Missing data may negatively impact on the quality of a research study. The need for a suitable missing data method has been a focus of many researchers, and many methods have been developed for handling missing data. The simplest, perhaps the oldest, method is the deletion method, in which the cases with missing values are deleted. The deletion method, however, could result in the loss of a substantial amount of data, which leads to increase standard errors and reduce statistical power. The imputation method, in which missing data are replaced by new statistical estimates, is a better and more commonly used method. Various statistical procedures have been developed for this method, and the mean substitution, regression imputation, expectation maximization (EM), and multiple imputation are a few familiar examples. Because statistical estimations in these methods are based on a statistical summary of the group information, they belong to the group-centered missing data approach. The group-centered methods may not be appropriate when a repeated measure design is employed because available individual information may provide a better estimation. This presentation will provide an overview on commonly used group-centered methods, and their strengths and weaknesses will be described in details. A new individual-centered method for studies with a repeated measure design, as well as new methods for several more complex research designs (e.g., the multi-level research) will also be introduced. Keyword(s): exercise/fitness, measurement/evaluation, research