Quant Courses in MPP to get Public Policy PhD?

[Ben414]

I'm not sure if I will eventually pursue a Public Policy PhD after a Masters, but I'd like to be able to do so if I changed my mind during the program. Assuming I have only one semester of calculus and several Masters-level courses (e.g. 2 statistics/methods courses, 2 microeconomics courses), I'd probably want to take Calc II and linear algebra before applying? Maybe Calc III as well? For those who are in or have graduated from an MPP/MPA, did you know of other students who were planning for a Public Policy PhD and, if so, how were they preparing for it?

[monocle]

Disclaimer: Not yet enrolled in an MPP program.

I spoke with an the director at a program I'm considering and her advice was roughly, "If you're considering a PhD, this isn't the program for you... but there are the occasional students who want to go that route and those who do should notify their advisors early so they can start to put together research portfolios and guide their coursework to prepare them."

Basically, once you get there you need to talk to the professors (i.e. the ones who have the public policy/economics phd's) and administrators who can help give you guidance.

My two cents: You should probably try to finish Calc II and III before the Master's anyways, since it will be helpful in the Master's quant courses..

[chocolatecheesecake]

Quick thoughts: it'd be a good idea to take your quantitative classes in an MPP/MPA program and do well in them, but you don't necessarily need more calc or more linear algebra. Plenty of people when coming into a public policy PhD program choose to focus in statistics, sociology, or political science, and you don't need as much hardcore econ. I've heard that analysis is actually pretty useful for econ-concentration public policy PhDs, but it depends on the program. Another piece of advice I've heard given to my MPP friends who are considering a PhD is to become a teaching assistant for quantitative courses if you want to do a PhD. 

[Ben414]

On 1/23/2016 at 0:09 PM, chocolatecheesecake said:

Quick thoughts: it'd be a good idea to take your quantitative classes in an MPP/MPA program and do well in them, but you don't necessarily need more calc or more linear algebra. Plenty of people when coming into a public policy PhD program choose to focus in statistics, sociology, or political science, and you don't need as much hardcore econ. I've heard that analysis is actually pretty useful for econ-concentration public policy PhDs, but it depends on the program. Another piece of advice I've heard given to my MPP friends who are considering a PhD is to become a teaching assistant for quantitative courses if you want to do a PhD. 

I would likely be pursuing a Political Science concentration with the PhD, but I had read that the field has become more quantitative-focused and it is now recommended to have 2 or 3 semesters of calculus and possibly a semester of linear algebra for admission to a Top 20. Have you heard differently? I would not mind being wrong on this matter.

[chocolatecheesecake]

23 hours ago, Ben414 said:

I would likely be pursuing a Political Science concentration with the PhD, but I had read that the field has become more quantitative-focused and it is now recommended to have 2 or 3 semesters of calculus and possibly a semester of linear algebra for admission to a Top 20. Have you heard differently? I would not mind being wrong on this matter.

The best thing to do is to go directly to PhD students at a few universities you are interested in, and get your information straight from the horse's mouth. You're not going to find too many political science or public policy PhDs on this board. 

[Ben414]

1 hour ago, chocolatecheesecake said:

The best thing to do is to go directly to PhD students at a few universities you are interested in, and get your information straight from the horse's mouth. You're not going to find too many political science or public policy PhDs on this board. 

I had already scheduled to meet with some professors, but I was curious if you had differently. Thanks for the help.

[publicaffairsstudent]

Core courses in quantitative methods at public policy and public affairs programs draw heavily on multivariable calculus and linear algebra. Many public policy programs now formally state the requirements, such as UNC Chapel Hill requiring a 1 year calculus sequence for admissions. In practice I think very few people are admitted without Calc 1-3 and Linear Algebra. 

Applicants should be comfortable with quantitative analysis. American culture has a collective math anxiety, but this condition is unbecoming of a researcher in a heavily quantitative field like public affairs and to a greater extent public policy. If you are considering math in terms of "do I need to" you probably aren't ready to be considering this path.

Edited February 11, 2016 by publicaffairsstudent

[WhatAmIDoingNow]

It depends. Most of the quant in an MPP program (statistics, multivariate regression, advanced modeling) are enough. Calculus and linear algebra are not necessary, I have never seen them used in policy or public affairs research.  The issue is you need to be doing research and connecting with a couple faculty on their research to get into a PhD program.

