Unacast Migration Patterns – Sample

Due to unprecedented circumstances, we are now seeing more dramatic and frequent migration patterns than before.

To analyze migration patterns, we designed two metrics/datasets:

* The first metric / dataset calculates changes of home location from one area (origin) to another (destination); we refer to this approach as Home-based Origin-Destination Flux (otherwise known as the Origin Destination Flux)

* The second metric / dataset – Population Distribution Trends – studies geographic areas (counties and states) and calculates the proportion of devices and time spent in them on any given day.

Schema

Home-based Migration

Field |Type |Description |Example
——|——-|——————|——-
origin_area_id | STRING | ID of origin area: FIPS (string to retain leading zeroes) | 39095
origin_area_county | STRING | name of origin area: County | Fairfax
origin_area_state | STRING | name of origin area: State | VA
origin_area_msa | STRING | name of origin area: MSA | Washington-Arlington-Alexandria, DC-VA-MD-WV
origin_centroid_lat | FLOAT | geographical representation of origin area (county level): latitude | 34.0773239
origin_centroid_lon | FLOAT | geographical representation of origin area (county level): longitude | -118.2420228
destination_area_id | STRING | ID of destination area: FIPS (string to retain leading zeroes) | 18003
destination_area_county | STRING | name of destination area: County | Arlington
destination_area_state | STRING | name of destination area: State | VA
destination_area_msa | STRING | name of destination area: MSA | Washington-Arlington-Alexandria, DC-VA-MD-WV
destination_centroid_lat | FLOAT | geographical representation of destination area (county level): latitude | 33.21535242
destination_centroid_lon | FLOAT | geographical representation of destination area (county level): longitude | -87.51079704
isoweek_number | INTEGER | iso week number of calendar year in the range 0-53 | 16
date_of_isoweek | DATE | processing date within the isoweek of interest | “2020-04-13”
moves_extrapolated | INTEGER | number of moves (sum of identifiers with observed move from origin_area to destination_area for this time period) (this can be aggregated with a sum() over bigger geographical areas e.g., state) | 100
moves_normalised | FLOAT | normalised number of moves (percentage of people moving according to origin area census) (this can be aggregated with a avg() over bigger geographical areas e.g., state) | 0.6
weekly_difference | FLOAT | change in moves_extrapolated in percentage compared to a baseline in 2019 based on the same week number. This metric represents “Null” in case that either in the baseline or in the period of interest there were less than 5 movements. | -22
processing_date | DATE | date on which data has been calculated | “2020-04-20”

Population Distribution Trends

Field |Type |Description |Example
——|——-|——————|——-
fips | STRING | FIPS code: 5-digit identifier of state and county | ‘01001’
state_fips | STRING | State FIPS code: 2-digit identifier of state | ’01’
state_name | STRING | State name | ‘Alabama’
county_fips | STRING | State FIPS code: 3-digit identifier of state | ‘001’
county_name | STRING | County name | ‘Autauga County’
isoweek_number | INTEGER | ISO week number of calendar year in the range 0-53 | 4
date_of_isoweek | DATE | Last date of week | 1/26/2020
year | INTEGER | Year of data | 2020
identifier_count_proportion | FLOAT | Proportion of identifiers per county in given isoweek | 0.017
duration_hours_proportion | FLOAT | Proportion of dwell time per county in given isoweek | 0.015
identifier_count_proportion_2019 | FLOAT | Proportion of identifiers per county in given isoweek in 2019 | 0.018
duration_hours_proportion_2019 | FLOAT | Proportion of dwell time per county in given isoweek in 2019 | 0.019
identifier_count_proportion_covid | FLOAT | Average proportion of identifiers per county during pre-Covid period | 0.02
duration_hours_proportion_covid | FLOAT | Average proportion of dwell time per county during pre-Covid period | 0.019
identifier_count_bl2019_percentage_change | FLOAT | Percentage change in identifier count compared to 2019 data baseline | -2.44827
duration_hours_bl2019_percentage_change | FLOAT | Percentage change in dwell time compared to 2019 data baseline | -10.8219
identifier_count_blcovid_percentage_change | FLOAT | Percentage change in identifier count compared to pre-Covid data baseline | -2.44827
duration_hours_blcovid_percentage_change | FLOAT | Percentage change in dwell time compared to pre-Covid data baseline | -10.8219

Want to explore? Check out our [Migration Patterns dashboard](https://www.unacast.com/covid19/covid-19-migration-patterns) to understand the migration patterns resulting from COVID-19.

For more details, refer to our [blog post](https://www.unacast.com/post/unacast-launches-migration-patterns) or contact us at [partnerships@unacast.com] (mailto:partnerships@unacast.com).

You May Also Like

Synthetic Healthcare Data: ADT, CCD, and DS Combo pack Sample

$0.00

Using a statistic population health model generator, this data set is made up of highly realistic, but synthetic patient data that can be used for testing purposes.

By: Heapery Staff

Hospitality Visitation Sample

$0.00

This product contains NinthDecimal visitation data on the top Hotel and Motel brands in the U.S.

By: Heapery Staff

vLife (Virtusa) – Synthetic data for hypothyroidism

$0.00

This listing consists of a dataset containing the complete Electronic Medical Records related to hypothyroidism related disease conditions over 1000 patients within the United States. The data is simulated based on clinical knowledge and census demographics.

By: Heapery Staff