Please ensure Javascript is enabled for purposes of website accessibility

Manipulating Statistics in Data


Statistics includes various aspects of analysis, such as the collection of data as well as its interpretation. In the media and throughout the statistical analysis process, there are known methods to intentionally influence the data to provide statistics to favor an author’s viewpoint.

Data manipulation can start with its collection. Suppose you searched online for the percentage of the population in favor of the legalization of cannabis. In that case, you’d likely receive ten different answers from the first ten sites you visited. This scenario is an example of sampling bias, a form of statistical manipulation. The sampling group’s sample size, location, and population can contribute to misleading results. Sampling a small number of people from one region might paint an accurate picture of that group’s opinion of cannabis legalization but might not be representative of the country.

Data manipulation is also common during the analysis phase. One example is to get a simple number: an average. Different statistical measures such as mode, median, or mean can all be used as an average and provide vastly different results on the same dataset. The validity of each measure might be technically accurate yet misleading when analyzing the whole picture. For instance, take a random community trying to determine how often people use cannabis in a week. Out of 50 people, 40 never use cannabis. However, if the other ten people use cannabis products five times a day, the statistical measures to determine an average would have vastly different answers.

Data can also be influenced through its delivery. Using incorrect graph types can intentionally or unintentionally mislead the reader. Altering a chart’s axis and using disproportionate scales can be used to fit the author’s intention. With different tactics to manipulate a visual, it is crucial to be aware of the unconventional methods used to display the data.

Data Manipulation using Graphs

There is an abundance of ways to manipulate data. Whether or not a statistic you see in the news or in an article is intentionally influenced or not, it is important to be aware of the possibilities. In today’s world, there are endless amounts of data to consume. Checking sources, looking at data collection methods, and examining any visuals to look for apparent misleading modifications are good starting points to ensure you are prepared to spot manipulated statistics.

About the Author Nick Purdom

Nick attended the University of Kentucky where he received a bachelor’s degree in Computer Science. Nick is currently attending Indiana University to earn a master’s degree in Data Science. Nick will assist the analytics team by utilizing his knowledge of Python, SQL, and Excel. Nick likes to spend his free time involved with sports, specifically basketball.

Recent Insights

Receive Key Insights