- 2 min read. Content may take a few seconds to load.
Visitor Arrivals & Non-Hawaiian Devices
Non-Hawaiian mobile devices in the SafeGraph panel sample represent any devices whose “home” location was not in the State of Hawaii over the last six weeks. A mobile device “home” location is determined by looking at six weeks of dwell times and determining where the device was during nighttime hours local time (6 pm to 7 am). Data on non-Hawaiian mobile devices in the SafeGraph panel sample should correlate with data on visitor arrivals from the Hawaii Tourism Authority. But does it?
After Hawaii traveler quarantine rules were put in place between March and October 2020, non-Hawaiian mobile devices detected in the State of Hawaii are higher than expected as compared visitor arrivals. It might be an artifact of the SafeGraph panel sample itself. But it also might be worth further investigation.
State of Hawaii Visitor Arrivals & Expenditures
The Data: The graph below mixes data from the Hawaii Tourism Authority on the number of arriving visitors, visitor expenditures, and expenditures per capita from January 1999 to July 2021.
What to Look For: You can clearly see the impact of the COVID-19 public health emergency response, which created by far the most significant decrease in total visitor arrivals to the State of Hawaii over the last twenty years. The Hawaii Tourism Authority presents similar graphs to show the resilience of tourism to major disruptions since 1970.
Scroll over the graph below.
Non-Hawaiian Mobile Devices
The Data: The two graphs below compare Hawaii Tourism Authority data on visitor arrivals with data on non-Hawaiian mobile devices from the SafeGraph mobile device panel sample. A mobile device in the SafeGraph panel sample is classified with high confidence as a Hawaiian “home” device by determining its location during nighttime hours (6 pm to 7 am) over the past six weeks. The difference between the total number of unique mobile devices detected within the State of Hawaii regardless of their home location (num_unique_visitors), and mobile devices specifically determined to be Hawaiian “home” devices (number_devices_residing), can be used as a proxy for tourists and other non-resident visitors to the State of Hawaii.
What To Look For: Look at the visual correlation between visitor arrivals from the Hawaii Tourism Authority and non-Hawaiian mobile devices detected within the State of Hawaii from the SafeGraph panel sample. Non-Hawaiian devices match well with visitor arrivals from the Hawaii Tourism Authority before the COVID-19 pandemic. But look at the period between April and October 2020 during the traveler quarantine. Non-Hawaiian mobile devices are much higher than expected compared to visitor arrivals, especially when seen as a percentage of all devices in the State of Hawaii. Is this real? Or is this just an anomaly in the SafeGraph algorithm for determining home locations? What were the actual home areas of these supposedly non-Hawaiian devices, and what points of interest did these they visit during the traveler quarantine period? It might be worth looking into.
Scroll over the graphs below.
TECHNICAL NOTES
SafeGraph data variable names are in parenthesis. For detailed information on how SafeGraph variables were derived, see the SafeGraph documentation. SafeGraph data was processed for the State of Hawaii using PostgreSQL and PostGIS then added into R. Publicly-available visitor data from the Hawaii Tourism Authority, updated monthly by the Research & Economic Analysis Division, State of Hawaii Department of Business, Economic Development and Tourism (DBEDT) Data Warehouse, was downloaded and then bound into a final table in R. Graphs developed in R included use of the dygraphs package.
