IBM® SPSS® Amos™ 28

The Data Imputation window is used to replace each missing value in a dataset by an estimate called an imputed value. Once each missing value has been replaced by an imputed value, the resulting completed dataset can be analyzed by data analysis methods that are designed for complete data. Amos provides three methods of data imputation.

▪In regression imputation, the model is first fitted using maximum likelihood. After that, model parameters are set equal to their maximum likelihood estimates and linear regression is used to predict the unobserved values for each case as a linear combination of the observed values for that same case. Predicted values are then plugged in for the missing values.

▪Stochastic regression imputation (Little & Rubin, 2020) imputes values for each case by drawing at random from the conditional distribution of the missing values given the observed values, with the unknown model parameters set equal to their maximum-likelihood estimates. Because of the random element in stochastic regression imputation, repeating the imputation process many times will produce a different completed dataset each time.

▪Bayesian imputation is like stochastic regression imputation, except that it takes into account the fact that the parameter values are only estimated and not known. For details on Bayesian imputation, see How Bayesian imputation works.

The Data Imputation window can be used to perform multiple imputation. In multiple imputation (Schafer, 1997) one of the nondeterministic imputation methods (either stochastic regression imputation or Bayesian imputation) is used to create multiple completed datasets. While the observed values never change, the imputed values vary from one completed dataset to the next. Special techniques are required to analyze the multiple completed datasets.

Latent variables do not have a special status in any of the three imputation methods. A latent variable is treated as an extreme case of missing data in which every observation on the variable is missing.

Data files containing imputed values can be saved for subsequent analyses by Amos or any other statistical analysis programs.

See Examples 30 and 31 in the User's Guide for an example of multiple imputation.