Sunday, August 17, 2008

On asking non-novel questions

One of my students, DC, wrote the following:

I started looking into the [student project that we discussed], and I found that there are already papers published about [that topic]. ... Since it seems like this topic has already been done, should I try to find another topic to look into? I've had the impression that if someone has already studied it, it doesn't work very well for a research topic but I know that must not be the case, otherwise no one would be able to prove theories wrong and there'd be nothing left to study... I suppose what I'm asking is if it's possible to still look into this, but in a way that doesn't only cover a portion of what another paper has already said (a paper that I'll have to cite, too)?
I responded as follows:
Hi DC-
An excellent question, and one that always needs to be asked. Very few people ever ask a truly novel question. Those who do are usually geniuses or lunatics or both. What we mostly do instead is try to ask the same question in a different context, or ask it better, or take a different and hopefully improved approach to answering it.
When LZ was hoping to design a project, she got interested in what caused mixis in rotifers, and I told her to go read the literature on that subject. She did, and came back upset because there were papers on the subject by respected rotifer experts, and they had already published answers to many of her questions. I told her to read those papers again with three questions in mind.
1. Which of her questions, or their own questions, had they failed to answer?
2. What areas of disagreement, apparent contradiction or uncertainty remained?
3. Where are the soft spots in the literature, meaning studies that could have been done better, analyses that are unconvincing or conclusions that aren't fully supported by the data they rest upon?

LZ, being both very hardworking and very clever, came back with answers to all these questions, and we used her answers, plus knowledge of our particular strengths, to design the study that became her senior honors thesis, and will become her first scientific publication.

Our strengths in terms of the primates, as compared to others who have written on this topic, are:
1. We have dispersal data on more species than they did.
2. We have longevity data for males and females of each species, where they did not.
3. We have their papers to use as references and examples of what to do (and what not to do) and they don't.

My suggestion to you is the same as what I suggested to LZ. I don't know if it is the best approach, but it worked for LZ.

Keep up the good work.

Best,
Dan
I wonder if my students know I make this stuff up as I go along?

1 comment:

jte said...

They'll never guess, and I'm not telling.