# Calculating demand and testing the effect of days of the week

I am running some data analysis on a site that lets users share a ride. What I am trying to determine is what separates users who are successful in offering rides and users who are less successful. This is in order to provide some tips to less successful users.

The research question I'm trying to answer now is: Is there a day of the week (or set of days) that gives you a higher probability of filling your car with fellow passengers?

The data I have is the ride offers from users with date and everything, and if a deal was made via our site. What's missing is how much request there is. Answering my question would be to find a mismatch between offers and potential requests.

What I have done now is to calculate the amount of offers and 'deals' for the last 3 months for every day of the week. If I plot this data as three boxplots you can easily see that (offers, deals and ratio between the two):

1. The amount of deals per day has less spread than the amount of rides offered.

2. The ratio between offers and deals fluctuates. On Mondays there is a high ratio of deals to offers; on Fridays it's the other way around.

Can I conclude from this that Monday would be a good day to offer a ride? How can I test the significance of this claim?