Biotech Articles
Publish Your Research Online
Get Recognition - International Audience

Request for an Author Account   |   Login   |   Submit Article

The Peril of Over-Interpreting Correlations

BY: Lisa Opdycke | Category: Healthcare | Submitted: 2013-06-04 06:26:11
       Author Photo
Article Summary: "The American public is acquiring more and more health knowledge thanks to the plethora of information available on the web. Consequently, the average person is more aware of scientific breakthroughs, despite their actual understanding of study findings or implications is slim. However, little in the scientific community has chan.."

Share with Facebook Share with Linkedin Share with Twitter Share with Pinterest Email this article

The American public is acquiring more and more health knowledge thanks to the plethora of information available on the web. While the millennium generation is becoming accustomed to instant internet gratification, their parents are diving into scientific research available online. Consequently, the average person is more aware of scientific breakthroughs, despite their actual understanding of study findings or implications is slim.

Studies are no longer read solely by fellow scientists. However, little in the scientific community has changed in response to larger, more diverse viewership. Yet, there are journalists and untrained individuals trying to grasp new developments in a way that is applicable to the average person. As a result non-scientists (and at times even scientists) are continually reading studies and searching for relationships so that research has larger meaning.

For example, a couple months ago, a physician reported in the New England Journal of Medicine that there is a high correlation between chocolate consumption and the number of Nobel laureates. While I, as a scientist, find this statement absurd, this correlation between Nobel prizes and chocolate was published in major magazines, printed in newspapers, and announced on television. After all, if a physician says it, it must be true, right? Not so much.

The author, Dr. FH Messerli, argues that flavanols in cocoa have beneficial effects on cognitive function. Consequently, Dr. Messerli recommends that increasing chocolate intake can improve individual's cognitive abilities which will in turn increase a nation's number of Nobel laureates. Although this reasoning is fascinating and appealing, it should be taken with extreme caution until you have analyzed the data and then drawn conclusions yourself. Specifically, the correlation was not drawn from data on the actual chocolate consumption of Nobel laureates. Instead, this ecological inference fallacy was drawn on countries' average consumption from the last two years. It should be noted that the Nobel Prize has been awarded for over a century and that chocolate consumption has quadrupled in the last fifty years

This study also neglected to mention that flavonoids, should they actually be the crucial explanatory factor, are found in many other nutriments, such as berries, citrus fruit, tea, and wine. However, the consumption of these flavonoid-rich foods do not show a statistical correlation with Nobel laureates in the same sample of countries.

A publication this week in the Journal of Nutrition set out to demonstrate the ludicracy of Dr. Messerli's extrapolation. These authors found "an incredibly high correlation between the number of IKEA furniture stores and Nobel laureates". While they could not derive a mutual causal relationship, they noted that someone, just as Messerli had done, could claim that IKEA limits its market to countries with Noble Prize winners or that "the need to understand and apply IKEA's furniture assembly instructions improves cognitive functioning at the population level". This is surely something to laugh about; yet, it is based on the same train of thought as the chocolate theory which was reported far and wide.

With both of these examples, we need to keep in mind that correlation does not imply causation. That is correlation, or a statistical relationship, between two variables does not mean that one variable caused the other. Or as the Nutrition article states, "a beneficial effect of flavanoid consumption on cognitive functioning on populations is plausible and certainly sounds more logical than the reverse effect of Nobel awards possibly leading to greater consumption of chocolate".

To demonstrate this principle, each country's Gross Domestic Product (GDP) was used as an indicator of standard of living, which could confer both the level of scientific research and luxury food consumption. As could be expected, the GDP strongly correlates with the number of Nobel laureates and chocolate consumption. Furthermore, the chocolate-Nobel correlation is not more statistically significant than the GDP-Nobel correlation.

These comparisons demonstrate the underlying principle that researches use correlation to show the degree of relationship factors. Correlation cannot be used in place of causation, as in the case of chocolate and Nobel laureates. Instead, correlation is useful in re-evaluating the hypothesis and directing further experiments. But as in the New England Journal of Medicine article, the results were over-interpreted and then shared with the masses, who have little or no experience in scrutinizing scientific data and studies. Consequently, the public is led to believe that consuming mass quantities of chocolate will make them more intelligent or have stronger cognitive function. I ask you, as scientists, to bear in mind that the studies you publish are no longer be solely consumed by your peers. They could end up in Forbes, as this study did, giving the public misinformation. Until science and its publications catch up to new social and cultural conventions, we must each be aware of the ways our data and publications can be misinterpreted.

