The aim of the study was to discover a satisfactory predictor of residential mobility among inner city families in light of the effect which the high rate of student transfer has upon curriculum development, staff morale, and student scholastic achievement. The sample of 358 families was administered a questionnaire dealing with demography, mobility, perceptions of education, and continuous learning. Relationships between previous mobility and other variables were studied; previous mobility, present address, and respondent's estimate were examined as predictors of mobility. Findings show that, when predicting mobility over a one-year period: (1) opinions of people themselves as to whether they would move is not a reliable indicator of whether they actually do so; (2) used together, previous family mobility and its present accommodations give the best indication of whether the family will move within one year; and (3) predicting future mobility on the basis of past mobility and present accommodations will give different results for different ethnic groups. Extensive appendices are included. (Author/SES)