 

That being said, I do know of a couple policy programs that do require Calc I prior to application. But in talking with people at those programs, it is a filter for bringing in applicants that are able to learn math rather than needing to know calc. 

Edited February 11, 2016 by WhatAmIDoingNow

[publicaffairsstudent]

Calculus isn't used in statistical modeling, but core courses in quantitative methods at good public policy programs teach a foundation in mathematical statistics, and this theory requires calculus and linear algebra to calculate probability distributions, regression coefficients and other key concepts. In practice a computer will do this for you, but good programs teach the theory.

Edited February 11, 2016 by publicaffairsstudent

[Ben414]

12 hours ago, WhatAmIDoingNow said:

It depends. Most of the quant in an MPP program (statistics, multivariate regression, advanced modeling) are enough. Calculus and linear algebra are not necessary, I have never seen them used in policy or public affairs research.  The issue is you need to be doing research and connecting with a couple faculty on their research to get into a PhD program.

 

That being said, I do know of a couple policy programs that do require Calc I prior to application. But in talking with people at those programs, it is a filter for bringing in applicants that are able to learn math rather than needing to know calc. 

I had a chance to talk with some MPP admissions and faculty recently, and they corroborated what you said above. Statistical software can do the math you need, and you just need to know how to use it. The faculty member who I talked to had a PhD in public policy, and he said calculus and linear algebra are not necessary for many PhD's in public policy. It would be if your concentration is in economics and it could be depending on your sub-field within other fields, but there are a lot of areas that don't.

The school is a well-regarded R1 institution and top 10 school in public policy, so they definitely disagree with what publicaffairstsudent is claiming.

Edited February 12, 2016 by Ben414

[Ben414]

34 minutes ago, publicaffairsstudent said:

It's your loss if you don't understand math. I certainly wouldn't trust your models. And I bet that prof would tell you he himself had taken those courses even if they were supposedly unnecessary. My advisor went to a top program and said most people there had completed all lower division math courses.

Actually I did ask him, and he said he didn't take those courses and doesn't use them. WhatAmIDoing is currently in a public policy PhD program, and he agrees with what he said. Most (maybe even all) of the MPP/MPA grads on here agree with what he said. You're the only one who's pretending you "need" Calculus 1-3 and Linear Algebra and that any modeling done by public policy PhD's without them should be discounted. You sound uninformed.

It's ironic that someone who "wouldn't trust my models" is displaying such poor formal logic. I certainly wouldn't trust your reasoning capability based on your showing here.

Edited February 12, 2016 by Ben414

[Ben414]

Based on your "Am I competitive" thread post, publicaffairsstudent, was your advisor from Cornell? Because they are specifically known for being extra quantitative-focused with their political science. If the advisor concentrated in economics or political science, that would explain why "most people there had completed all lower division math courses." It's not a great idea to extrapolate from one example to all public policy schools, especially when those with actual experience in other top public policy schools prove you are wrong.

As the faculty member I talked to said, Calculus 1-3 and Linear Algebra are necessary for certain quantitative applications. Some fields use these applications extensively, so you will need more advanced math for them. Other fields don't need them. To pretend that they are necessary regardless of the quantitative applications that will be needed is ludicrous.

Edited February 12, 2016 by Ben414

[publicaffairsstudent]

He didn't go to Cornell but he went to a top public affairs program. He said the quantitative methods core used extensive calculus and linear algebra, and most of his classmates had that background. 

As public policy is more heavily quantitative than public affairs, it will require more math. All good public policy programs have a two course sequence in microeconomics and a two course sequence in econometrics. You won't get through those without calculus. 

For example, UNC Chapel Hill which ranks about 15 requires calc 1 and 2 for admission, and states this on their website. Georgia state at #23 requires students without a quantitative background to complete a course that covers differential and integral calculus and matrix algebra before advancing to doctoral coursework. Higher ranked schools like Chicago, Harvard, Duke and Berkeley almost certainly have higher expectations.

So while a professor might say its unneccessary, in practice I imagine very few people earn admission without a decent quant background, unless they have some other unusual qualification.

Edited February 12, 2016 by publicaffairsstudent