Pierre Maurage, Alexandre Heeren, and Mauro Pesenti. (2013) "Does Chocolate Consumption Really Boost Nobel Award Chances? The Peril of Over-Interpreting Correlations in Health Studies". J. Nutr. 143: 931-933.

Messerli FH (2012) "Chocolate consumption, cognitive function, and Nobel laureates". N Engl J Med. 367:1562-4.

About Author / Additional Info:
Lisa is originally from a small town in northeastern Indiana. She completed her undergraduate coursework in Human, Biology, Health, & Society at Cornell University in 2011. She received her Bachelor of Science with honors after completing her honors thesis project entitled, "Bovine Seminal Plasma Protein Loss during Capacitation of Fresh and Frozen-Thawed Sperm". Lisa begins her master's studies in public policy at Brown University in the fall following two years at Weill Cornell Medical College-New York Presbyterian Hospital.

Search this site & forums
Share this article with friends:

Share with Facebook Share with Linkedin Share with Twitter Share with Pinterest Email this article

More Social Bookmarks (Digg etc..)

Comments on this article: (1 comments so far)

Comment By Comment
Mauro Pesenti
2013-06-04 02:59:29 640
Thanks for echoing our paper in J Nutr here. As you discussed, our take-home message was threefold: (1) Flavanoid concentration does not fully explain the high chocolate‚Ä"Nobel laureate correlation observed in the initial report as we found no correlation with other nutriments (tea and wine) presenting very high flavanoid concentrations; (2) Correlation is not causation, and it is never valid to infer a causal or directional link between two correlated factors, even when the story sounds plausible; (3) Always be careful when popularizing scientific results as over- and misinterpretation are never far away when a scientific finding is reduced to a simple punchy message. It is the researchers' and scientific correspondents' responsibility not to spread wrong or simplistic news to the lay people.

Leave a Comment   |   Article Views: 2507

Additional Articles:

•   Proliferating Demand For Probiotics

•   Fishing Nets - Evil for Water Birds and Snakes

•   Technologies Involved in Drug Development

•   Possible Factors Influencing Fruit Cracking in Fruit plants

Latest Articles in "Healthcare" category:
•   Health Care and WHO

•   Current Scenario Of Gene Therapy

•   Targeted Cancer Therapy

•   Custom Made Medicine - Pharmacogenomics

•   Nanotechnology and its Application in Medicine - What are Nanoparticles?

•   Red Biotechnology and Cure of Tuberculosis

•   Patho-Biotechnology: Solution to Fight Antibiotic Resistance Bacteria?

•   A Baby Please With a Cheek Dimple...

•   Ribonucleic Acid Interference

•   Biomarker in Cancer Prognosis, Detection and Treatment

•   Adjuvant Therapy For Treating Cancer

•   DNA Microarray and Protein Microarray

•   DNA Vaccine: Vaccine of Next Generation

•   Cancer Immunotherapy Using Monoclonal Antibodies

•   Vaccines to Treat Cancer!

•   Tumor Marker and Cancer Detection

•   Targeted Therapy and Cancer Treatment

•   Enzyme Linked Immunosorbent Assay : A Biotechnology Technique

•   Lycopene as Anti-cancer Agent

Important Disclaimer: All articles on this website are for general information only and is not a professional or experts advice. We do not own any responsibility for correctness or authenticity of the information presented in this article, or any loss or injury resulting from it. We do not endorse these articles, we are neither affiliated with the authors of these articles nor responsible for their content. Please see our disclaimer section for complete terms.
Page copy protected against web site content infringement by Copyscape
Copyright © 2010 - Do not copy articles from this website.

Agriculture Bioinformatics Applications Biotech Products Biotech Research
Biology Careers College/Edu DNA Environmental Biotech
Genetics Healthcare Industry News Issues Nanotechnology
Others Stem Cells Press Release Toxicology  

  |   Disclaimer/Privacy/TOS   |   Submission Guidelines   |   Contact Us