Manual footfall counts (counts of people who walk past a particular point) were collected at ten sites between the 5th to the 9th of July 2021 between 10:00 and 16:00 each day. At the time of data collection, footfall cameras were installed at three of the ten sites: Briggate, Headrow and Commercial Street. The coordinates of each site are available in the ‘LocationData.csv’ file. Footfall counts were collected at two sites per day and three data collectors were located at each site. Data collectors were replaced during breaks to ensure continuity.
The aim of the project was to capture manual counts of pedestrians to:

1. Validate estimates of the ambient populations produced using a statistical model in three geographic areas within the Metropolitan Borough of Leeds, UK: Headingley, Wetherby and Leeds city centre.
2. Validate the accuracy of footfall camera data captured in Leeds city centre.

The accompanying dataset consists of 24 files providing manual footfall counts at ten locations. Each file contains the collection date and a timestamp, with each timestamp representing one count/one pedestrian. For each location, the files are numbered 1 to 3, representing the three data collectors. Due to data collection issues, timestamped data for the 5th of July 2021 at Bond Street and Vicar Lane are not available; however, the total counts recorded by each data collector at these locations are included in the dataset.

Counting application
Counts were logged using the iOS application Counter+ which is available free of charge via the App Store. The application allows counts to be logged with an associated timestamp, enabling the data to be used in temporal analysis. The data can be exported from the application as a .txt file for further analysis. The application stores a maximum of 1000 records, thus the data must be exported prior to reaching over 1000 counts to avoid data loss.

Counting methodology
Data collectors were recruited from the student body at the University of Leeds. The data collectors were instructed to count all pedestrians who were not cycling, skateboarding, scootering, or using any form of transport. Individuals using motorised mobility aids were counted. All children were counted including those in pushchairs or being carried. Data collectors were located in positions which did not impede the flow of pedestrians and allowed them to have an uninterrupted view of the count location. To ensure that all data collectors were enumerating pedestrians in the same geographical area, pedestrians were counted as they passed a pre-determined physical marker, for example a lamppost. At locations with a footfall camera installed, the footfall camera was the selected physical marker.

Novelty
There are currently no, or very limited, publicly available manually collected footfall counts. This makes it difficult to assess the accuracy of automatically collected footfall counts, which are commonly used in mobility and urban analytics research.

Quality
On Tuesday 6th July, there was heavy rain from 10:00 to 16:00 at both locations, North Street (Wetherby) and B6167 Otley Road (Headingley) which may impact the number of pedestrians. During the other data collection days, the weather was dry with no cloud between 10:00 and 16:00. On Friday 9th July at Commercial Street, a sales cart was in proximity to the footfall camera. This partially obscured the data collectors’ view of pedestrian flows; it is unknown whether the sales cart would impact counts recorded by the footfall camera.

Representation and Bias
There are no expected representational issues or biases in the data. All people have a similar chance of being recorded by the human data collectors. See ‘Counting Methodology’ for details about people who were deliberately excluded due to their chosen transport method.

Citation
The following statement should be included when citing the use of this dataset:
“The data for this research were funded by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 757455), and provided by the Consumer Data Research Centre, an ESRC Data Investment (grant ES/L011840/1; ES/L011891/1)"