Table 1.
Summary of missing data categories
| Explainable logic (vs. random) | Identifiable pattern | Affects statistical inference | Action | |
|---|---|---|---|---|
| Structurally Missing Data (SMD) | X | Exclude entry | ||
| Missing Completely At Random (MCAR) | X | Impute | ||
| Missing At Random (MAR) | X | X | Impute | |
| Missing Not At Random (MNAR) | X | X | Impute |
Rows correspond to different missing data scenarios; columns correspond to relevant characteristics. First column stands for the origin of the missing data (randomly generated or not). Second column states if a pattern of missing data can be recognized in the data set. Third column shows the impact on statistical inferences. The last column shows the action that needs to be taken to address the missing data problem