Thursday, March 30, 2017

On service

I've written about service before in pieces of various posts. Service keeps the University running, and is essential to a functional department. However, the service load is distributed incredibly unevenly, which is a source of angst and frustration. I am the kind of person who likes to do a good job at the things I do, so I do put in quite a bit of effort in my service. This means I get asked for "favors" a lot. At this point, I've spent enough years at ProdigalU that I actually have (and can articulate) a service strategy that works for me.

1. Ignore the service slackers as best as you can
It is a truism that some people blow off service without any penalty, thus carving out more research time for themselves. This is incredibly and maddeningly frustrating to those of us who try to do a good job on service and STILL have the same sorts of research and teaching time pressures. This is exactly the same thing as those who blow off teaching (though we have very few of those in my department, to be honest), and I try my very best to ignore these facts as much as possible in my day to day life.

It is worth talking up your service contributions when merit raises are being considered. It is also worth bringing up service contributions when being asked for yet another "favor" from the Chair/Dean/whoever. Otherwise, I find that getting worked up about the unfairness of the service load is like getting worked up about the unfairness of life--yes, it is true, but getting frustrated and unhappy about it just hurts me and makes me more stressed out.

2. Try to pick up service obligations that you at least find interesting
No one really enjoys committee work, but it is much easier and more enjoyable to do committee work that 1) leads somewhere and 2) is at least somewhat interesting. I have enough years here to know which committees at the department level meet and do nothing vs. those that don't meet and do nothing vs. those that meet and get something done. I try to make all of my service tasks fall into categories 2 and 3.

Currently, my main service task does require a fair amount of effort, but it is vital to the department, and I can see the fruits of my labors pretty easily. I also enjoy the topic, and find it an interesting exercise. I don't really mind doing this sort of task, because it doesn't feel like wasting time. After my sabbatical, I asked to return to this position, because I would rather do service like this than many other possible options.

3. Since most people care about the department, but don't want to do the work themselves, they will follow your lead if you have a strong vision and can articulate it reasonably well
This is related to my point 2. One of the reasons I like my main service task is because it has a real impact on the department. In the course of doing my service tasks, I found that if I had an idea I wanted to implement, it was pretty easy to get people on board if I could explain the details clearly, concisely, and enthusiastically. Most people are aware when something is not working well, but don't necessarily want to solve the problem themselves. If you can give them a potential solution to back, most people are willing to give it a shot. This makes my work more enjoyable, since I know that my effort will go somewhere, and even better, that it is in the direction I think is best. As a result of my service position now (which concerns a topic I care about), I have an outsize say in some aspects of how our department works.

4. Saying yes to easy and/or interesting tasks helps you say no later to onerous or crazy-making ones
You have to say yes some of the time. You may as well say yes (or even volunteer) for some easy stuff, so that you can say with all honesty that you have a pretty full service load when future unpleasant tasks crop up. The last minute assignments are usually to difficult or frustrating things. May as well agree earlier to easier stuff.

5. People (even service slackers) do notice when you do service, so make sure they don't take advantage of you!
This may not be to your benefit (since it means people know who to ask for "favors"), but it is true. Everyone knows who the slackers are, even if there is no real penalty for being one. Having a reputation for doing the things you commit to and for getting things done is a good thing, even if there are no immediate rewards. That said, it is definitely possible to be taken advantage of. It is definitely possible to overdo it, and it took me a while to figure out when and how to say no. I think I am at a reasonable balance now, where I do enough useful service to keep things going (and feel like I am doing my part) without it being a major drain on my time. 

Thursday, March 23, 2017

Active learning in the classroom, flipped or otherwise

This started as a comment, but became a post when it started looking like a novel! There is a really interesting discussion going on at xykademiqz's blog in the comments section (starting here, where commenter idm asked about active learning). Xyk's comments about flipped classrooms sync with my experiences, namely that in a flipped classroom, the class covers less material AND that it is easy to screw it up so that the students learn nothing. While it is certainly possible to do a bad job in a lecture based class, I think it is harder to do it so badly that students may as well have not taken the course. In my field, less material per class means that even if done perfectly, flipping all the classes would mean that students either take much longer to a degree, or start out well behind colleagues coming from unflipped programs.

