Pictures help us make sense of data.
For example, when paleontologists unearth a single tooth, they can tell a lot about the animal it came from, including whether or not it’s a new species.
When they publish their findings for their peers, they will include pages and pages of data. But when the discovery is announced to the public, it will likely be accompanied by a highly detailed illustration of the animal in its natural setting. The pictured animal is based on the data, but if you didn’t know better, you’d think the paleontologists observed the animal in the wild—not just examined a single tooth.
A chart or graph can be a useful way to look at data. But again, it’s essentially a picture based on a careful selection of the data.
By choosing the units measured, the scale, and the density of the chosen information, the person making the graph can cause it to appear to support a conclusion that may not really be borne out in the data.
This is especially true for charts that illustrate a causative relationship between two things.
Researchers at Northwestern University conducted several studies to see how data-visualization techniques (different kinds of charts) tend to influence people’s perceptions of the same data.1
The first thing they found, unsurprisingly, was that charts and graphs can make people believe more strongly in a causal relationship between two factors, even though the data only shows a simple correlation.
For example, research has shown that kids who eat breakfast more often have better grades (correlation). But people viewing the right kinds of graphs will be convinced that eating breakfast causes better grades (causation), a result that has not been proved.
A second finding was that the less detail a graph shows, the more convincing it is of causation. This held true regardless of the type of graph used—bar, line, or plotted points. The more simple the graph you show people, the more they are swayed to your conclusion.
Financial charts and graphs can also be misleading. An investing guru with something to sell can choose just the right data framed with just the right parameters to make their speculations seem quite convincing. They are showing you “the facts.” Just not the whole picture.
As British economist Ronald H. Coase said, “If you torture the data long enough, it will confess to anything.”2
It’s good to be educated about what’s going on in the market. But when you come across charts and graphs that purport to show how some future price movement or investment outcome is all but guaranteed, it’s wise to exercise caution and healthy skepticism.
Sources: 1. http://go.pardot.com/e/91522/ns-false-assumptions-causality/77hy5p/1456313528?h=IoHCmpC5RKqr75qjNx5kI8sdOj9jFsPJYxGO4msbyzA 2. http://go.pardot.com/e/91522/nfess-to-anything-492786c30169/77hy5s/1456313528?h=IoHCmpC5RKqr75qjNx5kI8sdOj9jFsPJYxGO4msbyzA Disclosure: The views expressed herein are exclusively those of Efficient Advisors, LLC (‘EA’), and are not meant as investment advice and are subject to change. All charts and graphs are presented for informational and analytical purposes only. No chart or graph is intended to be used as a guide to investing. EA portfolios may contain specific securities that have been mentioned herein. EA makes no claim as to the suitability of these securities. Past performance is not a guarantee of future performance. Information contained herein is derived from sources we believe to be reliable, however, we do not represent that this information is complete or accurate and it should not be relied upon as such. All opinions expressed herein are subject to change without notice. This information is prepared for general information only. It does not have regard to the specific investment objectives, financial situation and the particular needs of any specific person who may receive this report. You should seek financial advice regarding the appropriateness of investing in any security or investment strategy discussed or recommended in this report and should understand that statements regarding future prospects may not be realized. You should note that security values may fluctuate and that each security’s price or value may rise or fall. Accordingly, investors may receive back less than originally invested. Investing in any security involves certain systematic risks including, but not limited to, market risk, interest-rate risk, inflation risk, and event risk. These risks are in addition to any unsystematic risks associated with particular investment styles or strategies.