Category Archives: Education

The Levels of Research

Colloquially, we use the word research to generally mean doing your own digging, studying, and analysis of a topic. For example “I need to research the best recipe for potato salad,” or “Go research what mayoral candidate you should vote for.” It usually starts with a question like “What’s the best potato salad?” Then you doing some searching and reading of articles, reviews, comments, and analysis by other people. Then based on that information, you can make a decision. In scientific research, it’s kind of the same idea. You ask a question (how or why does thing A cause effect B), then you do some reading on what’s been done in the area (Bob studied thing A and saw effect C. Tiffany saw effect B but with thing D). The last important step, making a decision, is where things depart from the common usage of research. After the reading step, there’s a test step where you use what you’ve read to try to answer your question, assuming it hasn’t been answered already. This involves doing experiments, getting data, wearing lab coats and the other trapping usually associated with the idea of scientific research. However, this is not usually the most important part. The important part is the analysis of the data and the drawing of fundamental data-backed conclusions to “make a decision” on the original question. It’s in this last conclusion step where science happens and is usually the hardest part for new students and researchers to grasp. Without a data supported conclusion, you just have a bunch of numbers that sort of say something, but you don’t really know the fundamental details.

At the university, there are usually 3 levels of research: undergrad, Masters, and Ph.D. They increase in difficulty, depth, and expectations. The level of detail expected can generally be broken up into the various steps in a given research effort: test and observation, correlation and comparison to known ideas, and proposal of new idea supported by the data.

The first, test and observation, is typically where undergrad research students function. Undergrads usually worked with or assist a graduate student. Assuming they’re not relegated purely to chores like cleaning test tubes, an undergrad’s main tasks are to assist with experiments. That may include sample preparation or experimental set up, test operation, and data collection. That’s what most people associate with research, the actual experiment. More experiences undergrads may move to the observation stage, also called data analysis. Here, you not only take the data, but you also try to process that data and look for trends or behaviors like does thing A cause effect B to increase or decrease. These are usually pretty straightforward observation. This is usually where an undergrad would end, since moving to the next step of correlation and comparison requires some specific knowledge of the subject the student may not have. And many undergrad thesis focus mainly on presenting data, trends, and maybe touching on possible causes based on existing knowledge.

The second step, correlation and comparison, is where Master’s research typically sits. To compare, you need to read and know what to compare against. Thus MS students is where a lot of literature reading begins. By reading and understanding the literature, you can make more educated arguments about your observations from the data. Often times MS research will use existing formulas and theories learned from the literature to support the conclusions. For example if you observe A causes B to increase, and you find that other people saw the same behavior in a similar but not the exact same experiment, then you could make the comparison to that work. Alternatively, your data may fit in the general trend of other people’s results, allowing you to correlate how much B should change for a given level of A. This level of research present a deeper understanding of the physics by using similar research and existing understanding and equations as a guideline for your work. You are still contributing new information to the greater body of knowledge, but there’s usually not enough time to try to develop new understanding and equations.

The last step, proposal of new ideas, is what’s required of Ph.D. students. A Ph.D. student has to do everything an undergrad or MS student does, but then has to go one step further. They must take the same data and observations, the existing understanding and equations, and discover new understanding or equations that describe the fundamental physics. Continuing the example of A causes B to increase, perhaps the existing understanding says it occurs because of a phase change in B. At the Ph.D. level, you must discover why does A cause B to change phase, that’s the fundamental physics. This is usually pretty hard and requires significant time investment to take careful measurements, analyze each set of tests to help determine what conditions to test next, and lot and lots of paper reading to understand what other people are saying or doing in your area. And there will be many set backs and errors along the way, resulting in redoing things until it’s right.

So in summary, the three level of research can be summed up generally in the questions of what, how, and why. What happens? How does it happen? Why does it happen? For a Ph.D. you have to answer all of those, and hopefully enjoy doing it, otherwise a research career may not be for you.


TA or RA

The Fall semester is ending and Spring will start soon. It’s about this time, or sometimes even earlier when departments and grad students will ask about funding for the Spring semester. For those not familiar with graduate school funding, grad students usually do not pay for school out of their pockets. There are monies available from the university, the research advisor, or external fellowships that pay for entirely or partially for tuition as well as a stipend. So it’s kind of like getting paid to go to grad school, though they money you see in the form of a paycheck is not a lot. A good chunk of the money is directly put towards paying for the tuition. So why the grad student may technically be budgeted $40,000 per year, half or more may go towards tuition. $20,000 isn’t a lot, but usually enough to live on for a few years while you finish school, assuming you don’t live in super expensive places like New York, DC, or LA.

