grouperlab

Get, share, and use information well

Tag: dissertation

Superposition

With a flourish and frenzy of activity, the cycle has completed itself and begun once more.  This week, of course, was particularly hectic.  On Tuesday, we in GROUPER were pleased to celebrate Liang’s successful defense of her dissertation (now to finish the writing!), and after a 2-hour lab meeting, I went home and got some pleasure reading in.  Wednesday was a travel day, with challenges of unstable weather leading to one canceled flight, and ground stops due to ramp closures at both the start and finish of the second leg of the trip.  Hours late, but with the air cooled from the rain, I finally finished the trip and got home for a good night’s sleep.

 

“Hold it.  You were at home, then you got stuck during thunderstorms, and finished the trip, and went home?  Did you not get to your destination?  Was your trip canceled?”

 

Even as I write this entry, there is a type of surrealism about this week.  I’m sitting in one of the chairs I have had for over five years, with Amber on my lap, looking at her cat tree and food and a bunch of other items I clearly recognize.  The window still faces east, but the image is different: instead of an empty parking lot across the street, I see a tall tree and an office building.  In other words, the shift has now occurred.  Amber and I have moved to Washington, DC to start my position as a Jefferson Science Fellow.  My first day is next Monday, and I have begun enjoying the exploration of the neighborhoods of McPherson Square and Thomas Circle, and picking up coffee and pastry and fruit at the White House Street Market.

IMG_3656

Amber exploring the new window view

My world is changing significantly, and yet some things remain consistent.  I am still an engineering faculty member, but I am thinking about a completely different set of issues this August compared to last August.  Our lab meeting addressed a very interesting topic based on students’ recent experiences, and one that I will be considering very hard in the coming months.  Academia, government, and industry are considered vastly different places, and representing wildly divergent career paths for those with PhDs.  And yet, we’re taught (and have first hand experience) that a variety of people all craft their actions, decisions, and processing of information based on perceived risks, costs, and outcome benefits.  There’s only one challenge.  The benefits that drive most industry teams (profit—hey, I hear the new Aston Martins are really nice) don’t come into play at all for most academic or government folks, and the primary risk keeping academic folks up at night (someone already published my idea in that journal I love!) don’t seem to bother industry or government people much.  A government employee may complain bitterly about the costs of having to work 5-10 hours of overtime one week; many academics and industry research folks take a 60-70 hour average work week as pretty much standard.

 

You may notice that the website (http://www.grouperlab.org) looks very different than a few months ago; that took us a while to work through.  Elliott, our wonder-undergrad, did a pretty thorough redesign and architecture job, but what was more important was not just the scripting language or tab sequences.  We went back to a very basic question: who comes to our lab’s website?  Three different types of groups (yes, it’s that “use case” thing) want distinct types of information, at varying grain sizes.  Even our own GROUPER alums represent different types of interaction profiles.  We’ve got government employees seeing how our work can inform improvements for federal agencies.  GROUPERs are also rising up the ranks in industry, and may be in a position to hire a new or recent grad (in this sense, GROUPER is definitely a distinct and valued brand).  With the lab’s traditional gender mix (somewhere approaching 2/3 female), it’s not surprising that a number of alumnae list their primary function right now as “Mom”.  I see no reason to hide that, or feel guilty or ashamed to highlight such life pathways.  If everything is a system, and GROUPERs look at information everywhere, can’t those skills be applied to everything from medical information use for family members, to understanding the daily experiences of neurodiversity, to getting a front-row seat to the miracles of how humans develop broad processes of learning and skill development sets in ways that still challenge our most sophisticated machine learning algorithms.

 

Over the next few months, the GROUPER blog will be more active again, but it won’t be focusing as much just on our current lab research.  We’re still researching, of course, but we have other stories to tell, and other forms of impact and effect to consider for how we get, share, and use the information we’ve been gathering and lessons we’ve been learning.  Thanks for visiting us again after our quiet period.  Hey, it gets busy learning how to be in multiple places at once.

 

Eaten up with Curiosity

The motto of all of the mongoose family is “Run and find out,” and Rikki-Tikki was a true mongoose.

–Rudyard Kipling

We find ourselves in the midst of a new academic semester, with the variety of challenges that face us in terms of schedules, task demands, and burdens both voluntarily and involuntarily shouldered.  In one sense, it is as it always has been; but for each individual, it may be the very first time of an experience that defines and influences the remainder of one’s life.  I have been thinking about this with the current configuration of GROUPER, and the need to help students make progress on existing dissertation topics or create new ones.   This is not always an easy task, and though I have gone through this process over 50 times (with over 30 MS students and 14 PhD students advised, plus the students whom I have assisted in various less formal ways) on this side of the desk, there are always elements worth learning and improving.

