I was writing this morning (a textbook chapter, as a favor for a friend) about the difference between unlinked and linked passenger trip measures. This distinction is important: unlinked trips measure a trip as every time a person boards and alights a vehicle. Linked trips capture the entire journey as one trip, even if there is a transfer in the middle. For unlinked trips, a person making a single journey with a transfer in the middle counts as two unlinked trips.
You can see how much uncertainty unlinked trips add to measuring service use, particularly in large regions where transfers are frequent: counting one trip as two is a fairly big measurement error; if transfers are timed well, the problem can actually get worse as more patrons are likely to be more willing to undertake trips with multiple transfers. And, of course, because we live in a world where nothing is easy, it’s far easier for transit agencies to count unlinked trips than linked trips.
Imagine how difficult it is to do performance evaluation on service changes using unlinked trips. If you reconfigure routes, alter schedules, or open a new line, you are hard pressed to figure out how much increase in service use might be due to lower trip times (ie service improvement) and how much might be due to simply introducing transfers on the one hand, or increasing demand amongst journeys that require transfer on top of basic demand.
For some reason this morning I am typing “unliked” transit trips instead of “unlinked” trips. It’s feeling very Freudian.