Now back to the different monies. First, grad funding is typically reserved and used to allow students to do research toward a thesis or dissertation. Section, not all degrees or programs will have graduate funding available. In STEM fields there’s usually a lot of research while in humanities there isn’t as much. All three will pay for tuition and stipend, though the amount for stipend may differ. TAs or teaching assistants are paid for by the university, or the department more likely. A TA’s job is to teach labs, grade, or otherwise assist the faculty with the undergraduate courses and sometimes graduate courses. That takes a significant chunk of time, maybe 15-20 hours a week in total. On top of TA duties and their own courses, the student will try to do their research towards their thesis. RAs or research assistants are paid to work on a specific research project defined by the research contract or grant their research advisor is working on. The RA’s job is to do the research as dictated. While this sounds perfect as you’re getting paid to do research which, it assumes the dictated research is what the student wants to do and can lead to a thesis. I’ve known many students who have worked on projects as a RA that they didn’t care for or was too short-term to lead to a thesis. They did it because it paid them, but didn’t move them forward towards their degree in a good fashion.

Fellowships are perhaps the best money because it is given to the student with no prescribed teaching duty or research project. So if the fellowship student can find the appropriate research advisors who has a project that fits the student’s desired topic and can lead to a thesis, then he or she gets paid to work on exactly what they want. This is of course a rare situation as there are many more grad students each year than there are fellowships.

Now the question of which is better, a TA or RA? It’s more of a personal choice I think. If the RA project is something you’re interested in and can lead to a thesis then it’s clearly the better option. However, an additional caveat about RA’s is since the funding comes from a contract or grant that will end at some point, the RA’s funding may end at some point. Hopefully it’ll last through your thesis or dissertation, but not always. On the other hand, TAs positions will usually continue year after year as long as the university has money, so it’s a more guaranteed funding. The TA is also more free to do the research on their choosing, as long as the advisor agrees. Of course the downside is TAs has to grade or teach labs which eats into their available research time. I have had students in the past that prefered a TA position to RA because the RA research was of no interest to them and they’re rather do the TA and continue their desired research and hopefully find RA funding for their research in the future.

If you’re considering grad school in a STEM field, you should definitely look at applying for fellowships since it gives you the most control and freedom in your research. But in general you shouldn’t be paying for grad school if you’re a good desirable candidate. Most fellowship deadline are in the Fall semester, a year before you’d start grad school or get the money. You should also contact faculty in your research interest area at the university to see if they have RA positions available. The answer will usually be no or I don’t know as most faculty use their RA funding for students to do the work as soon as possible, but never hurts to ask.

Are You Really “Bad at Math”?

I was visiting with a friend at a party and playing a card game. One of the players said nonchalantly “I’m bad a math” like it was a joke when we were totaling scores. Everyone smiled and continued on. I’ve heard that phrase a lot in the classroom and office hours. It’s something I’ve heard a lot growing up as well, usually from those of my peers who were struggling in math or science class. I also tend to hear it from adults. Being “bad at math” has become an odd badge of honor for people. In the U.S. it’s become socially acceptable to be bad at math. But if you visit any eastern country, no one would willing admit being bad at math. It’s an embarrassment. It’s almost like saying you’re illiterate, which is something no one wants to be admit because being able to read is one of the most basic skills. It should be the same with math. No one should be bad at math, and most probably aren’t in reality.

When we say math, people commonly think arithmetic, the nuts and bolts of addition, multiplication, fractions, etc. Math is really about patterns and logic, two things humans are quite good at naturally. Pattern recognition is instinctive to most animals and part of the survival instinct. Humans have developed that ability further and put numbers and logic to patterns. In everyday like, we all put that ability to use when we look at the clock and realize we’re late to a meeting just by the location of the hands. So we’re all naturally good at the basics of math, but it’s the technical details we tend to get lost in.

