The present study examines the direct and indirect relations between residential mobility during the year prior to kindergarten and academic skills (math, letter-word identification, and spelling) in the early elementary years for children from low-income families. Data were obtained from the Head Start Impact Study. The sample included 3627 children (50% male) who were eligible for Head Start. At base-line, children were on average a little over 4-years-old ([mean] = 49.12 months, SD = 6.79). The sample was ethnically/racially diverse and had a range of maternal education levels. Residential mobility was conceptualized as either not moving (70% of the sample) or moving one or more times (30% of the sample) during prekindergarten. Data were collected at baseline (fall of 2002) and in the spring of prekindergarten, kindergarten and 1st grade. When controlling for a set of demographic, family, and child covariates, results from three separate path analyses indicated that mobility had consistently negative associations with children's academic skills but that these links were quite small, especially when looking beyond the prekindergarten year. Mobility during prekindergarten had small direct associations with all three skill areas (math, letter-word identification, spelling) at the end of prekindergarten, and significant but even smaller links with two of the three skills at the end of kindergarten and 1st grade. In addition, mobility had small indirect relations with all three skill areas in kindergarten and 1st grade, mediated through these same skills in prekindergarten and kindergarten. Implications of study findings for supporting low-income families and directions for future research are discussed. (author abstract)