From the marketing department – or: blogging by science journalists vs scientists who are blogging?

A bit of context first. I received an “in need of an update” email from science blogger Rebecca Sato about one of my webpages I’ve written some 7 years ago and which summarises informally the benefits of red wine. In short, I had listed three negative aspects of red wine (alcohol, tannin, sulfites) and an amazing 7 positive aspects, being: reducing coronary heart diseases, maintaining the immune systems, polyphenols, Resveratrol (trans-3,5,4′-trihydroxystilbene), flavonoids, anti-bacterial activity, and anti-stress. This was based on 4 references (due to ‘article access secured for subscribers only’), a class at university (WAU, it probably was the food chemistry or fermentations course), and, taking into account health guidelines of the different countries, I had written at the end “The net effect of drinking at least 2 to 3 glasses of wine a day is supposed to result in a healthier life.

According to Ms Sato, I really have to update the page, because

you’re going to put uneducated readers at more than 25% elevated risk for cancer with your 2-3 glasses a day advice. International health organizations like the World Health Organization agree based on numerous studies that alcohol can elevate the risk of a range of cancers with as little as one glass a day. And by the way concord grape juice has been proven to have an even better effect on the heart that red wine, and the effects are longer lasting than alcohol, … to leave outdated and dangerous advice on your site could end up hurting people that don’t know it’s dangerous and outdated advice.

Ah. First, I did dutifully list alcohol in the “negative aspects” section to indicate that I’m not from the alcohol-promotion-board, so the reader was warned. Second, there were neither references in the email to back up the 25% elevated cancer risk claim—what about the balancing act that ingredients in wine actually reduce cancer by x %? And what if x>25?—nor the “as little as one glass a day”. Different countries’ health board gave different guidelines (from average 4-5 in Spain to 1-2 in the Netherlands), also based on research, presumably, so I took the middle ground. Third, some fun with concord grape (a cultivar derived from the Vitis labrusca) juice. Blasting in font size 29 in the email, “New study shows Concord grape juice has a heart-healthy effect not yet reported with red wine”, with some additional text reporting on the recently held WineHealth 2007 conference where this new scientific fact has been presented, presumably. The research presented there has not been published, but supposedly will be by December 2007, in a 1-page abstract (see last page of the conference leaflet). It was organised by Vin & Sante, where they have as aim to “to prove or disprove the phenomenon called the French Paradox and to help the scientific understanding of this phenomenon.” They are still researching.

It gets even better. A quick search revealed that the main text in the email was a press release, posted on physorg, with source attribution Hunter Public Relations. Hunter PR has its mission on their front page:

“marketing public relations firm that develops and executes programs that help brands achieve their sales and marketing objectives. Utilizing a creative and strategic combination of public relations, publicity and influencer outreach approaches…. helps [companies] become relevant in the hearts and minds of target consumers. Utilizing the power of the news and entertainment outlets, events, sponsorships and atmospheres, ….”.

So I “need to” update my wine page on the basis of non-published research that is being promoted by a PR firm??? I don’t think so.

I looked into the topic some 7 years ago out of curiosity and because I wanted to have something else to do than the daily grind of systems engineering. I know there are more references I can add to the webpage. But now I have access to most scientific journals, so, when I make an update on the pros and cons, it will be more scientific results (and, possibly, with my limit to six languages, to add a small table with the Ministries of Health guidelines one wine consumption). If you have any scientific publications on the French paradox, feel free to add them to the comments or email me.

A friend suggested I should try to get some corporate funding for my tiny little page on the outskirts of the Internet – after all, I’m only on a PhD stipend at the moment. I wonder how much those “science bloggers” earn to spread misinformation. Spreading misinformation, be it under the banner of ‘science reports’ or other, goes even quicker with blogs and instant news feeds and all that. Newspaper headlines generally exaggerate research results with the ‘new vaccine against AIDS’ kind of stuff to make it look sexy and sell copies, but at least there’s a reasonable-length article to go with it that has at least some of the usual nuances, explains the whole thing in laymen’s terms, and has the original literature reference(s). The Newsinternet headlines and sites are a simplification of the printed press (mostly); science journalism blogs definitely are (I know, not all of you science journalists. When I’m not working on my thesis, I’m thinking of an algorithm to mine the blogs to separate the wheat from the chaff and play with SPSS to back up this statement). In casu, googleing Ms Sato, the top result is a blog post “Immaculate conception?” from 16-4-‘07 twisting Prof Nayernia’s research focus from stem cells w.r.t. curing diseases to biological kids for lesbian couples, supposedly based on research published in “Reproduction: Gamete Biology”. It’s even the wrong journal title; it’s Biology of Reproduction and `gamete biology’ is a subsection of the journal. Nayernia’s article from Dec 2003 is entitled “Male Mice Lacking Three Germ Cell Expressed Genes Are Fertile” (open access). Or you would like a post on time travel and a follow-up with criticism by physicists? She does not seem to have her own blog but posts at others’. Need I say more?