Maybe it is simply the number of times that an issue presents itself within the period of a few weeks that it becomes more salient, and the gap between what is and what could be becomes more evident.  Let’s assume that it may be no more than that, although a friend of mine was just mentioning today how there can be periods when one becomes much more open to insight and jumps in one’s self-learning.  But there has been something about the question, “What should I do for my research?” that has struck me in very different ways this fall than in the past.  I am asking myself different questions about my own research and career pathways; I am reminded of writings and insights from when I first arrived at Purdue.  And of course, in the senior project design course, there is always the sense of importance to get the students—so used to textbook problem configurations and well-organized linkages between the information given and the equation to use—to start creating for themselves a system definition and sense of their own active participation in defining the problem to solve as a necessary part of being an engineer.    And as an engineer myself, such gaps between what is and what could be are always met in my head with, “What do we do about it?”

And yet, there was something that I couldn’t quite bridge on my own in the conversations with the members of the lab.  Where does one go to get ideas?  How does one start the organization of facts and methods and tasks that gets one from classroom student to nascent researcher?  Over the past month, I began to see that it was not just as simple as a statement in our “1:1 meetings” (as the nearly weekly individual meetings I have with all of members of the lab are known) to go figure out an interesting question.  Interestingly enough, this recognition for me comes from a couple of sources, as I am again reminded that I don’t seem to approach the world in a way that is like most of those around me.  Apparently, there are graduate seminars taking place in departments around the country (not just engineering departments, but bench sciences, literature, philosophy, sociology…) where students are encouraged and instructed to read through a bunch of journal papers or monographs or book chapters and determine which questions still required further study.  (For the record, I took such seminars myself: it’s how I first learned, in 1985 and 1986, of Daniel Kahneman and Amos Tversky’s work on cognitive strategies or the cultural specificity of the fundamental attribution error.  I loved those papers.  I just didn’t define my dissertation that way.)

Benson Snyder, in the 1960s and 1970s, discussed a critical issue affecting higher education, one that has come to be known as “the hidden curriculum”.   (The “hidden curriculum” of the book’s title and premise is the informal sociological and socialization process of how and when to learn, not just what to learn.) This book seems to have had a very significant effect on me—not just because I have read through it multiple times (I still own a copy of the 1973 edition of the book), but because I can now see that much of the curriculum I experienced at MIT was shaped in part by the studies Snyder reports of students there 20 years prior.  As I am teaching undergraduate statistics again after several years away from teaching it (but never far away from using it), I am also freshly sensitized to the processes of how to learn, and not just what.  And this is how I started to recognize some of what I was finding vaguely concerning in the lab.

As an undergrad, one of the most telling philosophies of innovation and excellence I ever heard was one that was directly told to me as to why I had so much latitude in organizing my activities for my work-study job.  “I’ve found that it’s best to give good people resources, and then get out of their way.”  For me, that was an excellent and empowering approach, since I was never at a loss for ideas or novel approaches or unusual ways of thinking (at least ways deemed unusual by teachers or professors).  In fact, I recently came to think about this as something I found exceptionally compelling in a cartoon I saw as an adolescent: Chuck Jones’ animated version of Rikki-Tikki-Tavi, narrated by Orson Welles.   Rikki is perpetually curious, and fiercely protective, and powerful in ways that are belied by his small size and friendly interactions.  But isn’t everyone curious in this way of mongooses?  Isn’t everyone driven to “run and find out”?  Isn’t that part of the essential “inside” of every researcher?

No, says the hidden curriculum.  Students are socialized to learn which questions are the “right” questions, and these questions are “best” defined in an outside-in way.  The existing corpus defines the way the field is configured, and thus how new questions should be approached.  But wait… 60 years ago, we didn’t have plate tectonics or the cognitive revolution—just working from existing papers published in 1953 wouldn’t have gotten you there, and certainly wouldn’t have gotten you accepted within the “standard” configuration.  The same is true with statistical process analysis or scientific project management 100 years ago, or pharmacy or aerodynamics 150 years ago, or electrical and thermodynamic processes 250 years ago.  And yet, my learning and research now derives from all of those innovations.  Someone has to move beyond the standard, outside-in framework, and be ready to do the new work and meet the new challenges (and face the inevitable questions and criticisms that such an approach will engender).

