Theoretical properties of nonresponse adjustments based on adjustment cells are studied, for estimates of means for the whole population and in subclasses that cut across adjustment cells. Three forms of adjustment are considered: weighting by the inverse response rate within cells, post-stratification on known population cell counts, and mean imputation within adjustment cells. Two dimensions of covariate information x are distinguished as particularly useful for reducing nonresponse bias: the response propensity p(x) and the conditional mean y(x) of the outcome variable y given x. Weighting within adjustment cells based on p(x) controls bias, but not necessarily variance. Imputation within adjustment cells based on y(x) controls bias and variance. Post-stratification yields some gains in efficiency for overall population means, and smaller gains for means in subclasses of the population. A simulation study similar to that of Holt & Smith (1979) is described which explores the mean squared error properties of the estimators. Finally, some modifications of response propensity weighting to control variance are suggested. /// L'estimation de moyennes de la population et de sous-populations pour une enquete soumise a nonreponse est discutee. Trois methodes d'ajustement pour nonreponse sont compares: le weighting, la post-stratification et l'imputation de moyennes en classes d'ajustement. Deux dimensions des covariates x sont distinguees pour la reduction du biais de nonreponse, la tendence de reponse p(x) et la moyenne conditionnelle E(y[divides]x) du variable y sujet a nonreponse. Les caracteristiques de methodes qui font usage de ces dimensions pour la creation de classes d'ajustement sont recherchees au moyen de theorie et de simulation.