UChicago's new program: MA in Computational Social Science

[sixis423]

Hi everyone,

Does anybody know something about UChicago's MACSS program? It starts this year, but I think it seems even better than MAPSS.

But no one is talking about it in this forum. Is there anyone who applied to this program?

This is UChicago's ad for it: (http://macss.uchicago.edu/page/why-choose-our-ma-computational-social-science)

Why Choose our MA in Computational Social Science?

Our two-year MA in Computational Social Science will give you the ability to apply cutting edge technologies and statistical techniques to data sets that are orders of magnitude larger than one normally encounters as a social science investigator.

Our four-course core will introduce you to the fundamentals of large data analysis, modeling, and computer programming.

Our structured curriculum will develop the tools and applications necessary to make important contributions across the social sciences. Depending on your field, that could mean problems of memory formation, language use, knowledge networks, changing historical patterns in the distribution of material and cultural objects, the conditions under which social movements succeed or fail, or how partisan identification shapes social and economic behavior.

Our three-quarter "research commitment” will have you work directly with a member of our Computation faculty, collaborating and co-authoring on papers.

Our summer practicum will get you embedded within an academic or professional organization, allowing you to take your research and computational skills directly into the field.

Our weekly academic mentorship, beginning with an advanced doctoral student “preceptor,” and continuing through every level of our faculty and staff, will assure that you are maximally competitive for PhD study.

Our Director of Career Services will provide a full suite of weekly programming, including resume workshops, networking opportunities, mentors-in-residence, and on-campus recruitment.

 

 

[juilletmercredi]

I've just had a similar post. I'm looking to apply, but my biggest concern is, looking through their sample course tracks, that most/many courses are undergrad CS/MATH/STAT, which is good on one hand if you don't have the background (I don't). On the other, you are paying gradschool fees for undergrad courses (?!). I would love to hear what others think.

Many excellent departments do this for statistics and math courses (and probably for computer science courses, although I don't know for sure). In many statistics departments, for example, graduate classes are simply accelerated versions of undergrad classes. At some departments, they are cross-listed (so the undergrads register for STAT 4360 and the graduate students register for STAT 6360, but it's the exact same class at the same time in the same classroom with the same professor). And in other departments - like Columbia's, for example - they don't even bother to do this; the graduate students and undergrads BOTH register for STAT 4360. There are lots of top statistics departments in which first-year students are sitting in the same classroom as the advanced undergraduates. To be frank, there's not a whole lot separating a senior statistics major and a first-year master's student in statistics; there's even less separating a senior statistics major and a first-year master's student with a background in the social sciences who doesn't have a strong stats background. In fact, the edge goes to the undergrad.

One quick glance at Chicago's course catalog reveals them as a latter case: the graduate course listings in statistics go from 30000 to 45000 level; 30000-level classes are advanced undergraduate and lower-level graduate student classes. The bachelor's paper, which is typically the highest-numbered undergrad class, is numbered in the 29000 range.

As for the intro math stuff, like linear algebra and introductory programming - the other alternative is to simply require it as a prerequisite, which most MA programs in statistics do. They're clearly trying to lower the barrier to entry, so they just require you to take it in your first year instead and increase the overall credit requirement (many comparable MA programs require around 30 credits and this one sounds closer to 36).

The program is aimed at social scientists without the background in math/stats and computer science; the idea seems to bring them up to speed in the latter area to enable them to be data scientists who work with social data. You use the basic statistical and CS fundamentals to do the analysis and the expertise in the social sciences to make interpretations and draw insights that can be of use to organizations.

Also, co-authoring is the name of the game in the sciences - a co-authored paper with an established scientist is a good thing. My master's thesis was a co-authored paper that I later published in a scientific journal. That is BETTER than a single-authored thesis that never sees the outside of the stacks.

There's nothing about this program that raises a red flag for me. I think the only concern is putting the program in its proper context. The basic curriculum of the program won't make you a data scientist capable of competing for data science jobs at top tech firms; it doesn't have the level of CS or stats necessary to do that. The stated goal seems to be to turn out quantitative social scientists who can manipulate and analyze very large social data sets and make interpretations. Think more analyst at the Census Bureau or economist at the World Bank than data scientist at Google. Given the name of the program, I'd imagine most comers would already know that, though, and that's how they are marketing it.

I also just want to mention that a program being a 'cash cow' (i.e., a professional program mostly funded by loans or student self-pay, with little to no financial aid) doesn't necessarily make it an objectively bad program academically or professionally.

[Alexander II]

I've just had a similar post. I'm looking to apply, but my biggest concern is, looking through their sample course tracks, that most/many courses are undergrad CS/MATH/STAT, which is good on one hand if you don't have the background (I don't). On the other, you are paying gradschool fees for undergrad courses (?!). I would love to hear what others think. 

[GradSchoolTruther]

Another money grab.

[Alexander II]

@GradSchoolTruther Thanks for your input! Could you elaborate what are some of the red flags for you?

[GradSchoolTruther]

Most of the courses are undergraduate courses. The thesis is co-authored with an advisor, so calling it a thesis is not accurate. The university is awfully loose in what it considers a "graduate" degree to be.

[Alexander II]

15 hours ago, GradSchoolTruther said:

Most of the courses are undergraduate courses. The thesis is co-authored with an advisor, so calling it a thesis is not accurate. The university is awfully loose in what it considers a "graduate" degree to be.

@GradSchoolTruther Thanks! I was also concerned with similar things.

[Alexander II]

@juilletmercredi Wow, thanks for such an expanded insight. It really helps to contextualize this program in the gradschool environment. Appreciate you sharing on the whole upper undergrand vs low grad course levels and the proper target audience for this degree.

[Alexander II]

@BirdKiller I share your concerns. From I understand though, US non-PhD track masters in social sciences are generally quite expensive as you're expected to pay in full and little financial aid is available. Though unlike most masters in soc.sciences this one offers merit-based aid ranging from 1/3 to full ride from what I hear.