It’s obvious to me now that it takes a lot more than a brief instruction to a graduate student to think in terms of the problems in the world of task environments, and interacting with people who live in those task environments.  (Although an introvert, I find it natural and obvious to talk to someone about the challenges of their work.  It’s easier for me than making other types of small talk.)  I begin to wonder, though—have I been assuming that, just by osmosis or creating a supportive environment, anyone and everyone will be “eaten up from nose to tail with curiosity,” as Kipling put it?  Might they need more help than that?

If you were expecting an answer to these questions… sadly, you will have to wait with me for that.  I’ve asked the lab to help me understand what I’m doing that’s different, and how the hidden curriculum has affected and shaped them up to this point (although I didn’t ask it of them quite that way).  But at the very least, asking the question is an important part of the process, and an essential element of making progress.  There are cobras threatening the bungalow of higher education… bringing in and raising a mongoose is not a bad idea.

Updating Documentation

Now that there are a few new members of the lab, it’s time to pay attention once more to making explicit some of our expectations and shared experiences.  It’s interesting to watch, and to test, how stories or catch phrases easily become part of a local culture… only to be met by blank stares when a new person experiences it.

“What’s the best dissertation?”  “The one *you* can do in a reasonable amount of time.”

“Is that a title of a song on the album?”

” Delta Pain” (which is a title of a song on the album)

All of these represent elements of tacit knowledge, in that they are shared and understood by people who were in the lab when the event occurred, or maybe in an individual meeting with me, and have learned to experience and internalize the informal lessons of the lab in a particular way.  That’s great for an individual mentoring interaction, but not really good for organizing the productivity of getting a population of students to finish high quality thesis and dissertation documents.  Thus, we have to do some of these things with more explicit intent, and a more focused and determined documentation of elements of the lab’s culture.

Since the beginning of the Spring 2013 semester, we’ve been working on this in the creation and updating of four distinct documents:  A Master’s Thesis outline; a Dissertation prelim outline (oh, even that’s tacit, or at least implicit: the proposal document written in order to describe one’s dissertation so that one can be advanced to candidacy); a Doctoral Dissertation outline… and most recently, a semester-by-semester timeline for progress towards degree completion.  These seem to be very helpful for students, and help to summarize and integrate and transmit my experience in a fairly efficient way.  And why not?  I’ve supervised over 30 MS theses and 12 PhD dissertations, and sat on committees for another 25 or more graduate documents.  Most students, on the other hand, only do this once.  (I did have one of my students complete a second MS with me after finishing a first one elsewhere; no GROUPERs have ever tried to do multiple PhD dissertations.)  Rather than make everything trial and error, or suggest that there is no pattern that leads to increased probability of success, some people (among those are many engineers) would like to have a sense of the path, the rule, the “game plan” of how this graduate experience is supposed to play out.

Does this mean that there is a fixed and rigid procedure that everyone must follow?  Of course not, for several reasons.  One is that GROUPERs are different people, with different skills.  They don’t want to work in the same stream, or using the same data collection or analysis tools, or ask the same sorts of questions.  Fortunately, I’ve worked in a bunch of areas, so my tolerance for procedure variability is fairly high–I can advise a variety of dissertations, because I have done a variety of projects in different areas and methodologies.  (Some people may think of this concept as akin to Ashby’s discussion of “Requisite Variety”; no strict quantification here is implied.)  So, the level of consistency is not at the detailed level of “you must design a three-level, two-dimensional Analysis of Variance studying the influence of…” Everyone is expected to be able to answer, “Why would anyone want to read this thesis / dissertation?”  or “Why do we care about the question, or the work you did to answer it?”

It may also be relevant that one of the recent dissertations now making its way to conclusion is specifically addressing the question of procedure reliability and complexity.  A very often-repeated task, with few new or challenging elements, can have a standard procedure that is rarely, if ever, inappropriate for completing the task as designed.  If you’re developing a brand new task that has never been tried before, it’s highly unlikely that you can write a perfect procedure on exactly how to do it.  Most procedures are somewhere in between, even if we assume the procedure is always right.  At what level should we expect a new procedure to capture all of the experience that we gain in the development of a new system?

