earthquake will happen. But how and why snakes? Please keep on reading to know.

Snakes can’t hear but they can feel the reverberation as their body is attached to the ground. So, if you closely observe the snake, depending on the gender, he or she can tell you when the earthquake can happen and in which direction and also how far. But why am I sharing this tale with you? The reason is just this.

So, here is what we did. First, we analysed the data and estimated the events that can happen because of the data. This is what a post-hoc power analysis is. In this blog, we will know the reason why we need Post-hoc power analysis in our PhD research and also many more. So, keep on reading. We start with a basic introduction to post-hoc power analysis.

Post hoc power analysis is a statistical method used to determine the statistical power of a study after the study has already been conducted. It involves analyzing the data collected in a study and using this information to estimate the power of the study to detect significant effects. Post hoc power analysis is generally considered a controversial practice because it can be misleading, as it is based on observed data and can overestimate the true power of a study. It is generally recommended that researchers conduct an a priori power analysis, which involves estimating the required sample size and power of a study before conducting it.

Now, we will not reveal the answer to the most important question of this blog. Maybe it's not the most important but it is the most interesting question. Then what will I answer in this part? The problems associated with Post-hoc power analysis. So, here we go.

• It can be misleading: Post-hoc power analysis can be misleading because it is based on observed data rather than a priori assumptions. This means that it may overestimate the true power of the study.

• It can encourage p-hacking: Conducting a post-hoc power analysis may encourage researchers to look for statistically significant effects that were not originally planned or hypothesized, which can lead to p-hacking or data dredging.

• It does not account for Type I error rate: Post-hoc power analysis does not account for the risk of making a Type I error (rejecting a true null hypothesis) due to multiple testing or other factors.

• It cannot justify a non-significant result: Even if a post-hoc power analysis shows that the study was underpowered, It does not excuse a non-significant finding or imply that a finding would have been significant had a bigger sample size been used.

• It cannot fix problems with study design or implementation: Post-hoc power analysis cannot fix problems with the study design or implementation, such as measurement error or confounding variables.

Now, can’t we understand its use cases after knowing the problems associated with it? Now why do we need “Post-hoc power analysis” for our research? The answer is….

• To determine the minimum detectable effect size: Post-hoc power analysis can help researchers determine the minimum effect size that can be detected with a given sample size and statistical test. This information can be useful for planning future studies or interpreting the results of the current study.

• To identify reasons for non-significant results: If a study did not find a statistically significant effect, post-hoc power analysis can help researchers determine whether the lack of significance was due to a lack of statistical power or a true absence of an effect. This information can be useful for planning future studies or interpreting the results of the current study.

• To justify a non-significant result: In some cases, a non-significant result can still be informative, particularly if it was a well-powered study. Post-hoc power analysis can help researchers justify a non-significant result by showing that the study had sufficient power to detect a meaningful effect if it existed.

But still, we need to answer another question. What is it? Tell us in the comments. 

How commenting and sharing our blog can help you only if you like it? Comment with your name and email id and find the magic.