Filtering Missing Values from Analyses

When variables contain missing data, any analyses of these variables need to exclude the cases with missing data. This is done by just about all software, provided that the missing data has been assigned an appropriate missing value code.

Note that excluding missing values from data is only valid if it is believed that the data is representative when this data is excluded. If this assumption is not likely to be true, it is appropriate to instead:

• Impute the missing values.
• Use methods designed for missing data.