RAing vs. taking Ph.D core courses

[pepper00]

Hi all,
If you are doing MS in Econ, planning to apply for the ECON Ph.D program this fall and
if you are offered a funded RA opportunity (beginnig this summer and continuing until next semester) for a project,
what do you think is the best course of action?

1. Take the RA job, and take Ph.D Micro sequence + metrics (not even sure if this is possible, considering the competition to get good grades in core sequence courses..)
2. Assist professor only during summer, Give up RA job during the fall semester, and focus on core Ph.D coursework (Micro + Macro sequence together+maybe metrics or math econ)

Also, I wonder how much impact the Ph.D Macro sequence course has on the application (when you assume that you take the course in addition to Micro sequence). Some people say it wouldn't have that much impact, since it's not standardized like Micro sequence, but the others say it's really essential to take Macro sequence and get a good grade, so I'm really confused...

Good luck for everybody preparing for fall 2010 admission!

[unitroot]

Most RA tasks assigned to undergraduates (and even beginning graduate students) are boring tasks like formating or manipulating data, and such. It's boring and takes a long time, plus you hardly learn anything new. I wouldn't continue it after the summer unless it's a truly interesting project.

[asquare]

Originally Posted by unitroot

Most RA tasks assigned to undergraduates (and even beginning graduate students) are boring tasks like formating or manipulating data, and such. It's boring and takes a long time, plus you hardly learn anything new. I wouldn't continue it after the summer unless it's a truly interesting project.

Learning how to prepare data for analysis is a really important skill, and one that is more efficiently learned through RA work than on your own time, on your first independent research project! Cleaning, reshaping, and merging data are non trivial and essential skills for anyone who intends to do empirical work. I think people who start graduate school directly from undergrad especially underestimate the importance of this skill and the prominent role it will play in their future time allocation. If you spend most of your time in an RA job preparing data, well, that's because you are highly likely to spend most of your time in our own research preparing data. Unless you are working with only single, well-established, well-cleaned, self-contained data for the remainder of your career, learning to prepare data is a great use of your time and will make it much easier to hit the ground running with your own research.

That said, a good boss will involve a capable RA in data analysis, preparing and writing up tables, and other parts of the research process as well. Of course, this has to be scaled to your training in econometrics and statistics.

Doing research, though, involves a number of skills that aren't taught at any point in the PhD process. IMO, the best way to learn how to do research is to immerse yourself in someone else's research by being an RA. Even if you wind up doing what unitroot describes as boring and time-consuming, you will be learning how to do things that are necessary for your own research! Boring and time-consuming doesn't mean obvious, and by doing RA work, you will be come much more adept and efficient at important skills you would otherwise have to learn on your own as you attempt your first research projects in grad school.

[LouisBD]

I would pick option 1. Micro and metrics + RA job => good combination.

[unitroot]

Asquare, I agree and disagree with you in some sense..

It certainly depends on what the project is. I have once been RA where the first 2-3 months were useful in terms of skills learned, but the rest of it was repetition or variations on the same old stuff, very time-consuming, boring, with little added value in terms of knowledge or compensation. The amount of satisfaction will depend on one's interests of course. If you want to be a financial economist but the RA job requires working with labor data, then the value added can be low because even if you learn many new things, not all of them will be useful. When it comes to learning data formating and analysis, I am generally a fan of learning things when I need to do them for myself. Maybe it's just my way of doing things. This was of course just a warning. Not all data work is like this, and not all RA work is data work.

[asquare]

Oh, I agree that not all RA work is useful, I just wanted to point out that boring data preparation isn't inherently useless -- in fact, basic skills in data prep are very important for anyone who intends to do empirical work. But looking for a good fit and checking that the previous RAs had good experiences and went on to jobs or schools you would be happy with is important. Not all RA jobs are worthwhile, but many are and the skills they teach might not be exciting on a day-to-day basis.

I also think that for a student in a good MA program with good grades and sufficient math preparation, RA experience could be more useful than the PhD core. I know I keep saying this, but having a PhD in economics is not (yet) a prerequisite for admission to a PhD in economics...

[notacolour]

I think I agree completely with asquare. RA work can be incredibly useful if you want to do anything applied, though of course it depends on the prof.

Moreover, I think it's kind of unreasonable to plan to take the entire PhD core before attending a PhD program. It is almost certainly harder than you think, especially coming from outside of the group of PhD students.

Taking a couple of PhD-level courses and doing well in them is a great signal, but I think that taking the entire core both increases the likelihood that your performance will be only mediocre (or worse) and fails to provide a better signal than just taking one or two courses.

[savingtheplanet]

RA + PhD Micro + PhD Metrics would be ideal, provided you feel you have the skills to get good grades in the phd classes.

and ra would be particularly valuable for an ms student, since a US-based master program usually won't focus on research, and is not a typical route into phd programs. a letter from the ra advisor would be good, especially if you can convince them to let you make some creative inputs to the project. (or at least impress them with your understanding of their project)