Dropped, just like that

Nov 28, 2006

Frank W. sent in a timely reminder of the start-at-zero rule.  This ad from Quaker Oats pitches the impossible: the smart consumer will never believe that cholesterol levels can be dropped just like that!  According to Frank's measurement, the column heights plunged 77% from Week 1 to Week 4 in this chart.

In fact, if the vertical axis had started from 0, then the drop would more appropriately appear to be 5%.  Now, even that would have been a miracle, in my opinion.

Thus, I would like to know what is a "point" of cholesterol, and what do they mean by "representative" drop.  I suppose they are asking me to call that number.

My previous posts (with commentary from readers) about starting at zero can be found here and here.

You can follow this conversation by subscribing to the comment feed for this post.

When trying to visualise grow of just one type data, it is often best to rebase the chart to 100 at the first datapoint. This way, one can easily emphasize growth or decline.

See this post as an example:

The one thing you loose in this example is a feeling for absolute scales. Jens

Thanks for keeping up the good fight! I wanted to say I used one of your examples in a journal-editors training session for What Manuscript's Graphs Should Not Look Like. It was very effective.

Best, JF

As a physician, I have had the concerns about these ads for 2 additional reasons: 1) is that drop statistically significant or merely a "trend". No mention is made of this; and 2), even if it is, is it clinically important (does dropping from ~208 to ~199 have any defined health benefits).

A bar chart is obviously a wrong choice.
As long as you observe JF scientist's principles, I think it's fine to leave out the zero in graphs like that.
To quote Edward Tufte:

In general, in a time-series, use a baseline that shows the data not the zero point. If the zero point reasonably occurs in plotting the data, fine. But don't spend a lot of empty vertical space trying to reach down to the zero point at the cost of hiding what is going on in the data line itself. (The book, How to Lie With Statistics, is wrong on this point.)

http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=00003q&topic_id=1

Quaker Agrees to Tone Down Exaggerated Health Claims on Oatmeal
CSPI Drops Plans to Sue
WASHINGTON—The Quaker Oats Company has agreed to drop certain claims on labels and in advertising that the Center for Science in the Public Interest (CSPI) says exaggerated the health benefits of eating oatmeal. Quaker will no longer describe its oatmeal as a "unique" whole grain food that "actively finds" cholesterol and "removes it from the body" and will no longer display a graph that greatly exaggerated the cholesterol-lowering potential of oatmeal. In turn, CSPI will not file a lawsuit that it warned Quaker company about in October.

The comments to this entry are closed.