Tag Archives: library science

A Sheep in Sheep’s Clothing

28 May

[Alternate Title: The Sheep that Wags the Wolf’s Tail]

Photo Credit: petsadviser.com

Photo Credit: petsadviser.com

A few years back, I made a deal with a friend to run the Chicago Marathon. I’d run several marathons previously, but none in a number of years. I was nowhere near the shape I wanted to be in when I made the deal and as both my friend and I were trying to be healthy, it was a win-win situation. At least it seemed that way. I trained in the winter, indoors on the treadmill. Once spring came, I moved outside. Week after week, I slogged through, never feeling like I was gaining any stamina, losing any weight, or getting any healthier. As summer came along and I started to stretch out my miles, I found myself getting pretty sick after running. I couldn’t eat anything without having serious digestive troubles. It got to the point that after running I’d stick with a smoothie and not much else. Even that didn’t always go well. I’d never had this experience in past training and kept chalking it up to being out of shape.

Then, one Friday evening I went for a 12 mile run – struggling through it as my gut rebelled against me. I finally finished, stopped by the 7-11 for a Gatorade, and drove home. After showering, I had a smoothie and settled in to watch the Red Sox. Next thing I knew, I was on the floor of my bathroom and next after that in an ambulance to the emergency room. I’d never been in an ambulance before – and that’s about the only positive I can think of regarding the experience.

After a night in the ER, tests that revealed nothing much, trips to my doc and a couple of specialists, the vascular surgeon told me that he suspected I had celiac artery compression syndrome (or median arcuate ligament syndrome). He also said, as I described my symptoms related to running with him, that he’d never heard of it being associated with exercise.

Being the medical librarian that I am, I set about searching PubMed (now that I knew some terms to search) to learn about what was going on inside of me. Mostly, I was looking for something that would link my training with this syndrome. Lo and behold, I found one. One. That was it. One case study about one individual – an elite runner who’d suffered something similar to what I was experiencing. Granted, I was hardly “elite” in my running, but the symptoms and situations described for this runner were just what kept happening to me.

I promptly sent a copy of the article to my surgeon and then, a couple of weeks later when I was wheeled into the operating room for an arteriogram to confirm his diagnosis, he said to everyone in the room, “If you have any questions, ask Sally. She’s read more about this than you have.” (Though fortunately not more than my surgeon!)

This is a long, round-about story to demonstrate a point – when it comes to evidence, a case study that resonates with you, the individual, is worth as much as any randomized control trial.

Evidence-based practice is THE term in medicine today. As noted by Stewart Donaldson, Christina Christie, and Melvin Mark in the introduction to their book, “What Counts as Credible Evidence in Applied Research and Evaluation Practice?” we live in an evidence-based society. Evidence-based medicine, evidence-based mental health, evidence-based management, evidence-based decision making, evidence-based education, evidence-based coaching, evidence-based policy, evidence-based sex education, evidence-based fill-in-any-blank are just some of the examples they list from a quick Google search of “evidence-based practice”. For those of us who have taught any course related to evidence-based practice, we know all about the EBM Pyramid and the hierarchy of quality when it comes to evidence. At the top of the pyramid sits the randomized control trial and systematic reviews. Further down, the anecdotal case study. In other words, something happening to one or two people – like me and that other runner – simply doesn’t qualify as enough evidence to state that there is any connection between exercise and celiac artery compression syndrome.

Except when it is enough. As it is/was for the two of us (and no doubt a few others).

I found myself thinking often of my personal case history and the evidence-based pyramid during a number of sessions that I attended during last week’s annual meeting of the Medical Library Association. Why? Well, mostly because I attended a lot of talks on the new roles that librarians and other professionals working within libraries, i.e. PhDs in bioinformatics, are assuming today. People are doing an awful lot of interesting things related to specialized services. I count myself in that lot. I may well be an evaluator now, but I personally think it’s simply an extension of the specialized work that I was doing in the library. But the thing that I kept noticing – and a point I raised in one of the sessions – was when, if ever, will we get past case studies related to these services? When, if ever, will we be able to say as a profession that the successful new roles and services that some libraries are offering today are roles and services that can be adopted broadly? When, if ever, will we have enough evidence that demonstrates the success is based more on the service and the role, and less upon the individual delivering it?

