MuscleHub A/B Hypothesis Testing

Here are my figures and conclusions from my MuscleHub A/B Hypothesis Test capstone project!

percent_filled_out_app
With a p-value of 0.00096 between the two groups, Fitness Test and No Fitness Test, we can reject the null hypothesis and state that the group that didn’t have to take the fitness test were significantly more likely to fill out the application for the gym.

percent_became_member_after_app
For transition from the application phase to purchasing a membership, there is a p-value of 0.432. With this, we can accept the null hypothesis, that there is not a significant difference between the two groups when deciding which group is more likely to purchase a membership, after filling out the application.

percent_became_member_after_visit
From the very first step of visiting the gym, to the last step of purchasing a membership, there is a p-value of 0.014. This shows that we can reject the null hypothesis, which means that there is a significant difference between which group, Fitness Test or No Fitness Test, is more likely to buy a membership.

Based off of these metrics, I recommend that MuscleHub implement the No Fitness Test route, because it will lead to more membership purchases compared to the Fitness Test route.

Side question: Does anyone have any other recommended resources to learn more about 1-sample t-test, 2 sample, ANOVA, chi2, etc.? I’m still having trouble deciding which is good for which scenarios.

Thank you!