Graphics Mapping ethnic migration in mainland China from the early 1960s to 2010 Jianzheng Liu, Jie Li and Weifeng Li The Department of Urban Planning and Design, The University of Hong Kong, Hong Kong Jingsi Shaw Department of Urban Studies and Planning, Massachusetts Institute of Technology, Cambridge, MA, USA Featured graphic. Mapping ethnic migration in mainland China from the early 1960s to 2010 This study aims to map ethnic migration activities associated with a significant change in the mix of ethnicities for a given area in China from the early 1960s to 2010, and to ascertain where and among which ethnic groups such activities were prevalent in the past 50 years. To our knowledge, this is the first effort toward a nationwide mapping of ethnic migration in China in the past 50 years that has been undertaken. To achieve this, we utilized the Geo-referencing of ethnic groups (GREG) dataset, a widely used ethnic distribution dataset developed by Weidmann et al. (2010); the LandScan 2010 population grid; and 2010 census data for China. The original data in the GREG dataset are drawn from the well-known 1964 Soviet Atlas Narodov Mira. The methodology used builds on the following rationale: if there is no ethnic migration, then for each administrative region such as a county, the estimated population for each ethnic group using GREG and LandScan should be consistent with the ethnic population census data within that county. Otherwise, we can assume that there have been several ethnic migration activities during the analysis period. We thus overlay the GREG data with the national county administrative map and census- adjusted LandScan 2010 to derive the estimated ethnic population for each county in 2010 and compare this data array with the 2010 population census data at the county level and compute Pearson’s r. If there has not been significant ethnic migration from the early 1960s to 2010, Pearson’s r should be very close to 1; otherwise, it should be closer to 0. The first map shows trends in ethnic migrations activities during the analysis period using a rule of thumb for interpreting correlation coefficients by Hinkle et al. (2003) (Figure 1). For counties with moderate, substantial, and extensive ethnic migration activities (Pearson’s r lower than 0.7), we identified major ethnic groups that penetrated the living areas of other ethnic groups as well as ethnic groups whose living areas were penetrated by other ethnic groups. Based on the results, we drew up five maps for the Han, and identified four ethnic Corresponding author: Weifeng Li, Department of Urban Planning and Design, The University of Hong Kong, Pokfulam Road, Hong Kong. Email: wfli@hku.hk Environment and Planning A 0(0) 1–3 ! The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0308518X15616632 epn.sagepub.com