Like Xyk, I find it really irritating that active learning now means no lecturing. I find that I can get students to ask (and answer!) questions in class, even in a room of 200 students. In my lectures, I often stop and poll the students/get them to ask questions/have them set up or solve a problem/demo something or show them a video showing a concept in action, etc. Even an audience of researchers really excited about a topic loses focus if a seminar goes on too long. A lecture-based class does not necessarily mean the professor drones on for the full class time every time. (Also like Xyk, I was an extreme introvert as a student, and would have hated flipped classes and found it difficult to learn if I were forced to interact with others the whole time).

Another commenter (Alex) points out a really interesting study in physics, which suggests that students learn concepts better in a flipped classroom (consistent with most studies), but learn problem solving better in a lecture-based classroom. This is not too surprising to me. Students learn problem solving by wrestling with problems, and they do more of that as assigned homework in a traditional class than in the 150 minutes of problem solving in a flipped class. Plus, watching someone problem solve in a video is not the same thing as doing it live, where you can interrupt if you get lost or confused.

Back in the olden days when I was a student (which was well before flipped classes became a thing), some of my smaller, focused, upper level courses were taught in a hybrid style, where at least some of the class time was used for interactive problem solving (usually one student at the board working a previously assigned problem, with the class discussing the strategy and/or comparing strategies). This was really effective--I still remember some of those classes many years later, especially ones where I was at the board! This mania for flipping things also forces people for whom that style doesn't work well to go against their strengths, just as forcing everyone to lecture would hurt those for whom a more active/flipped learning teaching method is better. As with anything, there is a time and place for everything, and perhaps entry level STEM is not that time or place (at least for folks expected to have problem solving skills like scientists or engineers).

Monday, February 27, 2017

Can we return to funding stability?

In the olden days (i.e. for my PhD advisor) in the physical sciences, the funding strategy was simple--get in with a program officer at one of the major funding agencies for your bread and butter grant. Then supplement that with additional proposals. For ProdigalAdvisor, his core grant was DOE. As long as he was productive, the renewal of this grant was more or less guaranteed. It was enough money to support 2-3 students, or a student and a postdoc if budgeted carefully. He knew his PO well, and had regular contacts.

The upsides to this, compared to our current situation are obvious. First, having semi-permanent funding enables risk taking in other proposals, since they aren't make or break for the lab. It also means less time spent proposal writing and more time on other things. Next, it makes the disconnect between PhD time (5-6 years in my field) and grant time (usually 3 years) less painful, since it is unlikely that funding will end mid-PhD causing a shift in research topics midway through. Finally, the reduced mental overhead of near constant worry about funding improves pretty much everything. Because funding was not guaranteed without productivity (and POs do come and go), there was still pressure to produce, but not as acute, and not to the exclusion of all else. 

However, there were also significant downsides to this system. The main one is that it HEAVILY favored people hooked into the Old Boy Network, who could be introduced to the POs they need to network with to get into this kind of favorable funding situation. POs tended to stick to people who reminded them of themselves when younger, which suppressed diversity big time. It also heavily favored older established scientists (who were hooked up with semi-permanent renewals) over younger, upcoming scientists (who had to fight tooth and nail for funding until they could hook up with a PO willing to fund their careers). POs tended to stick with "their" stable of researchers, even past the point of when they were producing good science. Finally, it reduced the "nimbleness" of the funding agencies to change research directions and priorities as situations changed.

Is it possible to get the good of stability in funding without the bad of locking in the status quo for years and years? I don't really know. It seems that both systems were about equally likely to favor famous, established groups and also about equally meritocratic (though the optics now MIGHT look a little better now). Diversity is better now--the Old Boy Network is still present, but not quite as dominant. Risk taking in research is probably less common than before, due to the danger of being unfunded. There always were people who wanted to push the envelope/got bored with their current areas/were creative scientists of any age, but these people could get to stable situations in the past that are much more difficult to reach now. My PhD supervisor, with his semi-guaranteed DOE grant, was not a superstar with a megagroup. He was just a "normal" PI.

I am much, much more worried about funding than my advisor ever was. All of my colleagues are also worried. I spend a huge fraction of my time writing proposals, and I know I write many, many more than my advisor ever did (not just conjecture--we've discussed it since I moved to the TT). Both systems favor flashy science that can be packaged up and sold, and extroverted people with good networking skills, which doesn't necessarily correlate with important science and research talent, respectively. The modern funding schemes favor trend following, and rapidly move funding into and out of areas when they become hot and cold, which is not necessarily a good thing in basic research if we want to develop a deep understanding of something, which takes time. It sometimes seems to me like we moved the deck chairs around to generate at least the appearance of a more meritocratic, inclusive approach and called it progress, while forgetting that the quality of the experience for those in the system is also important.

Thursday, February 23, 2017

Teaching a new class

The first time I taught a class, it took me (on average) 6-8 hours per hour of class time to prep, not including exam writing, homework assigning, or grading, which is obviously unsustainable. As I got more experienced with the material (and with teaching), I could drop this down to 30-45 minute per hour of class time. The first time I taught a new class, I was panicked to think I'd be back to 6-8 hours, but teaching itself is a learned skill, and I find it takes a lot less time to prep even new material.

Less time, that is, than 6-8 hours. It still takes me a long time to prep new material. I like to do a good job, so I try not to skimp on the class prep. Before I start a new course, I dread the extra work. It seems like a huge mountain of additional stuff I don't have time for. That said, I find that once I am into it, it goes faster than I fear, and I enjoy it. I like learning new things, and I find that having to explain things to students helps me deepen my understanding, even of material that I know fairly well. When the course includes things I haven't really used since I was a student, I find that my much deeper knowledge base now makes me appreciate things I glossed over as a student. It is still a ton of work, though.

My department here at ProdigalU does a good job protecting TT folks pre-tenure. Most people get to keep the courses they start with for their whole run to tenure. Post tenure, our department tries to give people a minimum of 3 years with a new course. However, since life happens (with emergencies and sabbaticals and family leaves and everything else), sometimes people have to shift more often.

I think it is important to rotate instructors. Not just from a fairness perspective (since we all know that some courses are more work than others), but from a teaching perspective as well. I find that 5-6 years is the ideal time for me to have a course before swapping. The first 2 years, I am still getting a feel for the material and how the students respond to it. I'd say I am at peak teaching performance (for me, anyway) in years 3 and 4. By year 5, I find that I am getting a bit stale in the class. Definitely by year 6, I am ready to move on.


Wednesday, February 8, 2017

Summer undergraduate research

'Tis the season for making summer plans, and undergrads are looking for research opportunities. I love having summer undergrads in the lab--they are usually super excited to be there, and I enjoy interacting with them. We've had great undergrad students in the lab so far. About a third of them end up co-authors on publications, which is win-win for everyone. I always ask my grad students if they want to mentor someone before signing them up for it, and many of my grad students ask for summer students even before I get to ask them about it first.

This year, for the first time, I am relying exclusively on interviews to decide on whether I will take a student or not. I have never been a huge believer that GPA is a good way to award these kinds of opportunities, so I thought I would put my money where my mouth is. With one caveat: I want students to be fairly compensated for the work they do in the lab, so low GPA students may need to figure out how they can do this (with my help, of course).

I strongly believe that students should receive cash or credit for research, which is after all, work. Since the coffers are a bit bare right now, I am telling all potential students that they must do one of the following: 1) meet the requirements to take research for credit (3.0 minimum GPA at ProdigalU) and sign up, or 2) apply for and receive a research award to cover at least half of their summer stipend, or 3) qualify for work-study so they can be partially paid through that program over the summer. This is something that actively troubles me, since while I don't WANT to have to consider GPA, with my funding situation the way it is, I have no choice. So far, half of the students I've interviewed are specifically looking for research for credit opportunities, since they are taking other classes anyway and would like to get some research experience as well. The other half plan on applying for summer fellowships.

We are fortunate because ProdigalU has loads of summer research fellowships students can apply for, but that usually doesn't help low GPA students. I haven't had any approach me yet about research this year, but if they don't qualify for work-study (and a large fraction of our students do), I will have to turn them away. I am not sure what to do about that--I literally don't have the money to pay full freight on an undergrad. At the same time, it is unfair to limit research opportunities to students wealthy enough to be able to volunteer. Summer research is a really different experience than undergrad research during the academic year when students are pulled in many directions by their classes. It is very sad that these possibly life-changing opportunities are pretty much the first thing to go when research money gets tight.