I give a short speech at the beginning of my lower level engineering courses. I explain how engineering is a skill, like music and sport. It’s something you have to practice to get better, and just about everyone can get better. It will be easier for some than others just like playing a musical instrument. Math is much the same. Being good at math means developing the skill to do math. And like all skills, it takes practice. You just have to be willing to do the practice, and I think that’s where people fail. Truthfully, if you don’t want to play the violin, you’re not going to want to practice, and you’re not going to get better. So if you have a distaste for math, then you’re not likely to want to practice. However it’s a skill we all should have. And I think you’ll find, like most skills, it starts hard and is a chore, but as you get better you’ll enjoy it more (or at least hate it less). So go out and practice the math skill.

The Changing Purpose of Professional Conferences

Last week, I along with two of my students flew to the west coast to attend the AIAA SciTech 2016 conference. It’s the AIAA’s largest conference each year and combines just about every discipline they serve except for maybe propulsion (there’s a separate conference for that). There were thousands of attendees (~4000 I was told at a committee meeting) and 1000+ technical papers and presentations. I never counted. I’ve been going to conferences for a long time and I’ve noticed what I’ve paid attention to and my personal goals at conferences have changed. That makes sense. As your career level changes, your job description changes, and thus what you need or look for to get the job done also changes. So let go through what a conference may mean for different people in academia and what they may want to focus on or try and get out of it.

Undergraduate Students
If you’re an undergrad presenting at or attending a conference, it’s pretty much a spectacle for you. You should be there mainly to meet people, see what research sounds cool, grab vendor swag, and have a good time. There’s almost no need for you to shake hands and schmooze besides for the fun of it. It’s always fun to talk to new people and learn new things. If you tell people you’re an undergrad, you will most likely get advice, asked for or not. The majority of conference attendees are academics, industry, and government professionals. In most cases they’re farther along in their career path. As social animals, humans like to share experiences and give advice. I think giving advice makes us feel good on some level. So take advantage of this if you don’t mind advice and ask questions that pertain to what you want to do. If you’re a senior, it’s also a great chance to ask about employment opportunities. A lot of times you may get directed to the company’s website, but if you hit it off with someone and give them a business card or get their email, that may help get your foot in the door.

Graduate Students
If you’re a grad student at a conference, you’re mostly likely presenting research. If you’re a first or second year grad student, you’re operating on a similar level at conferences as undergrads. You’re still considered “new” to the field and you should mainly go to present your work, meet people, learn things, and have fun. You’ll also get lots of advice.

If you’re a senior grad student looking to graduate, then your focus changes. Now you’re looking to meet people not just for fun, but to look for job opportunities. This puts a bit of pressure on you since there’s now something at stake, namely your career. Looking for employment at conferences is not like a job fair. Mostly people aren’t actively seeking people to hire at conferences. So your best bet is to strike up conversations, talk about what you do, ask about what they do, and see if there are mutual needs. One strategy I’ve found that works most of the time is to say you’re looking for a job and ask if they have any advice. Going back to humans are social animals, you’ll almost always get some kind of advice. They’ll also usually ask what kind of career you’d like, whether it’s academia, government, industry, or otherwise. If you don’t have a preference, then pick the answer that matched with the person you’re talking to. Getting their advice will put them in a more helpful frame of mind since it make the speak feel like a mentor which engenders a bit of desire to help. Then you can ask if they know of any openings in their place of work. Most of the time the answer may be no, but they’ll usually trade business cards with you. Then it’s your job to email them after the conference to follow up and see if the mentor feeling lasts long enough for them to poke around in their work place for you.

This is where I fall right now. As an academic, and a new one, my biggest goal is probably to try and get funding and collaboration. Both aren’t easy. In the same way people don’t go to conferences to hire people, they don’t go to give out grants either. The government agencies all have defined processes to give research funding. At a conference you can talk to the program managers and get a sense of what they’re interested in, but in my experience the program managers tend to be general in their advice. No one ever says “write this proposal and I’ll fund it.” Well, there is the exception that was the BP oil spill when certain agencies were funding lots of research in oil cleanup.  But unless there’s some crisis that needs immediate attention, program managers aren’t usually specific.

Industry managers are the same way, though they may be more willing to give you specific interests based on their current projects and products. It is possible to start the process to funding with industry from a conference by meeting them and setting up a later visit.

The second part of conferences is collaboration. We academics are very specialized in our research because we’re just one person. While the university may have hundreds of research areas, we don’t represent the whole university. So to do complicated research, you need help. A lot of times that help can come from other people at your university. Sometimes though the specialty you need is not at your home school, especially for smaller schools. Thus you have to look elsewhere. Conferences are a pretty good place for that since you get to see the kinds of research different groups are doing through the presentations. Most of the time people are willing to talk about collaborations, especially academics. The question of money and funding will always come up, but the first hurdle is to get the conversation started.



