[kingosticks]

The article says they do the survey on a particular day each year whereas the data set is less specific on that fact. Surely they do the survey over multiple days...? If available, there must be a ton of citibike usage data one could use to verify the citibike stats collected by this survey and generally validate this data a bit.

[aries1980]

Why they not rather contact with Fitbit, Garmin, Under Armour, etc. to provide data?

[mertd]

Perhaps because it would be biased? Cyclist population who wears fitness trackers might look different than the population of all cyclists. So now you have sampling bias to deal with.

[kingosticks]

Yes, this. But even so, I'd probably take that over basing my annual cycling stats on sample taking on a single day.

[danielsamuels]

Strava, MapMyRide and Garmin Connect would probably be the best data sources.

[thrownblown]

google fit probably has more