To pin down the relationship between rider birth year and trip duration, I would need to select the appropriate statistical test(s). Tests concerning the parent distributions of the data (for example, KS and Anderson-Darling tests) would have been useful to determine if there were significant differences in the shape of the distributions of ride data for Millenial versus Generation X riders. But these tests would not have answered my initial question concerning correlation between rider age and trip duration.
I also decided not to perform a Z test because this test would only reveal whether or not the two groups (Millenial and Generation X riders) had ride duration means that differed from each other to the point of statistical significance (depending on each group's standard error). Although such a test would tell us something about differences in ride duration for each group, it would not help us understand correlation.
I therefore decided to conduct Pearson's and Spearman's tests for correlations. Using the data grouped by birth year, I sorted the birth years and average ride durations and then ran my Pearson's and Spearman's tests, which produced the following results: