Sunday, August 21, 2011

Benefits of Teaching to Research

... [University of Virginia's David F. Feldon] and his colleagues gathered two sets of research proposals from 95 beginning graduate students in STEM fields—science, technology, engineering, and math—at three universities in the Northeast from 2007 to 2010. About half of those students taught, on average, one undergraduate course. The other half had no teaching responsibilities.

All of the graduate students submitted research proposals at the beginning of the academic year and provided revised versions at the end of the year.

Mr. Feldon's team used a rubric to rate several various aspects of the students' research skills, including the context of the proposed study, framing of the hypotheses, attention paid to the validity and reliability of study methods, experimental design, and selection and presentation of data for analysis.

The graduate students who both taught and did research scored higher on those measures, the study found. The results suggest that those students exhibited both superior methodological skills and greater improvement in those skills compared with their peers who focused on research alone.

"The findings resonate with people," Mr. Feldon said in an interview. "Of the people I've spoken to about this study, half said, 'Of course that's what you found.' The other half said, 'There's no way that can be true. Your data must be wrong.' Everyone's got an opinion on this, but there's been little data."

From Dan Barrett's story in CHE about Feldon's fascinating research published in the latest issue of Science (caution: requires subscription). Here's the abstract of that paper:

Science, technology, engineering, and mathematics (STEM) graduate students are often encouraged to maximize their engagement with supervised research and minimize teaching obligations. However, the process of teaching students engaged in inquiry provides practice in the application of important research skills. Using a performance rubric, we compared the quality of methodological skills demonstrated in written research proposals for two groups of early career graduate students (those with both teaching and research responsibilities and those with only research responsibilities) at the beginning and end of an academic year. After statistically controlling for preexisting differences between groups, students who both taught and conducted research demonstrate significantly greater improvement in their abilities to generate testable hypotheses and design valid experiments. These results indicate that teaching experience can contribute substantially to the improvement of essential research skills.


  1. ಅರುಣ ಪ್ರಕಾಶ said...

    Interesting article. I have always felt that the other way around is true. It has been my experience that people who are involved in research tend to be better teachers than those who don't (strictly at the universities, I cannot comment on the IIxs). They tend to be more sure and have a strong base of the subject they teach. They also find simple and alternate ways to explain things.


  2. Ankur Kulkarni said...

    I taught in my last semester as a graduate student and also found my research productivity increase. In fact, as much as the activity of teaching, it is the regimen that helped. Teaching gave a natural clock to the week, some discipline came about because some preparation was needed and some perspective and better articulation when I went back to relative simple UG level curricula after being in the thick of my Ph.D. area for a while.

    It may be worthwhile for IIXs to also try having graduate students teach.

  3. Vijay said...

    Graduate teaching assistants (TA) were around some years ago at IIT-Kanpur. I do not know if this is still so. Graduate TAs are, of course, common in the US. Teaching brings rigour to the TA's communication and likely feeds back on their research. Importantly, they bring a close link for the undergraduate to research: Bonds develop more easily between the TA and the undergrad, than with a more distant professor, and often results in the undergrad being attracted to research ( but sometimes repelled ;-) ?)
    PS: Abi: Will the IISc undergrads have graduate TA's?? Happy to enroll if you will consider 'outsiders' :-)

  4. Ankur Kulkarni said...

    In reference to Vijay's post... I am not sure if being a teaching assistant helped me in any way in my research. The work was light, required some mild preparation and could be accomplished with hardly any change in attitude. On the other hand, all the positive effects I mentioned came about only when I was an instructor for a course. Having the responsibility of conducting a course entirely on oneself makes a big difference.

  5. Vijay said...

    Thanks Ankur. That additional point in your 9:31 post is very useful to know.

  6. Abi said...

    @Vijay: While we at IISc have always had TAs; there has also been a wide divergence in the extent to which departments made use of their services.

    With the arrival of UG program, we now have a well-defined and well-organized program of teaching assistantships. Each of the first semester courses (except, perhaps, the humanities course) has about six to ten TAs who run tutorials and laboratory sessions.

    I don't think I can speak for IISc, but I presume our system would welcome outsiders to this program ...

  7. Nappinnai NC said...

    Sounds interesting although i disagree. I would say that a student who invests time on both teaching & research may not be able to do full justice to research as teaching requires lot of preparation(depending on the subject) and hence some energy is dissipated in teaching which could have been used for research. Although teaching improves one's communication skills, it is the quality of research a person does is an indicator of how strong he/she is in the field and also to see things/problems from various angles. Most Nobel Laureates and those who have done some ground breaking work but might not have hit the jackpot would fall in this category.

    Great example(wrt teaching/research) is comparison between Feynman & Gell-Mann. It is a known fact in the Physics circle that Gell-Mann was a better Physicist than Feynman but Feynman was an expert in explaining that Physics in an interesting way to Physics students or the general audience than Gell-Mann was! I don't think Einstein ever taught or had any student in his entire life although he had an assistant to do some Math calculation.

    Even Feynman(already won the Nobel when he wrote that book)makes a blunder in his famous 'Surely, you're joking Mr Feynman' that Einstein's assistant didn't know a simple problem in relativity when Feynman asked the assistant. When the assistant was interviwed by Einstein, he told Einstein 'I don't know relativity'. Einstein smiled and said 'Don't worry, i will handle that part!' Among all the years, 1905 is God's miracle year!!!

    So teaching skills do not necessarily correlate to great research skills. The advantage of teaching is, when research(even Profs, Nobel Laureates)work doesn't go well, one can atleast derive some satisfaction from teaching and the purpose of knowledge is to transfer. Teaching 'can' serve a person appreciate the psychology of people(as students can come from different cultural, social background) and also life. Funniest part of the impact of teaching happened in Feynman's life. When he was in higher secondary school, he used to talk a lot. So, the teacher said 'Richard, from now onwards you're going to sit in that corner and try to digest Advanced Calculus by Woods and don't open your mouth until you finish that book'. That's how Feynman developed problem solving skills in mnts in a short time but Hans Bethe & Paul Ulum were smarter than Feynman in crunching tough algebraic,calculus, contour integral problems.

  8. Wavefunction said...

    True, as you noted it's acknowledged that people like Fermi, Bethe and even Gell-Mann were greater physicists than Feynman. But Feynman was a genius :)