Missing Data

In this paper، we are presenting the basic concepts of missing data in a very simple but practical approach. Missing data are ubiquitous throughout the social، behavioral، and medical sciences. In Statistics، missing data occur when no data value is stored for the variable in the current observation. Missing data reduce the representativeness of the sample and can therefore distort inferences about the population. Missing data are a common occurrence and can have a significant effect on the conclusions that can be drawn from the data. All the methods of parameters estimation are based on the completion of data set assumption and only in this case the result will be a non- biased one، and with the increase of missing proportion، the rate of biased results increase too. For decades، researchers have relied on a variety of old techniques that attempt to “fix” the data by discarding incomplete cases or by filling in the missing values. Unfortunately، most of these techniques require a relatively strict assumption about the cause of missing data and are prone to substantial bias.