it is impossible to disaggregate them from Iranian Kurds and Vietnamese of Chinese origin. The ACS estimates also cannot be used to distinguish ethnic Somali Kenyans whose ancestors arrived in Kenya in the 1930s from those whose ancestors were in Kenya prior to that decade, who may have a lower incidence of statelessness. Likewise, these estimates cannot be used to differentiate between Thai-born ethnic Hmong, Lahu, Yao, Lisu, and Shan individuals with immediate ancestry in Laos or Myanmar, and members of these groups with immediate ancestry in Thailand, who may have a lower rate of statelessness, or between ethnic Nepalese of immediate Bhutanese ancestry and ethnic Nepalese with immediate ancestry in Nepal. In addition, this data does not allow the identification of individuals born in Pakistan and Sri Lanka who were stateless in India before arriving in the United States, as the ACS lacks data on country of last habitual residence for individuals who arrived in the United States more than one year prior to the survey. Nor does the data permit the accurate identification of Indian-born individuals of Pakistani origin from Jammu and Kashmir, since the ACS does not record region of birth within India. As a result, CMS excludes this profile from its final estimate. Finally, ACS data does not permit CMS to distinguish between Feyli Kurds and other Kurds born in Iraq. CMS’s estimates of Feyli Kurds may include some non-Feyli Iraqi Kurds that possess Iraqi citizenship. Moreover, many non-US citizens born in the countries set forth in the CMS profiles did not list their ancestries or they listed ancestries coded as “Other” in the ACS. About 403,000 of a total 5.5 million (about 7.4 percent) born in these countries did not list any ancestry and 20,100 listed “other” ancestries. Some of these individuals may match the criteria in CMS’s profiles. Consequently, this analysis may fail to capture small numbers of persons who are potentially stateless or potentially at risk of statelessness.49 Finally, the ACS does not permit the identification of persons who are potentially stateless or potentially at risk of statelessness who were not registered in their home countries at birth but do not belong to a minority ethnic or linguistic group. In particular, CMS cannot identify Bidoon that are not ethnic minorities in their countries of residence or Chinese-born individuals without hukou registration. CMS’s analysis of ACS data reveals 198,000 individuals who match its profiles of persons who are potentially stateless or potentially at risk of statelessness. To account for potential inaccuracies due to survey error, CMS also calculated upper and lower bounds for this estimate. CMS estimates that the number of persons who are potentially stateless or potentially at risk of statelessness, as defined by the above profiles and accounting for the data limitations described above, falls between 192,000 and 204,000 persons.50 49 About 100 persons meeting the criteria in CMS’s profiles also listed “other” ethnicities as secondary ethnicities. The majority of these persons were born in Soviet successor countries. In this analysis, CMS considers these individuals to be potentially stateless or potentially at risk of statelessness. 50 Specifically, CMS used Census Bureau-defined “replicate weights” that account for ACS sampling methodology in order to construct a 95-percent confidence interval around the 200,000 estimate. 43