Of course, I’ve read about bogus ‘think thanks’, fake tobacco and sugar research institutes, and the need to declare competing interests at the end of a scientific publication. But to encounter the nonsense first-hand was a new experience. I have to admit it was a not a bad try to bend tings, but it failed; instead of continuing with my research, I ended up wasting time tracing and looking up the info and writing this. What do such people hold me for that they think I would just swallow it?!


p.s.: no competing interests. I’m still on the university-paid PhD stipend; and this is not to be understood as a hint. The bad-publicity-is-good-publicity is bad enough as it is.

p.p.s.: yes, I very much enjoyed my wine yesterday evening.

p.p.s.s.: I emailed Ms Sato I’ve honoured her initiative to email me with a blog post. Now let’s see if I end up in purgatory.

7 responses to “From the marketing department – or: blogging by science journalists vs scientists who are blogging?

  1. Ah yes, the troublesome end of the long tail 🙂 – should come up with a name for that. Couple things this made me think of.
    1) real science journalists do exist and are generally much better at distilling the basic ideas of research in an accessible form than practicing scientists.
    2) I’m curious about your blog rating algorithm and how it relates to DIGG/Technorati/etc.. It would be excellent if you could do better than a simple voting system. I’ve thought for a long time about building a ‘slant detector’ for news articles – something that would first identify articles on the same topic, then identify bias in the article, then output a non-biased article based on the intersection of the inputs. Fox Nes + BBC + AL Jazeera => truth ?? Perhaps something like this could be added to your blog analyzer.

    mmm Resveratrol …

  2. most of the post is deleted on request by the author. I [keet] leave two bits here, as Ben is referring to the second bit, and I will on the first as well

    But I have to know, were you serious about thinking having your “own blog” is a sign of prestige?

    What he chose not to share was that there is nonbiased research supporting that his advice to drink 2-3 glasses of wine a day is garbage:

    A very recent European analysis of nearly half a million subjects came to a the conclusion that even one glass a day is a significant health risk. I didn’t give this blogger a long list of references because I assumed that he really was a credible expert on the subject then he already knew of all the many supporting studies done by both potentially biased AND completely unbiased sources. But just for kicks, here’s one listed below for you and all you have to do is google “cancer” and “alcohol” to find dozens of other reputable studies supporting this research.

    http://www.medicalnewstoday.com/articles/78391.php

  3. Darn it! I have such a guilty personality. I absolutely cannot call someone names without immediately feeling awful that I may have hurt them. I’m sure you didn’t value my opinion enough to be hurt by anything I wrote in the first place, but I’m sorry nonetheless. You were understandably offended that I implied you were endangering people, when all you did was write a post about some research several years ago. You were unreasonably snotty in your response, but I am also an unreasonably snotty person who shouldn’t be casting stones.

    So anyway, just delete my rude comment that I posted on your blog and please forgive me. You’re obviously a bright person who enjoys a good argument and in that respect we’d probably be friends if we actually knew each other. I won’t offer you any dubiously “helpful” advice in the future. I forgive you for being mean, as well, in case you also feel a twinge of guilt for needlessly mocking my writing. Take care.

    Rebecca

  4. Hi Rebecca,

    OK.. Soon to be Dr. Keet and everyone else knows that alcohol is not healthy. Seems that some of the other components of red wine are interesting, but thats really about the end of the story for now. Chill out everyone and try to have a conversation…

    To establish more credibility when writing for a scientifically savy audience about biomedical issues, I’d suggest citing links to PubMed or to the original journal rather than news sources. Reading the news is a good start, but, like every game of ‘telephone’ there is often information lost in every translation. The first sentence in the abstract of that paper you cite above makes a much weaker statement than the news article…

    “Alcohol consumption may be associated with risk of colorectal cancer (CRC), but the epidemiological evidence for an association with specific anatomical subsites, types of alcoholic beverages and current vs. lifetime alcohol intake is inconsistent….”
    http://www3.interscience.wiley.com/cgi-bin/abstract/114294429/ABSTRACT

    Hope that you keep talking.. but know one should assume a blog article constitutes real medical advice.. Its just a conversation.

  5. catching up on the events in the blog comments, I’ll respond first to Ben’s two points above.

