Infographics

Good infographics present data, visualize relationships and make clear surprising facts with style and flair using the medium of the web browser to view images tens of thousands of pixels long. Here is a whole blog devoted to the technique with great examples. I was recently asked my opinion of this newly created infographic on bad practices in science. Its creator is interested in feedback. My response is under the fold, but I encourage you to look at the graphic first, and, if interested, leave a comment or email me directly for his mail.


In general, I enjoy infographics; but I think this one is rotten. Not because of the message, but because of how it is presented. Skepticism of the world is healthy; but this can be taken to such extremes that it becomes ridiculous. And how this graphic conflates science in general with only a few disciplines is unjustified. The majority of “facts” (all of which appear taken out of context, or at least incomplete) presented in this graphic come from only a few disciplines. So why lump it all together into “science”? Of course, bad science is bad, and there are certainly incredibly bad and harmful examples (for instance, Penkowa and Wakefield), but is bad science really “rampant”?

The whole intro of the infographic is “cooked”. Is the number of researchers who falsify data really “enourmous”? It sets up scientists as “the most honest people around” (who says that?). I sure don’t believe that. That is why I submit my work to and participate in peer review and reproducibility (or try to as much as possible). Of course peer review doesn’t always work; but don’t we have cellular telephones that work? Computers that run? Satellites that orbit? Medical procedures and medicines that work? We are surrounded by good science, most of which goes unreported, and I see this infographic as encouraging ignorance and science denialism because of a few out-of-context results and false equivalences.

In a specific instance, this infographic claims “making up data”, “distorting data”, and “mining data” are examples of scientific misconduct. When designing new communication technologies, engineers typically “make up data” in order to run simulations. Text-to-speech is based on making up data. Data are inherently noisy, and we must process data (distort it by, e.g., averaging) to get to the information that is not noise. And Amazon.com uses data mining to recommend books, increase its revenue, and make my library richer. From this graphic, I fear that, people being people, people will text search scientific articles to find instances of “data mining” and throw away excellent work just because they have no idea of the particulars of scientific disciplines. We have seen such confusion in the recent climategate non-troversy.

Throughout this infographic, I am searching for specific references. 81% of “biomedical research trainees” said they would fabricate results. Reference please. I want to see if these are undergraduate students. I don’t think they are professors, or those that actually write grants. Besides, “would” doesn’t mean “did”.

“1 in 2 was found to contain a statistical error” sounds awful! But it really depends on the type of error made. Reference please.

As to the conclusions, I agree that all data and programs used to analyze the data should be made available and transparent. There is a current of “reproducible research” in signal processing that emphasizes this aspect. I, and most of my colleagues, make available data and algorithms to reproduce the figures in our articles.

The second conclusion about journalists has nothing to do with science. The vast majority of research never makes it to the newspapers, yet our cellular telephones are brimming with technology that is only a few years old. People have no idea the exotic aspects of digital signal processing and communications that are now being developed, e.g., to support WiFi on high speed trains. Also, “seeking out opposing views” assumes that an opposing view even exists. What is an opposing view to Fourier series? To evolution by natural selection? (Intelligent design and its discussion in the media as an opposing viewpoint is as fabricated a controversy as the “War on Christmas” of Fox News.) Seeking out opposing viewpoints doesn’t make sense at all to me on the majority of topics on which I work.

I think this infographic is entirely ignorant about the broad spectrum of scientific practice and its superlatively successful application over the past 2000 years. It presents an oversimplified and incomplete view of science that it itself is “bad journalism”. The color scheme is nice though.

Advertisements

One thought on “Infographics

  1. It seems to me that there is a division between the medical sciences and the rest of the disciplines. For a variety of reasons, it is harder to keep the sanity checks in place when you are working in this area. I think the infographic you mention is bias toward the medical sciences. Some of the data points it mentions are about psychology, is psychology considered a science :) ?
    Check this book for some analysis of how things goes (very) wrong in the medical sciences:
    http://www.amazon.com/dp/B0035IIDCG/

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s