Saskatoon data projectHow many Saskatoon bus stops are wheelchair accessible? It depends what 0 means
The City of Saskatoon publishes Saskatoon Transit’s stops as open data, in the GTFS format transit agencies use worldwide. Our teen students download the file, count one column and find that the obvious answer is wrong. Every number below is ours.
Our count of the Saskatoon Transit stops file, downloaded 22 October 2026 (1,368 rows), with each code read against the GTFS Schedule reference| wheelchair_boarding value | Rows | Meaning in the GTFS reference |
|---|
| 1, stop has no parent station | 998 | Some vehicles at this stop can be boarded by a rider in a wheelchair |
| 0, stop has no parent station | 340 (6 of them are the hub stations themselves) | No accessibility information, not “inaccessible” |
| Empty, no parent station | 9 | Also no accessibility information |
| 1, stop inside a station | 21 | For a stop inside a station: an accessible path exists from outside |
| 2, any stop | 0 | Boarding not possible; the file never says this about any stop |
This script prints those counts and the three stops nearest a point we picked downtown, using the haversine formula for distance on a sphere. It runs on Python 3 with no extra packages:import csv
import math
from collections import Counter
def km_between(lat1, lon1, lat2, lon2):
r = 6371.0
p1, p2 = math.radians(lat1), math.radians(lat2)
dp = p2 - p1
dl = math.radians(lon2 - lon1)
a = math.sin(dp / 2) ** 2 + math.cos(p1) * math.cos(p2) * math.sin(dl / 2) ** 2
return 2 * r * math.asin(math.sqrt(a))
here = (52.1300, -106.6600)
codes = Counter()
nearest = []
with open("transit-stops.csv", encoding="utf-8-sig") as f:
for row in csv.DictReader(f):
kind = "child" if row["parent_station"] else "own"
codes[(kind, row["wheelchair_boarding"] or "empty")] += 1
d = km_between(here[0], here[1], float(row["stop_lat"]), float(row["stop_lon"]))
nearest.append((d, row["stop_name"]))
for key in sorted(codes):
print(key, codes[key])
for d, name in sorted(nearest)[:3]:
print(round(d * 1000), "m", name)
The tempting mistake
A beginner counts the 1s, calls the rest inaccessible, and reports that about a quarter of stops cannot be used. The GTFS reference says 0 or empty means no information, and no stop in the file is coded 2, so the honest finding is: 1,019 marked boardable, 349 unknown, none marked impossible.
Split by a second column
The same code means different things for a stop inside a station. Splitting by parent_station first, as the script does, keeps the 21 child stops from being read with the wrong rule.
Nearest stop
From our sample point at latitude 52.1300 and longitude -106.6600, the closest stop is Downtown Terminal North at 82 m, then 3rd Avenue / 23rd Street at 102 m. Changing the point changes the answer, which makes a good first app.
One more check before publishing
There are 1,368 stop IDs but only 1,363 different stop codes. The gap is six rows with a blank code, and all six are hub stations rather than places a bus stops. Small audits like this decide whether a count can be trusted, and they are the habit we most want a student to take away.
The City of Saskatoon says the data shared on its Open Data site is “free to use, re-use and redistribute without any legal, technological or social restriction” (the city’s open data statement, dataset page). Code meanings come from the GTFS Schedule reference. The City and Saskatoon Transit do not endorse this page or our classes.
