MS at UT-Dallas as pathway to PhD

[oldprogrammer]

Originally Posted by CalmLogic

I thought you were in a rush I wouldn't do it, myself, as I don't have a BS in CS and would rather just take grad CS courses (either as part of an MS or as a non-degree seeking student) and catch up on my own.

It's one thing to take prereq courses in undergrad algorithms and discrete math, but the other courses I see as relatively optional.

So, in other words, I would rather spend 3 years working on an MS trying to do brilliant research than completing a BSCS degree in ~1.5 years and then the usual, hectic pace to complete the usual grad courework before the dissertation.

I hear what you're saying and completely agree. I think the MS approach would be the most comfortable. But in purely pragmatic terms, which do you think would likely yield the most favorable PhD admissions option: BSCS from UT-Austin or MSCS from UT-Dallas? And I'm assuming that I would be just as involved in research, though I don't know what kind of options undergraduates get at UT-Austin.

[CalmLogic]

BSCS from UT-Austin or MSCS from UT-Dallas?

I would definitely think the MS with a thesis at UT Dallas. Not only is the standard for an MS thesis higher than the standard for undergrad projects/research, you definitely won't get as much attention from professors if you are just an undergrad student and the professors will only be less likely to co-author papers with you as an undergrad. (And it's not like professors have much free time, anyway.)

Regarding research experience being more important:

Originally Posted by MDK

MIT rejects very good students from Berkeley and Stanford. The key to get into MIT is your research experience, not your undergrad school (as some programs do). As long as you have a great research experience and you have demonstrated the ability to conduct original research, you are welcomed to MIT. Otherwise they'll reject you even if you were attending MIT.

http://www.urch.com/forums/graduate-...y-profile.html (Pls evaluate my profile)

Letters of recommendation >> Records of projects and/or research >> reputation of undergraduate institution + transcript >> GRE scores

Adapted from: JHU Department of Computer Science

Regarding research/dissertation potential being uncommon among undergrads (not just CS undergrads) and sometimes even lacking in PhD students:

Even after completing all the coursework, fulfilling the research requirements (languages, computer skills, etc), passing the comprehensive exam, and getting an approved dissertation proposal, 50% still drop-out.

A PhD is like no other education. It's just you, and you alone, struggling to finish that dissertation. Even when done, there's no guarantee it will be accepted. Nobody helps you. Nobody understands.

To go for a PHD or not - Page 2 - Distance Learning Discussion Forums

http://www.urch.com/forums/computer-...-deciding.html (please help me in deciding!!!)

[oldprogrammer]

As always, thanks for sharing your encyclopedic knowledge! All my reading had led me to the understanding that LORs and research experience trump all else, but it's nice to get another opinion that applies specifically to my situation. Thanks again!

[CalmLogic]

Since I've been out of school for so long, I really have no means of getting meaningful academic LORs.

BTW:
http://www.urch.com/forums/graduate-...-question.html (LORs - unique question)

[oldprogrammer]

Reading through one of the other current threads, there were some questions about taking the GRE-CS, so I was curious what your opinion was for my particular situation. Since UT-Dallas is a, albeit lowly, ranked program, would the GRE-CS be helpful for me to gain admission to Top 10-30 CS programs? I ask in those terms because I was under the impression that the GRE-CS is most helpful when AdComs are unfamiliar with the quality of previous coursework, and I'm not sure if UT-Dallas fits into that category or not. And I'm not sure if it would be worth it to take time away from research to focus on studying for it.

And lastly, I can't imagine it would help at all in my applications to Business/MIS programs, who seem to only look at the GRE-Q score.

[CalmLogic]

Even if you got your MS degree from Columbia, I think taking the GRECS could help if the score was high enough (ideally 90+ percentile), especially for a school where such a score would be significantly above average, e.g. GA Tech rather than Washington. And studying for the GRECS can help one prepare for PhD qualifying exams, e.g. UIC may still use the GRECS as their written PhD qualifying exam: http://www.urch.com/forums/gre-compu...ying-exam.html (The GRE CS can serve as a PhD Qualifying Exam)

But I wouldn't count on the GRECS to do you any favors since, unlike the GRE that offers PowerPrep software, you really have no idea of what score you will get until you get scored. People expecting 80+ percentile have walked away with 50th percentile. But it is a good way to spend a Saturday morning, and you could train yourself by taking the GRECS every time it was offered by canceling the scoring In other words, just don't make the mistake of seeing the GRECS as a way to justify going to a lower ranked department since getting a 90+ percentile on the GRECS, even after finishing graduate level coursework, is very uncertain, and the GRECS can be more comprehensive (but certainly less in depth) than a PhD qualifying exam since you don't choose which subjects the GRECS questions should come from.

So, as you say, focusing on research is a more predicable, reliable way to ensure PhD admission, and I don't think MIS/IS PhD programs are going to really value of the GRECS since they may not regularly see GRECS scores and, therefore, may not realize that 80+ percentile is more uncommon than it seems.

[CalmLogic]

And I'm not sure if it would be worth it to take time away from research to focus on studying for it.

To further your point, a dramatic example that features a low GRECS score with an almost unbelievably high number of publications, though I don't know if he reported the score:

I scored in the 41st percentile. But last week I received acceptance into UIUC for a computer science PhD. With financial assistance. ...I have close to 20 papers, my undergrad was from a US State University not known for great research, my undergrad GPA was above 3.9, my master's GPA (from the same college) was above a 3.9. ...Good scores can help you but bad scores are not the end. Sometimes it seems like the messages on this board indicate that only the very highest scores on the test can get into grad school. Stop believing that and believe in yourself. http://www.urch.com/forums/computer-...ant-grecs.html (Just how important are GRECS?)

[oldprogrammer]

Originally Posted by oldprogrammer

So I'm thinking of doing an MS in CS at UT-Dallas to get the relevant coursework, LORs, and research experience to get into a PhD program. Would this be a wise course of action?

As another option, what about an MS in Applied Math? Since I don't have a formal CS education, I would also take the 5 or 6 "leveling" CS courses, plus a course each in databases, AI, Machine Learning, and software development since those are my interests. I would also focus on doing research or writing papers with CS professors for the purpose of getting research experience and LORs. What do you think? I'm just trying to explore all the options. Should I just stick with the MS in CS?

[CalmLogic]

As another option, what about an MS in Applied Math?

May not be the most effective use of your time, at least if you would get less credits being transferred into a CS PhD program (which is not something I would know about). And, relative to an MS in CS, it would not instill confidence in an admissions committee that you could pass a CS PhD qualifying exam.

[oldprogrammer]

Originally Posted by CalmLogic

May not be the most effective use of your time, at least if you would get less credits being transferred into a CS PhD program (which is not something I would know about). And, relative to an MS in CS, it would not instill confidence in an admissions committee that you could pass a CS PhD qualifying exam.

Those are all good points, and ones I didn't even think about. Thank you again.