Watching Twitter throughout the meeting, I noticed one person tweet a picture of a slide from Bart Ragon’s (University of Virginia) presentation, “Where is My Data Scientist?” (Disclaimer: I was in a different session at the time, thus am taking Bart’s slide out of context.) The slide read, “Unless you are Kristi Holmes or Michele Tennant – Most librarians lack any of these skills.” For those less familiar with MLA, for many years, Kristi (formerly at Washington University, St. Louis, now the Library Director of Galter Health Science Library, Northwestern University) and Michele (University of Florida) were known as the two PhD biochemistry people in our midst. They were anomalies; scientists working in medical libraries. Today we have more – Jackie Wirz at Oregon Health Sciences University, Meng Li and Yibu Chen at the University of Southern California, and Tobin Magle at the University of Colorado Denver to name a few – but they still remain oddities.

When I asked the panel of Kristi, Tobin, Jackie, Meng, and Jerry Perry (former director of UC-Denver’s medical library, soon to be the same at the University of Arizona’s health sciences library) this question about case studies versus a broader body of evidence, Jackie admitted that sometimes she does wonder if people call on her because they think “Jackie can help” or if they think the bigger, “the library can help.” I don’t mean to suggest that the two are mutually exclusive, but how much are our trends towards specialized services redefining the health sciences/medical library profession, as a whole, and how much are they simply taking advantage of particular individuals and the strengths, expertise, etc. they bring to a particular library?

One of the bits of advice that Kristi Holmes offered during this session was, “Build the best library for your institution, thus what works here may not work there.” It’s hard to argue that this isn’t good advice, yet at the same time I can’t help but wonder about how well it sums up our future as a profession. It’s becoming more and more specialized, more and more individualized, and the parts aren’t easily interchangeable anymore. This can be either good or bad for us. Thinking to what Mae Jemison said in her McGovern Lecture (I wrote about this for the NAHSL blog and will share the link when it’s published), innovation is not inherently good or bad; progressive or regressive. It’s the choices that we make around our innovations, our new ideas, our new roles, and our new services that ultimately make the difference.

I’m a librarian with the title of “Evaluator” sitting in an office in the UMass Center for Clinical and Translational Science. Jackie Wirz is a PhD biochemist and molecular biologist with the title “Biomedical Research Specialist” and an office in the OHSU library. What do we have in common? What do we do that can be teased out, taught to others, and adopted by other libraries, centers, or institutions? To me, that’s a BIG question and something worth continuing to try and answer/address as our profession continues to redefine itself for the future (and now).

Tags: , , , , ,

So DO You Want to be a Data Scientist?

28 Mar

Last week, a colleague that I follow on Twitter retweeted a post from the blog, NatureJobs, titled , So you want to be a data scientist by Michael Koploy of SoftwareAdvice.com. The colleague who originally brought the piece to my attention, Kristi Holmes, PhD, is a bioinformaticist at Becker Medical Library at Washington University in St. Louis School of Medicine. She’s also an all-around good egg and one of my absolute favorite colleagues in the field, but that’s beside the point. I would have read the piece regardless of who tweeted it to my attention. However, because it came from Kristi, we then engaged in a mini tweetchat that we’ve had before, i.e. Where and what is the intersection between data scientists and librarians, if there even is one?

One of the interesting things about this discussion, to me, is that Kristi is a scientist who happens to work in a library, while I am a librarian, trying to work in the arena of scientists. And from our different perspectives, she is the one who is routinely much more optimistic about librarians getting into the area of data than I. There’s probably a thing or two you can decipher from this, but that’s for another time.

Another thing that happened after I retweeted and commented on the post was that I got an email from Brittany Richards at Software Advice thanking me for the tweet and additionally, asking if I’d do a blog post of their article here on the Librarian Hats blog. Specifically, Brittany wrote, “You mentioned library science and I was interested to see your thoughts on how the two are related to each other.”

Now if you’ve read this blog for any time, you know my answer was an enthusiastic, “SURE!” So here goes – a recap of that article and some summarizing of conversations I’ve had with Kristi and other scientists on the topic:

scienceI once saw/heard a librarian give a presentation where he identified himself as a data scientist. I called him on it. I am a librarian with a graduate degree in library & information science. I also have a graduate degree in an applied biological science (exercise physiology). Given that background, I feel pretty comfortable stating that while the two share the word, there is a world of difference between the science that librarians do and that which takes place in laboratories, clinics, the field, etc. As I’ve stated in this blog before, my background in exercise physiology is what I feel gives me the extra tools that I need to be effective as an informationist. That’s the science background that is recognized in the sciences.

I hope you don’t hear me dissing my library degree, education, or career. I’m not at all. They are just different and when I read articles like Koploy’s, as well as many books on data, specifically library and librarians’ roles in working with data, I cannot help but keep this thought in mind. It’s what comes to my mind. Every time.

