This paper describes the Bureau of Mines development of a novel subsidence prediction methodology suitable to the mining and geologic conditions in the northern Appalachian coal region. It describes the computation of vertical and horizontal movements, inclination, curvature, and horizontal strains. The substance of this method is the separation of the effects of lithology by introducing a correlation between hypothetically homogeneous overburden and existing lithologic conditions, while providing for different mining conditions such as underground geometry and overburden thickness. The effects of lithology have been expressed in the form of a variable subsidence coefficient within the subsidence trough. The subsidence coefficient is considered a constant for other predictive methods. Field data from 16 test sites at 11 Bureau longwall panel studies were used in the analysis. For each panel, the characteristics of the variability of the subsidence coefficient were defined. Regression analysis of the subsidence coefficients from all test sites on their locations relative to the edges of their respective panels yielded a third- degree polynomial equation. Results from additional longwall panel studies not included in the regression analysis were used to prove the validity of this method. To facilitate the use of this precalculation methodology, a computer program was written in basic for use on a personal computer.