As mentioned earlier, you need to choose a representative sample before conducting the hypothesis test. Let’s take a look at the following video and understand how to do the same.
According to the given claim, the population mean for the given Search CTR = 40% or 0.4. Now the sample can show variation and may not be exactly equal to this value. It is only that on an average when you compute the mean CTR for several days, will the search CTR average out to 0.4.
Now once, you have understood this concept, you need to know which distribution you need to look at. And that will help you get at the representative sample.
Thus, as you saw, you need to focus on the “sample mean” for the entire set of the samples. You already know that the sampling distribution, or the distribution of the sample means, which is normal due to the Central Limit Theorem will help us out in this hypothesis testing procedure. With that in mind, let’s go ahead and compute the respective values for the sample mean.
Now, we have finally computed the sample parameters, or the sample mean which came out to be 0.39 with the sample size as 7400. The computation for this is pretty straightforward,i.e. all you did was find the total number of searches for all the 14 days and computed the average search CTR. You can check the calculations in the following image
Now in the next segment, you’ll be computing the necessary test statistic and then make the necessary inferences.