Missing data occur when some or all of the values for a sampled unit are absent in the dataset.
Missing Value: A special data item which indicates that the data in this cell does not exist. This may be because the member combination is not meaningful (e.g., snowmobiles may not be sold in Miami) or has never been entered. Missing data is similar to a null value or N/A, but is not the same as a zero value. Multi-Dimentional Array: See Data Cube.
Information that is not available for a particular case (e.g., person) for which at least some other information is available. This can occur for a variety of reasons, including a person's refusal or inability to answer a question, nonapplicability of a question, etc. For useful discussions of how to overcome problems caused by missing data in surveys see Hertel (1976) and Kim and Curry (1977).
Data values that are not known, values bellow the detection limits should not considered as missing.
Data that is incomplete for reasons such as recording errors, participants giving an inappropriate response, or participants giving a response that is difficult to decipher. It does not have to be catastrophic when a participant has one or two missing scores from a task as there are ways with dealing with their overall scores (see the book for details).
Data that the evaluator intended to collect but was unable to for a variety reasons (e.g., the inability to interview a key informant, limited access to a research setting, blank items on a questionnaire, data entry errors).
Data values can be missing because they were not measured, not answered, were unknown or were lost. Data mining methods vary in the way they treat missing values. Typically, they ignore the missing values, or omit any records containing missing values, or replace missing values with the mode or mean, or infer missing values from existing values.