Accurately assessing who lives in a household is often a critical step in both identifying the randomly selected respondent and understanding household composition and dynamics more generally. This is particularly important for large government surveys of social infrastructure. Many surveys begin by asking a series of questions aimed at generating a roster of individuals that compose the target household. However, in practices, there is substantial variation in how these rosters are assembled, which may result in different definitions of what constitutes a household across different surveys and lead to confusion. This chapter reviews some of the most common methods of rostering and suggests future research focused on harmonizing and standardizing these methods across different surveys.