Is Statistics MS a good background for research in government agencies?
May 23, 2014 1:43 PM Subscribe
Would Statistics masters make me competitive for research analyst jobs in state/county government agencies?
I am interested in becoming a research analyst for a local government agency, such as county department of social services. From my understanding, some of what they do is evaluating whether programs meet state/federal standards. Research analysts come from a variety of social science/psychology backgrounds, but they all know how to run a project, survey, and analyze data using Excel/SPSS.
My plan is to continue working while taking classes toward a Statistics Masters. I have read this question about statistics vs. social science background. Would a stats master degree make me competitive for research analyst positions? I know that there are PhDs in psychology/social science who are better qualified because they know research methods and have the whole experience of running a project themselves. So my concern is that a masters in stats is not project- or method-oriented enough. But I'm also pretty sure that I'm not going to go for a PhD, for financial and time reasons.
Would a Statistics MS be worth it for my goal?
I am interested in becoming a research analyst for a local government agency, such as county department of social services. From my understanding, some of what they do is evaluating whether programs meet state/federal standards. Research analysts come from a variety of social science/psychology backgrounds, but they all know how to run a project, survey, and analyze data using Excel/SPSS.
My plan is to continue working while taking classes toward a Statistics Masters. I have read this question about statistics vs. social science background. Would a stats master degree make me competitive for research analyst positions? I know that there are PhDs in psychology/social science who are better qualified because they know research methods and have the whole experience of running a project themselves. So my concern is that a masters in stats is not project- or method-oriented enough. But I'm also pretty sure that I'm not going to go for a PhD, for financial and time reasons.
Would a Statistics MS be worth it for my goal?
Once upon a time, I worked in operations research and management science, i.e. as a consultant to business. That work all but disappeared after the introduction of the personal computer. Financial analysts preferred doing the analysis themselves once they didn't have to rely on others to do the computer stuff.
So my experience suggests that you should have some subject matter education, not just research skills education. Besides, if someone has to go in a budget crunch, its going to be the ever dispensable generalist.
posted by SemiSalt at 6:22 PM on May 23, 2014
So my experience suggests that you should have some subject matter education, not just research skills education. Besides, if someone has to go in a budget crunch, its going to be the ever dispensable generalist.
posted by SemiSalt at 6:22 PM on May 23, 2014
A research analyst position is pretty generalist and requires a variety of skills. Straight stats will get you a foot in the door for some jobs, but social science will probably get you a foot in the door for more. Social science with a stats component will broaden your options further.
For context: social science is intended to teach you some ways that some people think the world works and provide you with tools to understand it. One of those tools is statistics. So focusing on social science gets you subject matter expertise. Focusing on statistics gives you technical expertise.
If you are a statistician, when you get brought in to projects all you will be doing is running data. That's fine if you love working with numbers, enjoy focusing on detail rather than the wider context, and like repetitive work. Don't expect variety or 'big picture' work.
Having said that, what you start out doing isn't necessarily what you will be doing in 5/10/20 years. There are so many examples of people excelling in government positions that aren't what they were originally trained in. So if you're just trying to get a job, and you see that there is demand in the labour market for government statistics work go for it.
Also - a lot of government based research isn't that sophisticated, so SPSS/Excel should be fine.
posted by bernardbeta at 1:26 AM on May 24, 2014 [1 favorite]
For context: social science is intended to teach you some ways that some people think the world works and provide you with tools to understand it. One of those tools is statistics. So focusing on social science gets you subject matter expertise. Focusing on statistics gives you technical expertise.
If you are a statistician, when you get brought in to projects all you will be doing is running data. That's fine if you love working with numbers, enjoy focusing on detail rather than the wider context, and like repetitive work. Don't expect variety or 'big picture' work.
Having said that, what you start out doing isn't necessarily what you will be doing in 5/10/20 years. There are so many examples of people excelling in government positions that aren't what they were originally trained in. So if you're just trying to get a job, and you see that there is demand in the labour market for government statistics work go for it.
Also - a lot of government based research isn't that sophisticated, so SPSS/Excel should be fine.
posted by bernardbeta at 1:26 AM on May 24, 2014 [1 favorite]
What is your undergraduate degree and work experience in? Knowing that might give us some ideas of how you can increase your education/ experience to leverage what you already have most effectively.
I work at a university-based research center where we get a lot of contracts to evaluate government programs. I only have a B.A. but had Ph.D. coursework and experience to get me into my position. Social science Ph.D.s are preferred because they look better on grant applications. My boss and many others I have had are happy to overlook that given good skills, but it can be difficult to get through bureaucracies that way... many job titles explicitly require Ph.D.s.
