http://www.streetfilms.org/bicycling-is-up-again-in-the-big-apple/

But Ben will put it up early….

From these notes, it seems that their methodology changed dramatically from 2006 to 2007, the most important change being likely the move from one count per year to 7+ counts per year. It could be that the increase in cycling is overstated because the one-count-per year method was getting inaccurate results– in other words, it could be that New York’s always had lots of cyclists but the old study format wasn’t getting good enough data. It’s notable that while the graph’s been increasing since 2000 or so, the really noticeable marked increase in the graph occurs right when they change methodologies.

Also notice that the earlier years (up to 2006, again) are graphed as three-year rolling averages, making them look less volatile and more reliable than they probably are. This also contributes to the impression from the graph as a whole that 2007-9 are radically different years from years previous. Equally important, while this may seem self-evident, is that years 2007-9 are not plotted as rolling averages. If they were, they might not appear as dramatic.

This isn’t to poo-poo the growth of cycling in NYC, which, from my personal and subjective experience, is definitely happening. Nor is it an attack on the NYCDOT, which is a great organization and which puts good resources into important research like this. But, that graph (and some of the other data in the report) definitely has the potential to be misleading, and the just-recently-bike-oriented DOT, which is being pressured from some directions to defend its bike infrastructure projects, does have a vested interest in demonstrating a link between infrastructure and ridership.

(Copy-pasted from my post on Ecovelo.info, where I first saw this)

If the full twelve-hour cycling volume is actually 2,294, then the daily volume is probably around 2500, which would mean that a cycle track plus the promenade would probably serve a total of 10,000 people per day. Still nothing to sneeze at, but not quite the open-and-shut case that it would be if it were 20,000 a day.

In the PDF, they state that the main problem with counting cyclists on the city streets (which would include cycle tracks) is that there could be working cyclists, either messengers or food delivery cyclists, who would only be going a few blocks. You run the risk of counting them every time they cross the line, thus inflating the count.

I am slightly concerned about the methodology, though. Since the indicator measures only commuter routes, what happens when, say, cycle tracks are installed on Columbus, Amsterdam, 1st & 2nd aves? These facilities would likely boost biking in general, but may actually draw people away from greenways to more direct routes. I know I’d take the cycle tracks, rather than going out of my way to the greenway, especially on the east side where the greenway sucks and shows few signs of progress. In the long term, more more would probably end up taking the greenways as more people bike in general, but in the short term the indicator may show biking decreasing. This is just a hypothetical question, and I’m interested to hear what people think.

Since 1985, DOT has been conducting an annual 12 hour count of cyclists entering and existing the center of Manhattan. Known as the NYC Bicycle Screenline Count, it includes counts of cyclists crossing the four East River bridges, entering and exiting the Staten Island Ferry at the Whitehall Terminal and each avenue and the Hudson River Greenway at 50th Street.

Just want to double-check: this is cyclists going both directions between 7AM and 7PM? So a cyclist who goes over the Brooklyn Bridge into Manhattan at 8:30 in the morning and goes back over the bridge at 5:30 in the evening counts as two cyclists?