The Trouble with Solutions Manuals

This semester I taught Statics for the first time. The material isn’t very difficult. It’s the same concepts applied to different situations. I assign homework problem from the book as a way for the students to practice the skills and learn the material. However I’m finding that students have easy access to solutions from online resources. I don’t mind if they look at a solutions manual to help them understand how to do something. The homework is not worth many points in their total grade. However it seems the availability of the solutions is making them dependent. Most of the students can do the math and write the equations, however many don’t know where to start a problem without the solutions. They can’t go blind into a problem and solve it, and that’s killing them on exams.

So I think the true issue of solutions manuals is not they give the answers, but they acts like a safety blanket and make the students dependent on the existence of solutions. Perhaps the solutions reduces the anxiety of working the problem, knowing the answers are right there if you want to look. But in real life there are no solutions manuals. In your job no one will be able to tell you the answer. You have to to be able to work the problem and know if the answer is reasonable and looks correct.

So my suggestions for all students is to always do your homework blind first, without the solutions. Then once you’re done, you can go back and check your answers. Don’t jump immediately to the solutions at the first stumbling point. That creates a dependency and doesn’t teach you how to examine and solve the problem.

To Ph.D. or not to Ph.D.

There are three main levels of academic degrees you can get in college. The first level is the bachelors degree. The name comes from the 13th century when junior members of guilds or universities were referred to as bachelors. It originated from “knight bachelors”, who were young or poor knights who could not gather their own vassals. The bachelors degree is kind of like the “you must be this tall” sign on roller coaster rides. It’s a basic requirement for most jobs, and it’s often sufficient.

The next level is the Master’s degree. The name originates from a variety of old words (Latin: magister, Old English: maegester, Spanish: maestro, etc) that reflect authority and in academics one who have the authority to teach at a university. The Master’s degrees gives you a little more knowledge in a specific area. Most professionals end up getting a Master’s degree at some point because it usually qualifies them for a pay raise, and often times the company will pay for the tuition.

The last level is the Doctorate of Philosophy or Ph.D. This is the highest academic degree you can obtain. (Note I’m not discussing specialized degrees like JDs or MDs). Interestingly, the Ph.D. as we know it, meaning requiring a research component, was first introduced in Germany in the 19th century. It was very successful and attracted many foreign students, especially from the U.S.

So, the point of this post is to discuss the pros and cons of pursuing that final degree, the Ph.D. When students, usually undegraduates, ask me about graduate school and going for a Ph.D., I always start with a warning. A Ph.D. is a double edged sword. It qualifies you for some specific kind of jobs related to research and development, but disqualifies you for a whole lot of other jobs. If I put this on a number line with all possible jobs from 1-10 with 1 being entry level, then a Bachelors and Master’s would give you access to 1-7 or 8 while a Ph.D. would give you access to 9-10 but deny you access to 1-8. The latter is not because you don’t have the skills for an entry level position, but because you are over-qualified and a company would not be able to pay you a proper salary in an entry level position. Even if you say you’re fine with less pay, the company would worry that you’re secretly unhappy and would quickly find a new position and leave. Getting a Ph.D. is also a long and often times arduous process. The typical Ph.D. takes 5+ years to finish and can take 10 years in some fields. Ph.D. students are also not paid very well. They tend to live just above the poverty line, and in some places they would technically qualify for welfare. So for the degree, you are giving up 5-10 years of a good salary if you had gotten a job with your bachelors or masters degree.

The other downside is a Ph.D.’s job usually doesn’t end at 5 pm. Most Ph.D.’s are hired into R&D or managerial positions where there are many tasks to do themselves or to oversee. The inevitable truth is it takes more than the standard 40 hours a week to get all the work done. So you will be taking work home with you, that’s almost a guarantee. For example, I often put in 10-20 hours a week outside of the office, either working on classes or research.

So what’s the upside of a Ph.D. then? Because there are a lot of downsides. The one single upside is you get to do research and hopefully interesting research. I know that doesn’t seem like much of an upside, but it works for some people. For all the years in school and long hours, your compensation is more work, oddly enough. Some may say this makes us crazy people who work only to do more work. I like to think of it as we find what we do fulfilling enough that the downsides do not seem that important.

