logo for solving-math-problems.com
leftimage for solving-math-problems.com

Coffee Statistics - Math 126

by Cody
(Manchester, NH, USA)










































In an article in USA Weekend, this statement was made: :More serious seems to be coffee's potential to raise blood pressure levels of homo-cysteine, a protein that promotes artery clogging. a recent Norwegian study found 20% higher homo-cysteine in heavy coffee drinkers (more than 9 cups a day) than in non-coffee drinkers." Based on this statement, should we give up our daily cup of coffee?

Comments for Coffee Statistics - Math 126

Click here to add your own comments

Apr 07, 2012
Coffee Statistics - Math 126
by: Staff


Part I

Question:

by Cody
(Manchester, NH, USA)


In an article in USA Weekend, this statement was made: :More serious seems to be coffee's potential to raise blood pressure levels of homo-cysteine, a protein that promotes artery clogging. a recent Norwegian study found 20% higher homo-cysteine in heavy coffee drinkers (more than 9 cups a day) than in non-coffee drinkers." Based on this statement, should we give up our daily cup of coffee?



Answer:

Based on this statement, should we give up our daily cup of coffee?

I wouldn’t. USA Weekend is trying to sell more issues of their magazine by using inflammatory language.

According to the article, if you’re only drinking one or two cups of coffee each day, the homocysteine levels should not increase significantly. The article states that a 20% increase in homocysteine levels occurred when drinking over 9 cups of coffee a day.

It is worth noting that the article does not state how many variables were taken into account when the study was completed. Unless only one variable was scientifically tested, the article is nothing more than anecdotal evidence.

For example, were the participants in the study doing other things which could increase homocysteine level (such as smoking cigarettes, eating large meals each day, or not exercising)? Unless all other variables can be ruled out, the results of the study are invalid.

The article does not define what “more serious” means. More serious than what?

The article does not say what methodology was used in the study, or how many patients were studied (did the study compile results from 5 patients, or 50,000 patients?). What were the control groups: children, old men, vegetarians, athletes, people from a specific location (such as Oslo, Norway), etc?

The article is vague about the effects of homocysteine. The article does not say what the exact relationship is between elevated levels of homocysteine and artery clogging.

----------------------------------------------------------

Apr 07, 2012
Coffee Statistics - Math 126
by: Staff

----------------------------------------------------------

Part II

There are also several other things about this advertisement that create a credibility gap.

First, the ad does not tell the reader who conducted the “study” in Norway. Was it a nationally recognized medical center, or the corner drugstore?

Second, the words “recent study” do not tell the reader when the study was conducted. The word “recent” could mean anything from yesterday to 50 years ago.

The confidence interval for the results of the study is not given. The confidence interval could be 95% or 25% - or anything else.

IMPORTANT: the advertisement does not tell the reader HOW MANY STUDIES were CONDUCTED.

The number of studies is important. If you want research to confirm the results you decide on in advance, it is not hard to misuse statistics to prove your point – no matter what your point is.

For example, if product ABC was completely useless with a 95% confidence interval that means 2 ½ % of the time a study will show it is effective. A study may actually show the product to be highly effective ½ % of the time.

To promote a completely useless product, all the seller needs to do is conduct one study after another UNTIL one study shows the product is superior. The seller (or distributor) can then selectively report that particular study without telling the reader about all the other studies which show the opposite result.

But, . . . there is an even easier way to get the statistical results you want. Change the definitions of the words used in the study.



The following website (named “Bad Science”) highlights the negative consequences of using statistics improperly:

http://www.badscience.net/





Thanks for writing.

Staff
www.solving-math-problems.com




Click here to add your own comments

Join in and write your own page! It's easy to do. How? Simply click here to return to Math Questions & Comments - 01.



Copyright © 2008-2015. All rights reserved. Solving-Math-Problems.com