    You make a good point on that science journalists may be better in transferring the science to the rest of the world in comprehensible writing compared to overfocussed scientists. What I do wonder about is if most of those good science writers did have a masters or PhD education first and happened to have good writing skills and liked doing it, so made the move to science journalism. That is a different starting point from doing a BSc in journalisms and then to choose science journalism as specialisation. This is not to put down the latter group, but the former have had training in both reading & digesting scientific papers and have learned skills of doing research. Some research is needed on that, if there is a difference in type, quality, depth of the articles & posts they write.
    This brings me to the second point of how to automatically detect the good ones from the others. Apparently, each person has his/her own writing style and with some computational linguistics it should be possible to make and use such profiles (things like average length of sentences, use of the word “the”, passive/active voice, recurring sub-sentences, etc). That is, if the CIA and the like can use it to assess if message x is really written by terrorist y and not one of his accomplices, we might as well use the same technology of pattern matching for detecting good science journalism. Problem with this is that it puts lesser-known science journalists at a disadvantage. This brings me to another re-use of existing technology: plagiarism detection software. If (parts of) the journal article borders plagiarism, then one can postulate it stays close to the content of the scientific publication. Also, one can make use of the keywords that authors have to provide to classify their publication: if they do not, or hardly, appear in the journal article (ok, with some WordNet synset flexibility during the string-matching or classification with an ontology, like with GOPubMed), one may wonder if the gist of the article is properly reflected; conversely, if far off keywords appear in the journal article, then the contents of the scientific article may be totally put outside the proper context and thus not really reporting the scientific results.

    Besides these three technologies that may have their place in the algorithm, and focusing on online news/blog articles, it is appropriate to provide the reader with the full reference & link to the article and/or the link to the researcher they are referring to or quoting, to at least make it easier to verify the end of the tail: if there’s nothing to hide, then there’s no problem to add it.

    Regarding the user-voting, I’m a bit hesitant. I can imagine some groundbreaking paper, say, stomach ulcers caused by the H. pylori and then some concerted effort to vote it down. Nature does not operate on a democracy by humans and sometimes the results are not to everybody’s liking, but it may be an excellent science journalism article that transforms the results to an article digestible by the interested laymen.

    This for the first brainstorming; other ideas are more sketchy. And they all have to be tested on validity of course.

  6. Responding to Rebecca, concerning blogs: no I certainly do not consider having one’s own as a “prestige” over posting on other people’s blogs and news outlets. In the context of traceability, it does give a single point of reference; that is, it is part of the provenance issue on the ‘who says what and where does it come from?’. This is easier when a journalist has his/her own blog as compared to having to trace previous posts through finding out with Google searches. Given the ease of getting one’s own blog, it is odd when one only posts at other outlets but does not have a single point of reference that aggregates the contributions together. Given that the technologies I mentioned in my comment to Ben aren’t implemented in a way for Jo and Joanne Soap to use them, this leaves each reader to assess journalists and bloggers manually though the contributions they make over time. For instance, you keep promoting the Hunter PR press release. Now, if you would have had your own site or blog, it would have been easy to check if you’ve been their, or another PR firm’s, mouthpiece before, likewise on the sloppy referencing to the scientific publication (if it is a recurring thing or just a few odd ones out of all the posts you’ve made). Ranting and name-calling does not help building up credibility as a science journalist, but things like traceability, proper referencing etc do. Probably, scientists and other people with a science background are more finicky on such matters.

    Like Benjamin, I did check the medicalnewstoday link and reference as well, which has the link to the abstract of the source article at the bottom of the news item to simplify checking the claim of the news article. In fact, the news article is a reasonable digest of the fully referenced paper, with the main conclusions as well as ‘complicating’ factors that tone down the results (or any exaggerated claims), such as the person’s diet (in casu, folic acid, although it may well be thanks to the whole food item and not only due to this component, but that’s another topic).

    Anyway, the issue is not the alcohol thing, but reliability of data/information and if and where there is a difference between contributions from science journalists blogging and blogging scientists.
    As for bioinformatics research and the (Semantic) Web, I get the impression that topics like provenance and trust have started on their way to the top of the hype cycle. That is, research on how these layers of the “SW layer cake” can be implemented, which will take several more years to trickle down to mainstream applications. And for the meantime, it’s manual assessment and self control. There are, however, already several federations and associations of science journalists and science writers; for a list of links, see the end of the wikipedia entry on science journalism (http://en.wikipedia.org/wiki/Science_journalism).

  7. Pingback: Marketing Research » From the marketing department – or: blogging by science journalists vs scientists who are blogging?

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.