The vis is about the various purported health benefits of the health supplements on offer, their popularity and their effectiveness at actually helping. Hopefully this will be the first of many of these kinds of projects for me, it’s something I really enjoy.
UPDATE: We just released a new version with an all new data set, proper live data loading from Google Docs and some visual tweaks.
Check it out here.
I am interested in your “Snake Oil?” chart. I personally have found some of the things below the “Worth It” line that definitely work for me. I know that not all the herbal suppliments work well, depending on the guaranteed potency, and freshness of the product. For example, Saw Palmetto work for me. However, Echinacea, I just don’t know.
I understand that your stats are from scientific information, but really, how many studies have been done on these. As far as I know, not many have went through that much testing because companies or people don’t want to spend the money. I take the herbs, whether “science” has proved them or not. Most of the ones I take work. If you want a list, I will send them along. Thanks!
[AP - Full information on the nature, results and scope of the studies can be found in the original article linked to above. I am pleased to hear that some of the supplements found to be of little use in large, human, randomized placebo-controlled trials *do* work for you, however I am sure you understand that creating an infographic based on personal anecdotes rather than hard scientific data would be both near impossible, and mostly useless. ]
1.Theinteractive chart is cuotoff on the right side on the Info site
2.Reishi is missing from your data spreadsheet
An amazing chart to be sure from what I can tell! Imagine that I would be reffered from a pc gaming clan to a major interest of health supplements being here!
Is the bubble race chart code open licence? I would like to play around with this type of charting tool with different data sets, do you have any suggestions?
Hi Andy — Love this Flash visualization. My question is same as person above. Just checking if you’re going to share the file/code you used to transform the info from the Google Doc. We’d love to experiment with this!
Hi, I’m afraid the code will not be released as an open source library… If you have any questions i’ll be happy to try to answer them though.
My first impression upon looking at this was “cluttered mess”
The longer I look at it, the more issues I find with it.
1) What are the units for the Y-axis? How does one quantify strength of evidence? And what’s the basis for the “worth it” line?
2) What is the meaning of the colors? Other than the “one to watch”, the color seems to be redundant, simply displaying the same information as the Y-axis position represents.
3) Why not make use the X-axis? As far as I can tell, position along the X-axis is arbitrary. If you had plotted popularity on the X-axis and strength-of-evidence on the Y-axis, then you might expose some trends in the data – how are popularity and evidence correlated?
4) While its a nifty visual trick, having the circles expand as you mouse over them distorts the data.
5) Why not show the # of google hits? You indicate this with the size of the circle, but the key does not show the actual numbers. Is the number of hits proportional to the area or the radius of the circle? Since you don’t show this in the key, we are left to guess.
And, this use of circles is questionable. “Area and volume representations fool people with the square/cube law: an increase in linear size leads to a square of the increase for areas and a cube of the increase for volumes.” – (Edward Tufte)
Those are just a few of my issues with the graphical presentation – beyond this there are clearly some issues with careless gathering of the data, cherry-picking of studies (the St. John’s Wort study was from Germany, and according to Wikipedia, “results from German-language countries are considerably more favourable for hypericum than trials from other countries.”), and apparent misunderstanding of medical terms (for example, mousing over “Licorice Root” says “coughs”, while the study linked is for “A plant extract and its modified preparation in functional dyspepsia”)
It just goes to show, you can put a pretty Flash interface in front of something and people will be impressed with the gee-whiz aspect of the presentation, despite the fact that the underlying data is being badly mangled. The purpose of a plot is to present information clearly and cleanly, not to be a game. Information is beautiful when it is presented clearly.
You did an amazing job with this visualization. I’m sure you’re aware of the wide audience it’s reached. This information would never have been passed on to so many folks without such a stunning visual display. It is so difficult to convey medical information in a visually elegant way and you have hit a home run on this. While I don’t necessarily agree with all the findings, I applaud your work.
Andyperkins.. Super
Hi Charles Waldman – why so bitter?
I agree with Charles. This looked really easy. The challenge would have been using the y-axis as well. As it, it seems messy and needlessly confusing.