Why null hypothesis is Ho <= 40 % ?
In the example presented in video, market analyst was asked to check if email open rate of competitors is above 40%.
So according to previous lectures, null hypothesis should be the one which we test and we assume "current state" as null hypothesis.
According to this logic, null hypothesis should be Ho >= 40%.
But video says null hypothesis is Ho<= 40%. I could not understand this.
Please clarify and if possible point me to other material where I can find how to choose null and alternate hypothesis.
Thanks for reaching out!
Here, we want to estimate if the competitor has a higher rate. This basically tells you the alternative hypothesis. Usually, what you try to achieve or prove is what drives the alternative hypothesis.
If we cannot reject the null hypothesis, it means we have been wrong and the competitor has the same or lower email rate.
The 365 Team
Aren't we actually doubting that our competitors have email open rate > 40%, aren't we? So shouldn't this be H0? we want to reject this.
In the example, why would want to reject (if in general, in hypothesis testing we aim to reject Null hypothesis) email open rate <= 40%?
If I would take the above situation, I would get a false positive result.
Well, In the exercise we want to estimate if OR is exactly 40%. Using the same logic the alternative hypothesis should be OR = 40%? In the solution provided it's been made the null hypothesis. I am confused
Hi Timur and Sayed!
Thanks for reaching out!
Let's first clarify the roles and meanings of the null hypothesis (H0) and the alternative hypothesis (H1).
The null hypothesis indeed represents the status quo, a baseline, or a default position. In our case, where we want to check if a competitor has a higher email open rate than our company's 40%, the null hypothesis would be that the competitor's open rate is less than or equal to 40%. Meaning, the baseline assumption that their open rate is not higher than ours remains unchanged.
The alternative hypothesis represents a change or difference from the null hypothesis. It is not necessarily a desired outcome or a 'wish' but is the condition that we are trying to find evidence for. In our case, the alternative hypothesis is that the competitor's open rate is more than 40%.
The alternative hypothesis isn't about what we wish or want to happen. Rather, it's what we suspect might be happening and are testing for. The null hypothesis is the default assumption that there is no change or difference.
Hope this helps.