Thursday, January 26, 2017

Scientists march on Washington

Just in case you haven't heard about it elsewhere. Please check out their website.

Tuesday, January 24, 2017

Tenure, job security, and academic freedom

Recently, loads of people inside and outside of academia have been calling for an end to tenure, saying essentially that is it outmoded job security that no one else has, so why should professors? The common counters to that include that tenure is part of the compensation package (true), that it keeps academic salaries down compared to scientists elsewhere (maybe, but I don't think so), that it is needed for shared governance to function via preventing administrative retaliation (probably). Many people ignore or downplay the academic freedom part, or associate it with teaching or with research in social science, arts, or the humanities rather than science or engineering.

I never bought this argument, especially in light of what happened with climate research. In the late 80's and early 90's, climate science was not controversial. Scientists did their thing, and no one got really upset (or even noticed much) about the results. But then climate change became news (and more obvious as lived experience), climate change denial became a thing, and more importantly, a thing associated with one of the parties in our two party system. Suddenly, people like Virginia Attorney General Ken Cuccinelli were investigating climate scientists like Michael Mann for fraud. And suddenly tenure meant something, at least for climate scientists.

There have always been political attempts to control research (I was even mentioned in one!) And now as per DrugMonkey we are here on day 4 of the Trump administration, which has now gagged EPA scientists and frozen EPA research grants, stopped USDA researchers from communicating with the public, and told HHS staff to stop external communications. The Trump administration is neither the first nor the last to actively interfere in the research enterprise, which makes the academic freedom guaranteed by tenure all the more important. We got complacent. Now we need to be vigilant.

Tuesday, January 17, 2017

What does it mean to be not cut out for PhD research?

Inspired by a comment from SEBASTIAN RAMIREZ on a previous post here, I started thinking about people I've met who have not been cut out for PhD research in various ways. Before joining the TT, I would have though this was an easy thing to determine. However, experience has shown that there are many things that can make a person not cut out for PhD research, some of which are harder to see than others (for both the student and the mentor). The stereotype is that people not cut out for a PhD either lack technical/intellectual skills or are lazy, and therefore this can be determined right away. However, I find that this is not usually why people leave our PhD program without a PhD (though it does happen, usually in the first year).

Lack of desire: This is the most common reason people leave our PhD program. I've seen plenty of people who like the idea of research but not the reality--some don't like the uncertainty (is there even an answer? will I ever find it?), some don't like the repetitiveness required to ensure reliable data, some don't like the large problem solving component, some don't like the required background reading/lit review and just want to do experiments. Some people think they like doing science, but actually just like reading about discoveries. Some people did well in their science classes and apply to grad school because they don't know what else to do (and at least they can get paid). Some people apply to grad school due to family pressure. Some people apply to grad school because it provides a way out of their country. Some of the people who start a PhD program without a direct desire to do research discover they enjoy it. Some don't.

People can fool themselves for quite a while about what they actually want, so students can be pretty far along when it becomes clear that they don't really want a PhD and/or dislike research. Sometimes they attempt to tough it out anyway to try to not "waste" time sunk into a degree or because they don't know what they want to do instead/family pressure/need the paycheck/don't want to leave the country. This doesn't always go well because it is difficult to keep working hard for something you don't really want when you also have no interest in the work.

Inability to mature as a researcher: Many of the students who leave our program with an MSc instead of a PhD have good lab and data analysis skills, but just cannot make the leap to doing PhD level work. This can be hard to spot until 2-3 years in, when students in the PhD program typically are driving their own projects. These students can design and complete experiments, but have difficulty deciding what experiments should be designed. The cliche way to put this is that these students can't see the forest for the trees, but this is not exactly it--they are not overwhelmed by details, or overly detail oriented, but they cannot see how their work fits into a big picture in a way that lets them run a project themselves. Sometimes PIs push these students through the PhD program, but that really isn't in their best interest, since future employers will expect PhD-holders to be capable of driving a project.

