The use of graphics in statistics can misrepresent data if they are used wrongly or for the wrong reasons. A good use of graphs in statistics is to plot them in order to have a demonstrative data set for the variables. The tendency of a variable or distribution is demonstrated using different types of graphs. The misrepresentation of data comes about when the researcher does not know the best type of graph to use for a particular variable (Hill 10). The lack of knowledge on the best graph to use with a specific set of data leads to random plotting of graphs for any data set. The result of this is a wrong impression about the true meaning of the data set. The other likely misrepresentation of data in graphs is when the researcher changes the test that they used and the set of data, with a view to match the graphical display with the results of the tests.

Misrepresentation of data in statistics can also occur as a result of an error done during compiling of data. This error may come about when recording the data using the previously recorded data in the stage of secondary compiling (Ercan et al. 131). In this stage, the researcher may miss the exact variable they want to examine in the recordings. In addition, they may find the variables but they may have been measured in scales that are different. This may bring a need to change the structure of the data, or even to increase the number of data. Such errors may also occur when a clear definition of variables is not made. Censoring and truncating data may also be a source of error in statistics. When data with such errors is represented graphically, it will be a misrepresentation of the true nature of the results.

Works Cited

Ercan Ilker, Yerzici Berna, Yang Yaning, Ozkaya Guven, Cangur Sengul, Ediz Bulent and Kan Ismet. Misusage of statistics in medical research. European Journal of General Medicine, 4.3 (2007): 128-134. Print.

Hill, McGraw. Chapter one: what is statistics? n.d. Web. 17^th Feb, 2014.