Dear All, I am new to this listserve and to multiple imputation in general. I have what might be a foolish question about using multiple imputation to estimate values for a variable that has not been measured at all phases of a longitudinal investigation. When data are arranged in a multivariate format (i.e., all observations for each sample unit are placed on a single line), it is clear that multiple imputation cannot be used to estimate missing values because you would have empty columns for those phases in which a variable was not measured. However, when data are arranged in a univariate format this is not the case. In the example below, all values for Variable A at phases 2 and 3 are missing. But because the data are arranged in a univariate format, there are no empty columns for variable A.
My question concerns whether or not it is possible to estimate the missing values for Variable A using multiple imputation once data are arranged in a univariate format. The imputation program that I am most familiar with is NORM. The documentation for this program seems to indicate that data must be arranged in a multivariate format. However, I was hoping that there might be other programs that will accept data in the univariate format depicted below. Assuming that this is the case, would it be reasonable to estimate missing data points in the manner I have described? Or does this go well beyond what any multiple imputation program can be expected to do? Any help that people can provide will be most appreciated. Paul (Note: Initially, I had decided to insert a table into this e-mail message so people could see my example data. However, I now understand that some people may not be able to read the table using their e-mail program and so I have typed out what should appear in each column of the data set as well). Column 7: x, x, x, x, x, x, x, x, x, x, x, x, x, x, x, x(Variable D) Column 6: x, x, x, x, x, x, x, x, x, x, x, x, x, x, x, x(Variable C) Column 5: x, x, x, x, x, x, x, x, x, x, x, x, x, x, x, x(Variable B) Column 4: x, _, _, x, x, _, _, x, x, _, _, x, x, _, _, x(Variable A) Column 3: 1, 2, 3, 4, 1, 2, 3, 4, 1, 2, 3, 4, 1, 2, 3, 4(Phase) Column 2: 1, 1, 1, 1, 2, 2, 2, 2, 1, 1, 1, 1, 2, 2, 2, 2(Spouse) Column 1: 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2(Couple) Couple Spouse Phase Variable A Variable B Variable C Variable D 1 1 1 X X X X 1 1 2 X X X 1 1 3 X X X 1 1 4 X X X X 1 2 1 X X X X 1 2 2 X X X 1 2 3 X X X 1 2 4 X X X X 2 1 1 X X X X 2 1 2 X X X 2 1 3 X X X 2 1 4 X X X X 2 2 1 X X X X 2 2 2 X X X 2 2 3 X X X 2 2 4 X X X X --------------------------------- Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.utsouthwestern.edu/pipermail/impute/attachments/20021217/7713a2e0/attachment.htm
