Methodology

Figure 3 is a normalized histogram of trip duration for young  and old riders with a fitted normal curve.  The data have a right skew in each distribution, but are unimodal and close enough to normal that I apply an unpaired t-test to assess the significance of the difference in means.  Because we are drawing sample means from two samples without knowing the underlying population, and because there is a good reason to believe that old and young riders do not belong to the same population, an unpaired t-test is suitable for the analysis.  Our t-test yields a value of 29.448 with a p-value of 0.00, indicating that there is a statistically significant difference in mean trip duration between old and young riders.  This result is contrary to the alternative hypothesis that young riders would have a longer mean trip duration relative to older riders.