Data

COVID-19 caused widespread disruption to normal lives and human activities. In China, the mobility behaviour response to the COVID-19 pandemic at the intra-city levels is largely unknown, mainly due to a lack of individual-level publicly available mobility data.

Globally, the Google Community Mobility Report provides a percentage change in place visit frequencies in six types of locations to help understand people’s social activities within cities, but this dataset does not include mainland China.

Here, using 210 million geotagged posts from 10 million users in Weibo social media platform, our team provides fine-grained mobility data describing the daily percentage change in visits to functional amenities in China during 2020 compared to the baseline in 2019.


Features

  • Our data is probably the first mobility dataset depicting the diverse mobility response in China during the pandemic from the perspective of amenities.

  • The data shows the changes in visits to six categories of places: Residential, Workplaces, Retail & recreation, Parks, Transit stations, and Grocery & pharmacy.

  • The fine-grained mobility data could supplement the available city-level mobility metrics and fill the data gap in Google Community Mobility Report in mainland China.

  • The data could help design targeted policies to promote outdoor activities and stimulate economic recovery.


Changes in visits to six categories of places in China during 2020

A: Start of the Chunyun period
B: Announcement of COVID-19 human-to-human transmission
C: All provinces in mainland China (except Tibet) launched the highest level of emergency responses
D: 18 provinces lowered the level of emergency response
a: New Year’s Day
b: the Spring Festival
c: Tomb-sweeping Day
d: May Day
e: the Dragon Boat Festival
f: National Holiday & Mid-autumn Festival


Changes in visits to residential/non-residential places in China during 2020

To understand how people balanced their outdoor activities and stay-at-home choices, five non-residential categories (Workplaces, Retail & recreation, Parks, Transit stations, and Grocery & pharmacy) are aggregated into one category and then calculated the corresponding percentage change in the number of visits.

A: Start of the Chunyun period
B: Announcement of COVID-19 human-to-human transmission
C: All provinces in mainland China (except Tibet) launched the highest level of emergency responses
D: 18 provinces lowered the level of emergency response
a: New Year’s Day
b: the Spring Festival
c: Tomb-sweeping Day
d: May Day
e: the Dragon Boat Festival
f: National Holiday & Mid-autumn Festival



About data

Data source

The fine-grained mobility data is aggregated from 210 million geotagged posts uploaded by 10 million Weibo users from 2019 to 2020. Weibo (https://weibo.com), the Chinese version of Twitter, is the most widely used social media platform in China, with 582 million active users in the second quarter of 2022 (41% penetration of the China population). The Weibo social media platform allows users to share their geographical location by posting POIs. Specific POIs can be used to determine the detailed address and geographical coordinates of the sites visited by users.


Data process

  1. Referring to the taxonomy of Google Community Mobility Report (https://www.google.com/covid19/mobility/), the POIs attached to posts were grouped into six categories (Residential, Workplaces, Retail & recreation, Parks, Transit stations, and Grocery & pharmacy) based on their social attributes.

  2. Counting the daily visits to these categories of places for each cities.

  3. Smoothing the daily visits by the method of 7-day moving average.

  4. Calculating the percentage change for each day by comparing it to a pre-pandemic baseline value for that same day of the week estimated over the second half of 2019.


Data description

The mobility data fields are described as follows:

Column Description
ct_adcode city code
city_ch city name (Chinese)
city_en city name (English)
pr_adcode province code
pr_ch province name (Chinese)
pr_en province name (English)
created_at date
category six categories of places (Residential, Workplaces, Retail & recreation, Parks, Transit stations, and Grocery & pharmacy)
visits_7MA_percentage_change percentage change in number of visits compared to baseline days



Download

https://github.com/CASGIS/Socialmedia_Mobility/tree/main/publication/geotagged_data



How to cite

Using these data should cite:
Zhu, K., Cheng, Z. & Wang, J. Measuring Chinese mobility behaviour during COVID-19 using geotagged social media data. Humanit Soc Sci Commun 11, 540 (2024). https://doi.org/10.1057/s41599-024-03050-0



More details

Based on the fine-grained mobility data, our team performed the following analysis, which is described in the paper: Zhu, Cheng & Wang (2024).

  1. To examine the representativeness of users in this data, their spatial distribution and demographic structure was compared to the entire population derived from the national census.

  2. To validate the mobility results externally, the geotagged data was performed post stratified correction and further comparison with Baidu Qianxi data.

  3. The changes in stay-at-home and outings across functional amenities and socio-demographic groups in China during the pandemic were quantified.

  4. Adaptation to mobility restrictions was analysed in terms of visiting diversity and travel distances.



Contact us

wangjh@lreis.ac.cn
zhukaixin@lreis.ac.cn
Team: https://github.com/CASGIS

Institute of Geographic Sciences and Natural Resources Research,
Chinese Academy of Sciences