OBJECTIVITY: BE TRUE TO YOUR DATA

This is one that is difficult for some analysts to follow. In many business analytics projects, there may be some pressure to show findings in a certain way or to ascribe a certain meaning to your data analysis. For example, people in your marketing department may expect that your analysis will show that their recent spend of $2 million on an online media campaign was very effective at driving a large audience, or your product department may pressure your team to show an analysis that proves that its new product results in higher customer loyalty, average order size, and retention rates. It is sometimes tempting to go along with the group and skew the data presentation slightly to most favorably represent the conclusions that the business might like to see. Data visualization techniques offer many opportunities to do this. However, your business analytics team must aggressively avoid this at all costs.

Take a look at Exhibit 6.3, which shows a less obvious (and therefore more insidious) way to create a misleading impression with your data.3 William Playfair is considered by many to be the creator of statistical graphing techniques, and his charts are often considered fine examples of how to convey information through images. However, in the case of Exhibit 6.3, which shows the relationship between wheat prices and worker wages, his chart leaves something to be desired. He shows three time series–weekly wages, wheat prices, and the reigns of British ...

Get Win with Advanced Business Analytics: Creating Business Value from Your Data now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.