One major problem complicating the task of effective humanitarian protection is the lack of quality data on the populations most affected. If protection agencies cannot identify those who need help, then their ambitions of assisting them are unlikely to be realized. This is especially relevant when considering "invisible," hard to reach, or historically marginalized groups for whom we have little baseline data and whose presence is a source of contention for national authorities.Unfortunately, undercounting is not simply a matter for statisticians and social demographers. It is often a political matter. As Sarfaty writes, "numbers display governmentality because they serve as a technology of power that constitutes populations and makes individuals calculable and therefore governableboth by others and themselves." 1 Who is counted also tells us about governmental and institutional priorities and exposes biases about what counts, and how resources should be allocated. For example, voter registration may not include the total adult citizens because African Americans have been repeatedly denied the right to register to vote in some US states. Official processes may be exploited to discriminate against certain groups and published data may inaccurately reflect population trends.The rationale for collecting data also reveals political and institutional priorities. National statistical offices play an essential role in governmental planning and are central to the state's claim to legitimacy. Similarly, international organizations, including the United Nations High Commissioner for Refugees (UNHCR), have * The author would like to thank the editors, Molly Land, Kathy Libal, and Jillian Chambers, as well as Alessio D'Angelo, Jeff Crisp, Margaret Okole, and Don Kerwin for their helpful advice on earlier versions of this chapter. I am also grateful to Bronwen Manby for her helpful comments on this chapter.