Need structure: This can be a huge issue. Students starting out get lots of help with planning and carrying out their research, plus they often start out doing some coursework. As students go further into a PhD program, students are expected to take on more and more of the planning and scheduling themselves. Classes (if any) are completed. Most of the days are wide open to get work done. Some people cannot be productive in an unstructured/unscheduled workday. Some people need deadlines--for these people, even the weekly deadline of group meeting or a progress meeting is not enough pressure to help them focus. Most PIs are not interested in having daily scheduled meetings with senior students (which can help people who need deadlines), but some will do frequent meetings. The lack of structured days needs to be solved by each person though. Sometimes, they just can't do it.

Monday, January 9, 2017

Common mistakes at the on campus interview

In honor of academic interview season, I thought I'd discuss a few common pitfalls I see candidates fall into on campus. Way back when I was interviewing, I made some of these mistakes myself, so I know how common a trap it can be. For those job hunting, xykademiqz has a great post on advice for Skype interviews, some of which carries over to on campus interviews as well.

The research talk: A good research talk will engage both experts and non-experts for at least a portion of the talk. It will also demonstrate your technical and research skills to the experts. Thus, it must have both breadth and depth. This is where I fell down in my first year interviewing. The most common mistake is to focus too much on depth (I did this), but it is equally problematic to focus too much on breadth (I've seen this) and not come across as an expert in anything.

The chalk talk: The chalk talk is hard, because it is unlike anything else most people have done in the past. One major issue is in presenting a bunch of projects that look like just a bunch of projects. From years of experience, most postdocs are used to thinking about their research in a project by project way. However, in the chalk talk, we want to see an outline of what the candidate's research program will look like over time. Pretty much all of the candidates invited to campus had some long term research plans in their research statements, but this needs to come out in the chalk talk as well. Our department wants to hire someone who will have a successful career, and that means making sure that the planned research leads somewhere beyond the immediate 3-5 year project horizon. What will your lab specialize in? When we discuss our new hire, what is the one sentence summary about your work that we will lead with?

Related to this issue is candidates who position themselves in competition with their previous groups. It is great that you love your current (or previous work). It is great that you can take it with you. HOWEVER, unless you can clearly articulate the differences between your approach and your supervisor's approach to non-experts, some in the audience will wonder if you will be successful in funding your work. Towards that end, a really common question is "why would a funding agency give you this money instead of your more established adviser?".  You need to have an answer.

The opposite problem (moving to a brand new area) can also be problematic. Without a track record, some will be reluctant to take a risk on a candidate. Also, proposals into a new area can come across as naive or not fully formed. Realistically, at ProdigalU we do consider the experience level of our applicants, so if one of three proposed projects in a new area comes across as naive or unrealistic for an inexperienced candidate, we are usually pretty forgiving if the other proposed projects are well done. This is particularly an issue for candidates trying to force their research to fit into a targeted job search, and it becomes obvious in the chalk talk unless the candidate is really interested in moving in that direction for the research's sake rather than just to get a job.

A final chalk talk issue is particularly common with those from large and well-funded groups. Really innovative research is a great attention grabber, but as a PI, you will have real students who really want a degree and therefore need to have some research success. You need to be able to articulate not just Plan A, but also Plans B and C for your projects, and also discuss what important (i.e. publishable) work will come out of your planned research directions even if the project is ultimately unsuccessful. There are no great rewards without risk, but you need to show some awareness of the other requirements of your research program (like that you will be training students, not just doing cutting edge research).

Personal interactions: Do not make flip or ironic remarks that could be misinterpreted (I did that when answering questions after my research talk my first year interviewing, and it was a mistake). Personal experience aside, every aspect of the visit is part of the interview, including the meals, the walks between offices, the interactions with non-academics, and the transportation to/from different locations! Be polite to everyone (including all staff members and students). Do not drink more than you can handle and remain in good control of yourself, regardless of how much your meal companions indulge. Do not make disparaging comments about pretty much any person, group, or place. Do not bring up politics as a topic of meal time conversation. I have seen candidates do all of these things, and it did not make a favorable impression.

Thursday, January 5, 2017

Happy new year!

Happy new year to my readers! My science resolutions this year:

1.  Set aside regular time to just browse through the literature and read what is interesting, rather than directly useful.

2.  Spend more time with my undergraduate researchers. 

3.  Try to blog regularly.