Cleaning data can be a lengthy process that is very time consuming, especially with complicated data. In my opinion, I believe that there should be some sort of data cleaning in order for the article to have more structure and make more sense to the reader. On the other hand, I do not think that an article needs to be completely straightened up. There should be some sort of messiness to an article in order for it to be relatively interesting and easier to read. Likewise, I think there should be some sort of visualization to help the reader understand the data. Data cleansing can allow for many margins of error and create a sense of confusion within a data set. For example, How to Lie with Statistics demonstrates this by portraying that statistics can be used to sensationalize, confuse, and oversimplify information. Additionally, over cleansing your data can allow for comparisons and extreme exaggerations. Furthermore, the article shows us when you are cleansing data, critical thinking is important and crucial for success. Another source, Data Visualization: Ambiguity as a Fellow Traveler, demonstrates how data can have missing parts. The article mentions how refinery is a good way to keep track of analyses performed on a data set. It goes on to talk about the importance of visualization within statistical data sets. “Visualization methods have to keep up with large data sets that are big, complex, and noisy, but they can not replace statistics.” (Marx). In other words, visualization within data sets is crucial. However, they cannot replace regular data. All in all, data should be cleansed to a certain point in order to be easily understood.