Coursera

[mbrown0315]

Last March I decided that I was missing some critical courses to make my application competitive, so I took a statistics course and a microeconomics course through UCLA Extension and a Calculus II course through UC Berkeley Extension. These are not degree-granting institutions, but I did get credit and an official transcript (each class cost roughly $700 not including textbook).

If I end up having to apply again next year, I think I would want to take additional courses to improve my application, but I don't want to keep spending money on these courses, so I was thinking about using Coursera, which is free (at least for now). Most Coursera classes award a "certificate of completion" with, I think, some indication of your performance, but they do not grant credit.

Does anyone know how these classes are perceived? On the one hand, presenting records of these courses should reflect a certain level of initiative (especially if combined with credit-granting classes and an explanation of my budget constraint). On the other hand, I can see how some professors would consider it a joke.

[gilbertrollins]

I have no idea what Coursera is, and professors are likely not to either. So you're asking them to go and research the quality of the certificates, which I suspect they won't do.

Moreover, I usually hear about people reapplying after doing additional research in some regard, not just taking more courses. Also committees will be looking for advanced (200+ level) coursework. So introductory courses get discounted, and probably even further if taken through an unaccredited institution.

[mbrown0315]

Fair enough.

Regardless, you should definitely have a look at Coursera if you see yourself in academia.

[Willows]

Is there something utterly lacking in your applications? With a 3.9 GPA and considerably high GRE scores for the social sciences, it seems strange that you wouldn't be competitive anywhere. I could understand additional courses if you had a low GPA, but I'm not sure what you're compensating for by taking them.

[jacib]

If you are looking for brownie points for taking a course in classical theory or a course on globalization, that probably won't mean much from Cousera (or from an extension school).

However, if you are learning a specific skill, then I would guess knowing the skill trumps where you learned it from. If you took a course in, say, GIS and got a certificate for it, all the better, but I think that'd have about the same impact as "Through an independent research project (my writing sample), I taught myself a year's worth GIS. I am now able to do [these analyses]" in your personal statement. Okay, if you had a fancy certificate, maybe slightly more of an impact, but I think people care more that learned how to do, say, multi-level modeling than how you learned it or where.

Is there something utterly lacking in your applications? With a 3.9 GPA and considerably high GRE scores for the social sciences, it seems strange that you wouldn't be competitive anywhere. I could understand additional courses if you had a low GPA, but I'm not sure what you're compensating for by taking them.

GPA and GRE scores are a necessary but not sufficient part of your application. If you look at the results pages, you regularly see people with high schores rejected from top schools. I agree with you, OP probably shouldn't take classes to change their GPA. However, if they are learnign skills they need for their analysis, then they'll need to learn this stuff sooner or later. Demonstrating that you can and will learn this stuff shows something to a admissions committees. If you come in with a limited stats background and say you want to become a quant person, that'll probably take several years of course work. It happens exceedingly rarely at my school (though maybe it's more common at Michigan or Wisconsin, where they really emphasize that stuff and are better at teaching it in department). And if you've already got a stats background, all the better. One kid in my program taught himself basic social network analysis in the six or so months between being accepted and the start of the semester.

[Willows]

If you are looking for brownie points for taking a course in classical theory or a course on globalization, that probably won't mean much from Cousera (or from an extension school).

However, if you are learning a specific skill, then I would guess knowing the skill trumps where you learned it from. If you took a course in, say, GIS and got a certificate for it, all the better, but I think that'd have about the same impact as "Through an independent research project (my writing sample), I taught myself a year's worth GIS. I am now able to do [these analyses]" in your personal statement. Okay, if you had a fancy certificate, maybe slightly more of an impact, but I think people care more that learned how to do, say, multi-level modeling than how you learned it or where.

GPA and GRE scores are a necessary but not sufficient part of your application. If you look at the results pages, you regularly see people with high schores rejected from top schools. I agree with you, OP probably shouldn't take classes to change their GPA. However, if they are learnign skills they need for their analysis, then they'll need to learn this stuff sooner or later. Demonstrating that you can and will learn this stuff shows something to a admissions committees. If you come in with a limited stats background and say you want to become a quant person, that'll probably take several years of course work. It happens exceedingly rarely at my school (though maybe it's more common at Michigan or Wisconsin, where they really emphasize that stuff and are better at teaching it in department). And if you've already got a stats background, all the better. One kid in my program taught himself basic social network analysis in the six or so months between being accepted and the start of the semester.

Definitely. I just question the impact of taking non-credit courses (without even a grade?) in the hopes of it boosting your application when you have a 3.9 already with stats and microeconomics. It seems like that effort might be better spent on other methods, like developing independent research projects/RA's, which may have a greater impact on competitiveness than taking the above style of course. (Unless the OP already has a plethora of both, but this is unstated.) I don't see it as a bad thing; rather, I just wonder if there are better alternatives, if that makes sense.

Edited November 28, 2012 by Willows

[Edugy]

I think that taking any of the stats or math courses to fill any gaps in knowledge would be useful, but besides that, I don't think it should matter.