As time goes on, all of these documents will need to be updated–not necessarily because we were wrong, but because our knowledge evolves.  (OK, I was explicitly wrong on this item. I forgot to include a version of the well-known advice: “Tell them what you’re going to tell them.  Then tell them.  Then tell them what you told them.”  In other words, the last section of the introduction chapter should include an outline of the organization of the remainder of the thesis / dissertation.)  Even the working of the updating process is a helpful way of sharing the experiences and telling the stories of the lab.  And when we’re done, current and future generations of GROUPERs can know that I won’t get upset if they haven’t taken 15 credits or completed their plan of study or research proposal by the end of their first semester in the program.  Really.

 

I know, you’re wanting the links to these documents.  They’re still under construction.  Check back when they’re done and posted.

Stream Management

“Hire the best people, and then get out of their way.”  — Arnold O. Beckman

 

While I was an undergraduate, I frequently heard variations of this management philosophy; as an independent-minded student, often working by myself on projects, the mindset was very appealing.  However, as a major professor and lab director, things are not always that easy. 

 

Recent updates of the lab webpage have focused on further elaborations of the range of projects (“streams”) that GROUPERs are involved and productive.  Instead of just listing the two primary research application areas (healthcare and spaceflight), we’ve now elaborated all three application areas (including STEM education), as well as the three more general methodological / theoretical considerations (communication effectiveness when resources are constrained; the effects of delays and information asynchrony on performance; and processes of knowledge sharing in teams and communities).  Why all of the additions?  Not just to sound more impressive and recruit more students (with seven grad students, two undergrads, and a 50% administrative appointment, it’s not clear that a much bigger lab is needed or wanted).  With the new population of students (see the “Restocking GROUPER” entry), it is almost inevitable that the interests of this collection in 2012 would “load” on the streams differently than the lab circa 2004 or 2008.  They even change their allegiance to a particular stream as their understanding of their topic (and their understanding of how the streams are organized) evolves through their development as students and scholars.

 

So, how does one remain sensitive and aware of these issues?  One problem, of course, is that you don’t know what someone doesn’t know, or that you didn’t agree on basic elements, until implicit expectations aren’t met, or unquestioned assumptions aren’t grasped.  This seems to show up most frequently, and with the greatest sense of immediacy, when students are in the midst of a critical written milestone (a thesis proposal or dissertation preliminary document, or a thesis or dissertation draft in preparation for a final defense).  And with that, I again recognize the challenges of managing a diverse lab that works on novel and bleeding edge problems or approaches.  There really isn’t a good way to give someone a past completed thesis and say, “Read this, and do exactly the same sort of thing with the same format, and you’ll be fine.”

 

In some cases, even that might not be sufficient.  After 22 years, a dozen dissertations chaired (and another 20-30 as a committee member), and over 30 theses chaired, it’s easy for me to have the different pictures in my head.  Research methods questions?  I just go back to my psychology background.  Systems engineering definitions?  Yes, the style of that is covered, all the way back to my sophomore Unified Engineering experience.  But what if your writing style was formed and honed by those with backgrounds in mathematical optimization, or physics, or political science, or software simulation?  I’ve experienced all of these from GROUPER students in the past two years… and they’re not the same.  Good luck with putting them all into the same box, or worse yet, into a blender and hoping that something palatable comes out.  (Suddenly, I cannot help but think about the old “Bass-O-Matic” Saturday Night Live skit.)  Yes, it would be much more straightforward to have a single stream and project focus, with all of the students having very similar backgrounds.  However, I confess that I wouldn’t enjoy it as much. 

 

There is a balance between being too far away, and being too close.  Each person, and each group, needs its own balance.  So, I’m looking for dynamically stable equilibria in academic research management and educational personal / professional development.  Last academic year, I spent a fair amount of my time working with two students both trying to finish their dissertations at the same time.  This year, the challenge has been a dissertation, a thesis, and dissertation prelim document, in addition to other research projects.  I keep insisting on being directly involved in each project, and each document, and each student—to the extent that I felt guilty and nervous when I mentioned during a lab meeting that I would be gone for a few weeks and unable to keep to the weekly schedules.  With this level of commitment, I can be both overwhelmed and concerned when I am not there every moment—exactly the opposite of Beckman’s admonitions.  And yet, when I expressed worry that the GROUPERs would be left vulnerable and exposed in the “dangerous jungle and forest” of graduate student life, the response was one of the most rewarding and valuable validations I’ve received in a number of years:

 

“We’re not in the forest.  We’re in the Lab.”

 

That’s also a philosophy I can live with.