In his post, Koploy recalls the description of a data scientist that he got from Bruno Aziza, a big name in Big Data. Aziza called a data scientist a “business analyst-plus.” He highlights mathematics, statistics, and business strategy as their core skills. Koploy himself adds, “While programming and statistical expertise is the foundation for any data scientist, a strong background in business and strategy can help jettison a younger scientist’s career to the next level.” Further, he notes that successful data scientists are drawn from the fields of biostatistics, econometrics, engineering, computer science, and the like. I’ve read the article several times. Library or information science is not on the list.

Again, this isn’t a slight against my field, but rather an observation that there are different skill sets required for different jobs and the job of a data scientist is not the job of a librarian. And vice versa.

So the question then becomes, how much does a librarian – or an informationist – need to learn to become a data scientist? I say, “A lot.” However, that “a lot” comes with the assumption that one isn’t entering data science from one of those previously mentioned fields. If this is the case, then of course, that individual is well prepared. You’ll note though, that even with the background, Koploy points out that data science is (1) fast-growing, (2) extremely competitive, and (3) new. Even the most seasoned statistician needs to learn some new skills and/or subjects to keep up.

The optimistic among us – those who believe the cross-over between information and data science is broad – focus upon those characteristics that are, in fact, mentioned by experts in the data science field as ones that separate the exceptional data scientist from the average; inquisitiveness, the ability to spot trends, and the tendency (skill) to ask the right questions. It’s the latter where librarians, informationists, and information scientists both have experience and often excel. We know how to ask the right questions that get to the heart of information problems, e.g. How does the business work? How does it collect data? How will it use the data? (per Krishna Gopinathan, Global Analytics Holdings)

So, do you want to be a data scientist? If you’re a librarian or an informationist, depending upon your background, you may or may not have a little or a lot of work to do to get ready to take on the role. If you don’t have the background, I see two possibilities:

  • Get it (hit the books!)
  • Find the right partner(s) where your skills can be paired to produce a good data science team

We choose careers for a lot of different reasons, but I like to believe that in the best case scenario, we choose something that we’re both interested in and good at. Remember those aptitude tests you took in the guidance counselor’s office in high school? They were (and still are) meant to measure something. They measure what we like and what we have an aptitude for. They measure what career would fit us best. It means something to be a librarian. It also means something to be a scientist. I believe that it’a a sign of the times, and a bit of a challenging time at that, that careers and skills and tasks that once sat neatly within cubicles and labs and computer workstations are now all mixed up together. This melting pot of vocations is difficult to navigate. On the one hand, it opens a wealth of new opportunities. On the other, though, it means for everyone working with information and/or data, we will never enjoy sitting back and doing the same old same old for very long.

If you’re interested, I also encourage you to read the original piece that Michael Koploy wrote, along with some of the links he suggests for further reading. In particular, I really enjoyed Hilary Mason’s blog. Good stuff there. I also happened to notice, just this morning, that Coursera’s free Introduction to Data Science class that’s listed is starting up in the not too distant future. If it piques your interest, give it a go. You might well find that you have a hidden talent that will take you far in this new area.

Which brings me full-circle to the question I began with, i.e. Is this new area in the library? Well, quite obviously there are individuals like Kristi, bioinformaticists and data scientists who find their home in libraries*. There are also librarians or informationists with training in data science who find their homes outside of the library. And then there are librarians. And then there are data scientists. In other words, there’s a big mix of us. If you’re comfortable in the mix and you’re up to the task of getting and/or honing new skills, you’ll likely do really well wherever you are.

The times they are a changin’, sings Mr. Dylan, and we look to change with them. At the same time, though, we need to be realistic. We need to see clearly what we know, what we do well, what we like, and more. We need changes in graduate education across the board to address these issues, and likewise those of us working need to accept that we’ll be learning for a lifetime. These are the times we live in. You can’t just call yourself something different. You need to do something different. Or do things differently. Likely all of the above.

special agents rockin

Rockin’ out with my pals, The Special Agents, at Houghton Elementary School. Support art, music, and physical education in your public schools, people! You could get a band out of it.

Now I’m off to play drums with a friend’s band, dressed up like the Cat in the Hat. You’ve got to have a really big tool box o’ skills, friends. Really big!

* And then there’s the matter of money. If you have the chops to get a job as a data scientist, are you willing to work for about half of what you could make in business or industry than you will in a library? It’s a question that comes up in our professional discussions often. If you want to have at it in the comments section to this post, go for it!

Tags: , , , , , , ,