Data from the Safe Blues Experiment at UoA

The Safe Blues Campus Experiment is taking place at The University of Auckland during 2021 and 2022. Live data from the experiment is presented on the Safe Blues Dashboard. Further, participant prize statistics are here.

Importantly, data from Phases 1, 2, and 3 of the experiment that took place during 2021 is now available. Download the 2021 data zip file (43.8Mb). Open publication of this fully anonymized data is approved within the experiment's ethics application and its revisions (UAHPEC22143).

Description of the data

This is a dataset generated by the Safe Blues campus experiment at the University of Auckland. This experiment involved the simulation of several simultaneous real-world epidemics by spreading safe virtual virus-like tokens (or strands) between participating smartphones. Here, we present anonymised and aggregated results regarding strand transmission and campus attendence.

Dataset Structure

Dataset Structure

The Safe Blues dataset is organised within the data folder, which includes the strand parameters, the transmission data, and the participant data. This folder is arranged in the following layout:

data
├── daily
│   ├── strands
│   │   ├── strand1.csv
│   │   ├── strand2.csv
│   │   └── ...
│   └── participants.csv
├── hourly
│   ├── strands
│   │   ├── strand1.csv
│   │   ├── strand2.csv
│   │   └── ...
│   └── participants.csv
└── strands.csv

Strand Parameters

The strand parameters determine the epidemiological behaviour of each virtual virus-like token, including its transmissibility, incubation duration, and infection duration. A table of these parameters is stored in data/strands.csv. This table contains a row for each strand circulated during the campus experiment and contains the following columns:

strand_id: the unique numerical identifier given to each strand,
batch: the release batch or group of strands within which each strand was released,
model: the epidemiological model used by each strand (either "SI", "SEI", "SIR", or "SEIR"),
start_utc & start_nzt: the time at which each strand's reporting begins in either the UTC or NZST/NZDT timezone,
seed_utc & seed_nzt: the time at which each strand's initial infections occur in either the UTC or NZST/NZDT timezone,
stop_utc & stop_nzt: the time at which each strand's reporting ends in either the UTC or NZST/NZDT timezone.
initial: a participant's probability of initial infection for each strand,
strength: the strength of transmission for each strand,
radius: the maximum radius of transmission for each strand,
incubation_mean & incubation_shape: the mean and shape parameters of each strand's gamma-distributed incubation duration (missing when model == "SI" or model == "SIR"), and
infection_mean & infection_shape: the mean and shape parameters of each strand's gamma-distributed infection duration (missing when model == "SI" or model == "SEI").

Transmission Data

The transmission data provides aggregate measurements of the spread of strands throughout their reporting periods. We store the progression of the iᵗʰ strand (strand_id == i) as a daily time series in data/daily/strands/strand(i).csv and as an hourly time series in data/hourly/strands/strand(i).csv. These tables have a row for each time point (either daily or hourly) and have the following columns:

strand_id: the unique numerical identifier given to the strand,
time_utc & time_nzt: the time of each measurement in either the UTC or NZST/NZDT timezone,
susceptible: the current number of participants that are susceptible to the strand,
exposed: the current number of participants that are incubating the strand (missing when model == "SI" or model == "SIR"),
infected: the current number of participants that are infected with the strand,
recovered: the current number of participants that have recovered from the strand (missing when model == "SI" or model == "SEI"), and
distance_factor: the artificial distance multiplier used to emulate social distancing.

Participant Data

The participant data presents aggregate information on the engagement of participants throughout the course of the experiment. This is also available as either a daily or hourly time series in data/daily/participants.csv and data/hourly/participants.csv, respectively. Again, these tables contain a row for each time point (either daily or hourly) and contain the following columns:

time_utc & time_nzt: the time of each measurement in either the UTC or NZST/NZDT timezone,
count_campus: the number of participants on campus during the current NZST/NZDT day,
count_reporting: the number of participants whose phones are sending reports during the current UTC day,
count_registered: the number of participants who have registered for the experiment by the current NZST/NZDT day,
hours_mean: the mean number of campus hours collected by participants who attended campus on the current NZST/NZDT day (missing when count_campus <= 5),
hours_min: the minimum number of campus hours collected by participants who attended campus on the current NZST/NZDT day (missing when count_campus <= 5),
hours_q1: the lower quartile number of campus hours collected by participants who attended campus on the current NZST/NZDT day (missing when count_campus <= 5),
hours_q2: the median number of campus hours collected by participants who attended campus on the current NZST/NZDT day (missing when count_campus <= 5),
hours_q3: the upper quartile number of campus hours collected by participants who attended campus on the current NZST/NZDT day (missing when count_campus <= 5), and
hours_max: the maximum number of campus hours collected by participants who attended campus on the current NZST/NZDT day (missing when count_campus <= 5).

Note that, as mentioned above, all statistics reported in data/daily/participants.csv and data/hourly/participants.csv are in reference to NZST/NZDT days except for count_reporting, which is in reference to UTC days.