The 2013 Strava Metro data for Glasgow City is publicly available through the City’s open data portal. This dataset can be used to analyze cyclists’ activity in Glasgow in different ways including route preference, for which the data is used here. This study visualizes and discusses route choice preference as the density of bicycle trips in Glasgow using the 2013 Strava Metro data.
Route choice density is visualized in this study by highlighting the BIKECNTN data in the open source geographical information system, QGIS. This BIKECNTN variable is described in the metadata as ‘the total number of bicycle trips taken on a road segment regardless of the direction of travel’. Our workflow includes importing the Strava Metro dataset into QGIS, and selecting a graduated aesthetic style to generate the maps below. There are ten classes defined, broken up by QGIS’s natural breaks (Jenks) algorithm. QGIS’ natural breaks algorithm is intended to create classes of data where the variance within each class is minimal while the variance between classes is maximal.
The three maps below visualize macro and meso perspectives aggregated route preference data Glasgow from by Strava users in 2013. After each map is a list of visible high-density corridors, which highlights the importance of looking at the transportation system at different perspectives.
Problems with the data and some final thoughts are discussed in the final two sections.
From a broad view of Glasgow, we can start to get an idea of preferred route choices.
- A761/Paisley Road West
- National Cycle Route 75 (North of the Clyde River)
- National Cycle Route 7 (North of the Clyde River)
- A81
- A77/Eglinton Street
A slightly closer view shows additional routes:
- Govan Road
- Shieldhall Rd
- Victoria Road
- A74/London Road
- Great Western Road
Finally, the nearest perspective shows greater detail.
- Byres Road
- Argyle St -> Dumbarton Rd
- A8
Please note that the style class parameters were not changed between maps, only the perspective. In all we can see a high level of activity by Strava users on:
- A761/Paisley Road West
- National Cycle Route 75 (North of the Clyde River)
- National Cycle Route 7 (North of the River
- A81
- A77/Eglinton Street
- Govan Road
- Shieldhall Rd
- Victoria Road
- A74/London Road
- Great Western Road
- Byres Road
- Argyle St -> Dumbarton Rd
- A8
Problems with the Data
Two found problems with the 2013 Strava Metro dataset include: unclear bike count numbers and missing time stamps.
What is not made clear by the data are how many trips were made at the street level. Looking at the dataset, we can see the bike count numbers are not whole numbers and therefore not a precise measure of how many cyclists were counted. Further, Glasgow City provides no metadata to clarify this issue. It does seem data are normalized data (BKCNTN = Bike Count Normalised). However, yet again, this hard to know without the proper metadata.
An additional problem with the publicly available Strava Metro data is the missing time stamps. Time stamps would allow for a variety of temporal analyses from the seasonal frame down to the minute. The time stamp is marketed by Strava Metro as part of the product.
Urban Data Cyclist 