[gilbertrollins]

What are the modal math courses for sociology applicants? I'm thinking I'm quant heavy, but that might not be the case. I have four semesters of calculus, linear algebra, intro to proofs, econometrics, applied stats with calculus, and analysis. Is that heavy or average for people who want to do networks and multivariate estimation et. al.? Are there a good deal of math majors that transition into quant sociology? I keep up among the math-o's, but I don't compete with them (so for instance, not going to write graph theoretic proofs for network theory).

[gilbertrollins]

Also, what are the texts commonly used for grad level stat courses in soc programs? I want to get under the hood with stats so I can develop novel empirical techniques for textual analysis. Was thinking about doing the graduate econometrics sequence wherever I end up, but might miss estimation techniques more germane to sociological questions. Betting sociometrics and econometrics are very different courses, but maybe not.

Edited November 28, 2012 by econosocio

[AaronM]

im not sure about math/econ majors transitioning into sociology, but overall I think your pretty quant heavy. In my soc undergrad program we have to take a basic stats class, a data analysis class (pretty much a class on how to do analysis in stata/spss), and a regression analysis class (and/or two qualitative research classes, but we're focusing on quant here). I'm guessing here but I think that is pretty standard course-wise. I gained most of my quant experience while working as an RA and the book in my regression class was useless, so I can't really answer your question about texts. So to answer your question, I think your going to have more quant experience than most applicants, even those who want to do quant heavy work, but I'm not sure the transfer rates of math/econ/stats into sociology so I might be wrong.

[RefurbedScientist]

Most profs won't know what Coursera is, but some will. Mitch Duneier taught Coursera's intro to sociology course this Fall. It's one thing to say "I took an online introductory course"; it's entirely different to say "I took an online introductory course with Mitch Duneier". While in fact the quality of the course probably doesn't change much based on who teaches it, name-dropping signals that the course was serious, even if 60,000 people enrolled (which is not uncommon for Coursera). And you can assume that at least some people at the participating faculties (Mich, Princeton, Berkeley...) might be aware of it if their colleagues are participating.

I concur with the above post that Coursera's biggest payoff might be in methods, which are harder to just pick up from independent study and are more-or-less the same regardless of who's teaching. Coursera's social network analysis class this Fall was taught by Lada Adamic, a big name in SNA. The formal modeling class was taught by Scott E. Page, political scientist at UMich.

Point is, no Coursera class will get you out of program requirements. But, they do signal self-motivated learning. Perhaps more importantly, you can impress people with a little "Oh, network analysis? Yeah I did a little of that with Lada Adamic". If you have nothing but time on your hands, then I say go for it. There's nothing to lose. But I wouldn't sacrifice GRE study time for Coursera work.

[Willows]

im not sure about math/econ majors transitioning into sociology, but overall I think your pretty quant heavy. In my soc undergrad program we have to take a basic stats class, a data analysis class (pretty much a class on how to do analysis in stata/spss), and a regression analysis class (and/or two qualitative research classes, but we're focusing on quant here). I'm guessing here but I think that is pretty standard course-wise. I gained most of my quant experience while working as an RA and the book in my regression class was useless, so I can't really answer your question about texts. So to answer your question, I think your going to have more quant experience than most applicants, even those who want to do quant heavy work, but I'm not sure the transfer rates of math/econ/stats into sociology so I might be wrong.

This is essentially what my school required; algebra/cal->then elementary stats-> then stats for soc majors -> then data analysis -> then methods.

[gilbertrollins]

Scott Page is a very, very smart guy.

[magicunicorn]

I honestly don't think Coursera would add any benefit to your application and could seriously hurt your application - simply because the bulk of the review of your application will examine 1. your undergraduate record, 2. your statement of purpose, 3. recommendations, 4, masters program work, 5. extra-curriculars. Where would Coursera courses get included? If they wind up in your SOP, that is a huge waste of space. The SOP is an incredibly short and concise statement by academic standards - and throwing in a line about taking an online course (who cares who taught it) would seem to most adcoms completely unnecessary and almost seem as if you either didn't have anything more exciting to mention in your SOP, or wonder why you even needed to take make up classes (did you bomb your undergrad?). Also, the way the online applications work is that there is a place to upload each item and there is really no upload space for a "certificate of completion" from Coursera. Including it in your CV too just screams that you are trying to pad since your aren't confident in whats there. For the CV, the rule is concise and less (but high quality) is always always better than a CV that has more in it, but is bad or "filler" resume stuff.

Here is the benefit of those courses: you learn stuff about Sociology. You know more about theory and can incorporate these ideas into your writing and your SOP.. and you can learn more before you start your grad program so you can get a leg up on your classmates.. but just please save yourself a potential frown-face from an adcom and don't mention the coursera courses in your application.

[jacib]

If it's a general sociology course on theory, there's not a way to seemlessly include it (other than awkwardly trying to slip it onto the bottom of your CV). Which will look bad because every course will take up the same about of space as your university degree.

However, if you learned a skill in your online course (python, formal modeling), you could easily put it under "additional skills" on your CV and/or a line in your SOP that won't be too disruptive like, "Additionally, over the past severals months I have taken an online course on formal modeling taught by Scott Paige. It gave me the skills to ____ which will allow me to ____ in the future.