This report proposes an empirical Bayes approach to the problem of equating scores on test forms taken by very small numbers of test takers. The equated score is estimated separately at each score point, making it unnecessary to model either the score distribution or the equating transformation. Prior information comes from equatings of other test forms, with an appropriate adjustment for possible differences in test length. Derivation of estimates for y[subscript prior] and var(y [subscript prior]) is appended. (Contains 4 notes.)