The relative precision of four methods of estimating missing data in principal components analysis was investigated. Artificial data with known characteristics, obtained from Cattell's "Plasmode: 30-10-4-2," was used with one third of the data on half of the variables being systematically eliminated. The four methods of missing data estimation were: means substitution, simple regression, stepwise regression, and multiple regression. In order to extract all possible variance, the Principal Components Analysis was employed without rotation. Factor scores from complete data and each of the estimated data solutions were obtained. Goodness-of-Fit was judged on the basis of cross-correlations of each estimated data solution with the solution derived from complete data. The study showed that all four methods of estimation compared fairly well with the criterion. The average correlations improved from the method using least concomitant information (means substitution) to that employing most (multiple regression). Indications of the study were that means-substitution may be a viable method of estimating missing data . (AE)