Statistics is a discipline and it combines mathematical and non-mathematical procedures. It helps people to get conclusions about a greater population is possible by using a limited sample. There are different kinds of ways such as tables, graphs, and charts and all those are really important for presenting the data in order to draw conclusions.
During my university education, mostly I had knowledge theoretical about the purpose of statistics, theorems, history of statistics, formulas, and roots of formulas. As one would expect, statistics is largely grounded in mathematics, and in order to understand the behind of the statistics you need to know some concepts such as probability, estimation, and data analysis. However, as I mentioned there are non-mathematical methods such as data collecting, coding data, reporting results, and summarized data. I would like to give you an example of how statistics are like. For example, when you read a book, you can feel or predict what the ending will be like based on a context. It is similar to statistics because, with the fields and information, you can have accurate thoughts about the flow or end of the book using a small amount of information. It is like the population and sample relationship. Basically, most of us can summarize the data and predict the conclusion of daily basis facts. Visualization is also significant as much as data collection and preparation of cases because if you cannot visualize your data, there is no way to transfer your results to audiences in the proper way even your cases are prepared in the proper way. However and unfortunately, even though there are a bunch of people who understand statistics, some of them are not able to implement the correct statistical methods. In this article, I would like to explain some most common data visualization mistakes.
Data Visualization Mistakes
This is one area that can give a nightmare to both parties the presenter as well as the audience. Incorrect data presentation can skew the inference and can leave the interpretation at the mercy of the audience.
Bar charts are the most common charts among others as they are more simple to create. I remember a day that I was taking visualization class and our instructor said almost 99% of statisticians are using Excel and bar charts to show their results even they know another visualization tool such as PowerBI and Tableau. I think this is true! Okay even it is the most common charts that used, we need to know when we need to use it. It uses in order to show the categorical data by the number or percent for a particular group and there are several variations of the standard bar chart including horizontal bar charts, grouped or component charts, and stacked bar charts.
It summarizes the distribution of a univariate data set and based on it we can get statistical information. This statistical information includes the mean value, the maximum and minimum values.
I think some problems occur because of selecting a pie chart instead of other charts. A pie chart is best used when trying to work out the composition of something. There might be some confusion about when should we use a pie chart. One of the only good use of a pie chart is to show the relationship between parts or percentages of a whole. But, for sure the slices should be equal and if some of them are larger than the rest of the slices, it makes the pie chart vague. Also, another important point is that percentages should add up to 100%. I would like to warn you if you still the persistent person to use it is that never use a pie chart if it has more than 5 slices. Instead of using a pie chart, you can prefer to use a bar chart and in this way, you can understand which one is better to present your results.
A time chart is used to show how the measurable quantities change by time. Your questions are important in order to show why do you use the time chart, that is why you need to ask ‘ what is happening’. Another point to remember is that if there is no data, your chart should show empty spaces.
The basic definition of when we need to use a line chart is that if you have continuous data that you would like to represent through a chart then the line chart is a good option. Especially, if you want to identify a trend and pattern in your data, it fits well. For example seasonal effects over time.
We know that the charts and graphs are ready for us to help our questions, but the problem is that it is not easy to know which one of them is the best. We need strong visualization to tell the most truthful story and that is why first, we need to know about data types and then we can decide what kind of graphs we need to use. For sure, in order to decide the graphs types, we have to know the differences between graphs and charts.
Lastly, I would like to give you a few examples of data visualization tools. PowerBI, Tableau, Sisense, IBM Cognos Analytics, and SAP Analytics Cloud are few of them. Also, Python and R programming languages are good options for creating visuals as well.