"I think part of conducting data analysis is accepting that issues are going to crop up. One needs to try to avoid these problems in the first place, but if they do arise, it's a matter of knowing how to handle them so ultimately your end result is accurate. One issue that frequently comes up is duplicate data entries and spelling mistakes. This is only natural given that people make mistakes when they enter information into databases. I run validation rules to catch and eliminate these kinds of errors. Sometimes the data source isn't reliable, in which case much more time needs to spent cleansing the data. To avoid this, I try to use only reliable and quality sources to obtain my data. Incomplete data creates problems, and another common issue occurs when data is gathered from multiple sources. In this case, it's a matter of structuring the various datasets so they are all compatible and don't cause any delays when I perform the analysis."