Now admittedly I do wish I had more free time to do other things like play games, read for fun, and spend with my family. But I don’t mind reading for work or thinking about research when I’m sitting on the couch. I find it interesting if not exactly enjoyable. So I do not recommend a Ph.D. for most people. A “normal” person has hobbies outside of the job and would rather spend their time at their hobbies. But if you thinking a life of research and scholarship may be right for you, go talk to a Ph.D. I would strongly recommend doing some research at the undergraduate level when the risk is minimal to you. Undergraduate research gives you a chance to see what you would be spending 5+ years doing for the Ph.D., and an inkling into what your career may be like. Now in truth, there is a lot more to a Ph.D.’s job than just research. There’s also a lot of people management, time managements, budget management, and sometimes coming up with ideas for funding. The first three are not specific to a Ph.D. job and can be found in any upper managerial position. The last is less common, and usually difficult to teach. But I’ll leave that for another time.

What, Why, How, and So What of Paper Writing

One of the most difficult and perplexing part of research and by extension graduate school is writing and publishing your work. It’s a key part of the research process and important for building your own career. But how to write a technical paper well is often unclear to new students. It took me a few papers before I got there hang of it, and sometimes it’s still hard to figure out where to start. The easier part for a student to write is usually the results since they have the data. But just data is not enough. You can tell you’re getting better when there are fewer revisions between you and your advisor and they make fewer red marks.

Your research paper or presentation is a bit like a sales pitch and debate rolled into one. You want to give a clear, concise, and logical argument for the validity and impact of your work. On one hand you want to convince the audience that your work and results are significant and add to the greater body of knowledge like a sale’s pitch. On the other hand you need to defend your techniques, results, and theories like a debate. For my students, I tell them the paper and/or presentation should answer the “What, Why, How, and So What” of the research. These are usually what the reader or audience is looking for, consciously or not.

What: What are you doing/what did you do? This is the simplest question to address. This could be a few sentences in the abstract or intro that talks about what the work you’re taking about is and the goal of the project.

For example: “This work conducted experimental measurements of the size distribution of the Whos in Whoville using electron microscopy. The goal is to determine if exposure to elephants alters their growth patterns.” (I recently watch the movie Horton Hears a Who.)

Why: Why are you doing this work? The immediate answer maybe “because my advisor/boss told me to.” That is not the answer you’re looking for. The why question address the big picture. Why is this work important? What kind of impact can it have if successful? This is your selling point if you will. Maybe your work can lead to a cure for cancer, or more efficient car engines, or a better way to do space travel, or a better technique to teach children. The why question gives the reader and audience the reason to care about your work.

To continue the example: “This work is important because Whos are an endangered species that is easily affected by outside forces. The recent exposure to Horton the elephant is a major change to their environment, so we need to know if that event changed their anatomy.”

How: How did you do your work? This question covers the methods and techniques you use. You may have done experiments that used a particular probe or calculation, or developed an equation using a set of theories. So the how question address the validity of your process. In answering the “how”, you need to put on the debate hat and provide logical and technical backing and justification for the steps you too.

Example: “The size measurements were done using electron microscopy. Volunteers from Whoville were recruited to participate in exchange for ipods. A survey of each Who’s daily routine and their activities during the Horton incident were done.”

So What: So what’s the significance of your results? The last question is the take away the audience should have from your work. Until now you have presented on what you did, why it’s important, and how you did the work, and presumably the actual results. Now you need to bring it back to the larger picture and answer the question of “so what’s the point?”

Example: “The results have shown that the Horton incident caused the Whos to growth 5 microns taller on average, based on the sample measured. This indicates that the Whos are greatly affected by external factors. However it is not yet clear if the elephant was the primary factor in the height increase. Another external factor may be the high altitude of Whoville’s current location that increases their exposure to UV radiation compared to their previous sea level location. Further research is need to clarify the triggers.”

Amusing example aside, this is my technique in a nutshell for writing papers, and it generally lays out the paper in a logical order. What and Why would be in the abstract or into. How would be in the experimental setup or theoretical background. Then there’s the results section which is usually the first thing students think about. Finally the So What covers the discussion and conclusion. As you gain experience in paper writing, you’ll develop your own methods and you’ll recognize how others write. The goal is always to present a clear, concise, and logical argument for your work.