BUS 625

Question 3(Catherine)

There are several methods to display data in ways that make it easy for the consumers of that data to understand. While things like pie charts and bar graphs allow us to visualize variances for one data point, a table chart allows us to look at multiple data points at once. Tables utilize rows and columns to keep these data points organized.

A table would be preferred in situations when lots of information needs to be organized. Maybe an organization wants to know how many people signed up for a new credit card in the past six months. A table would allow the organization to display information like the name of the person, when they signed up, how they signed up and how much they have spent, to name a few examples. Another type of table that might be easily recognizable is an income statement. A table on an income statement would allow a financial manager to display what the money is for, the amount of money, and the year the company earned or spent that money.

A good example of a table is the segment reporting from the Starbucks Q2 Financial Report from 2021.

Other charting techniques that can be used to visualize data are pie charts, bar charts, segmented bar charts, mosaic plots, line graphs. All of these charts serve different purposes, and some may be more appropriate than others depending on the situation. Pie charts show a total divided into categories by showing a circle divided into wedges that correspond to different variables. Bar charts represent percentages displayed as bars. Segmented bar charts are a variant of bar charts but allow the user to display more conditions or variables in their chart. Mosaic plots is a representation of a contingency table that divides the information into rectangles proportionate to the number of cases in the cells around it. Finally, a line graph shows information over time by connecting points with lines.

Each of these charts have their uses but choosing the wrong one might create the wrong visual. If a financial advisor wanted to display the growth of sales over time, they would not choose a mosaic plot or a pie chart.

The Simpson’s paradox is a phenomenon that occurs when multiple percentages are taken across different groups and then the group averages of these percentages appear to contradict the overall total. A good example of this paradox frequently occurs in the batting average of baseball players. They can bat consistently high in separate years, but these batting averages can appear low when you begin to combine years.

Question 4(Andrew)

Choosing the correct data visualization technique is critical to ensuring the data being presented is understood.  One data visualization technique is a table.  Tables are the preferred method of presenting data when specific, precise data needs reviewed.  Tables allow the audience to examine the exact values and are especially useful when multiple dimensions could make charts confusing (Blitz, 2018).  An example of when a table would be helpful is when presenting financial data to a Finance team.  Since Finance teams are detail-oriented, they often want to see exact numbers to analyze the trends.  Below is an example of good use of a table showing financial data for Carvana from 2015 through 2019.  This example is a good use of a table as it allows the audience to see the increased revenue and gross profit over the five years.

(Source:  Carvana Co, 2019, p 46)

                The five charting types that I have chosen are line graphs, dot maps, word cloud, bar charts, and pie charts.  Line graphs show quantitative values over time and easily show trends by connecting the data points with a line.  The visual of seeing a line go up when a number increases or go down when a number decrease makes an excellent graphic.  However, line graphs can get confusing when there are more than three or four lines (Line Graph, n.d.).  Dot maps show the distribution of data over a geographical area.  However, they do not do an excellent job of showing the representation of exact data, especially when data points are close together (Dot Maps, n.d.).  Word clouds show the frequency of a word’s usage by increasing the size of the word in the cloud.  While the final image can be aesthetically pleasing, the results can look skewed by some word types.  For example, longer words or words with ascending and descending letters show up more prominently and may look to be more represented (Word Cloud, n.d.).  Bar charts show data on a horizontal or vertical column and are suitable for showing how many items are in different categories.  However, labeling can be confusing when there are a large number of categories (Bar Chart, n.d.).  Finally, pie charts show the percentage of the whole data.  This chart is a good chart for a quick visual where the total equals 100 percent.  However, when there are too many categories, the pie slices can get small and distort the visual (Pie Charts, n.d.). 

                While all these charting types have advantages and disadvantages, some are more suitable for displaying data.  Line charts, bar charts, and pie charts are suitable charting methods to show data because the audience can see the exact amounts easily.  However, the dot maps and the word cloud are not suited for displaying data because the imagery used can distort the data.

                The Simpson’s paradox is when two or more averages, or percentages, are grouped from different groups.  The paradox occurs when the total averages contradict the separate group averages (Sharpe, De Veaux, Velleman, 2019, p 39).  An example of Simpson’s paradox is a study of SAT scores between 1992 and 2002, which showed that each GPA group decreased their SAT score, while the overall SAT score for the same group stayed relatively flat.  Simpson’s paradox occurred in this situation because of grade inflation that occurred between 1992 and 2002.  The grade inflation caused the brightest students from each group to go up a GPA group.  For example, the top of the B+ students in 1992 would be an A- student in 2002.  Therefore, the former B+ student would reduce the A- student’s average SAT score.  However, the overall SAT scores were staying consistent (Sprenger, Weinberger, 2021).