However, statistics can be a weakness of social science folks, so it's not necessarily a bad route.
posted by metasarah at 7:41 AM on May 24, 2014 [1 favorite]
I work at a university-based research center where we get a lot of contracts to evaluate government programs. I only have a B.A. but had Ph.D. coursework and experience to get me into my position. Social science Ph.D.s are preferred because they look better on grant applications. My boss and many others I have had are happy to overlook that given good skills, but it can be difficult to get through bureaucracies that way... many job titles explicitly require Ph.D.s.
However, statistics can be a weakness of social science folks, so it's not necessarily a bad route.
posted by metasarah at 7:41 AM on May 24, 2014 [1 favorite]
Response by poster: My undergrad is in international relations.
Would taking additional social sciences courses during my masters help?
posted by ichomp at 12:28 PM on May 24, 2014
Would taking additional social sciences courses during my masters help?
posted by ichomp at 12:28 PM on May 24, 2014
Research design classes would help the most if you're in part of a formal program.
A few suggestions/ possibilities:
Research coordinator jobs, which usually involve data collection (like conducting surveys), keeping track of follow ups, sending out mailings, etc., could be a good way into a higher level position. They do not require advanced degrees. Doing that while taking classes in data analysis may be more effective and efficient than going to school full time.
Looking at specific job listings, or better yet having an information interview at a place you want to work would be the absolute best way to figure out your best path.
If you go to grad school, find the social science research centers at the university and see if you can get an assistantship there. It will dramatically improve your attractiveness to future employers (or they may keep you on). Grad students are cheap labor which may make you attractive, especially if you have something specific you can offer them. For example, you're mosast likely to impress them if you look up some of their current projects and talk about how you can assist with them based on your current skill set or credentials.
posted by metasarah at 12:57 PM on May 24, 2014 [1 favorite]
A few suggestions/ possibilities:
Research coordinator jobs, which usually involve data collection (like conducting surveys), keeping track of follow ups, sending out mailings, etc., could be a good way into a higher level position. They do not require advanced degrees. Doing that while taking classes in data analysis may be more effective and efficient than going to school full time.
Looking at specific job listings, or better yet having an information interview at a place you want to work would be the absolute best way to figure out your best path.
If you go to grad school, find the social science research centers at the university and see if you can get an assistantship there. It will dramatically improve your attractiveness to future employers (or they may keep you on). Grad students are cheap labor which may make you attractive, especially if you have something specific you can offer them. For example, you're mosast likely to impress them if you look up some of their current projects and talk about how you can assist with them based on your current skill set or credentials.
posted by metasarah at 12:57 PM on May 24, 2014 [1 favorite]
I work in a research area of an Australian federal government department and have been responsible for the hiring of research officers.
Unless you find a job for something highly quantitative, I think a masters in statistics is probably going to be overkill. Very little policy research requires high level stats skills, and when it does it's most likely to be outsourced (usually to econometricians, unfortunately). You'll find most of your specific skills are wasted, I suspect.
In my own professional life I've never had to do anything more complex than a regression analysis, and you can get to that level of statistical sophistication by reading a book. These days I don't do much actual stats, but I need to know that the people who work for me are doing their stats properly and be able to explain it to my bosses.
Last time I was hiring research officers what I was really looking for was someone with a psychology degree. This was largely a proxy for having a bit of stats, a bit more methods (which is not just stats), and the ability to read research and write about it. In my experience, that skillset is much less likely from someone who has studied political science or sociology or IR or criminology (almost entirely for the methods stuff).
The most hardcore stats people in my area have a demography background. I actually don't know much about demography as a field of study, but our demographers are worth their weight in gold. That might be something worth exploring. I know our demographers have worked in health and in criminal justice policy areas, so it seems to be a pretty useful skillset.
Also in relation to having the experience to "run a project yourself", I don't think anyone comes out of a university course knowing that, so I wouldn't get too hung up about that as a criteria. I learnt more about doing research in the real world in six months of one job than I did doing a PhD. It's not somethg I'd specifically look for in a course.
posted by damonism at 3:48 AM on May 25, 2014 [1 favorite]
Unless you find a job for something highly quantitative, I think a masters in statistics is probably going to be overkill. Very little policy research requires high level stats skills, and when it does it's most likely to be outsourced (usually to econometricians, unfortunately). You'll find most of your specific skills are wasted, I suspect.
In my own professional life I've never had to do anything more complex than a regression analysis, and you can get to that level of statistical sophistication by reading a book. These days I don't do much actual stats, but I need to know that the people who work for me are doing their stats properly and be able to explain it to my bosses.
Last time I was hiring research officers what I was really looking for was someone with a psychology degree. This was largely a proxy for having a bit of stats, a bit more methods (which is not just stats), and the ability to read research and write about it. In my experience, that skillset is much less likely from someone who has studied political science or sociology or IR or criminology (almost entirely for the methods stuff).
The most hardcore stats people in my area have a demography background. I actually don't know much about demography as a field of study, but our demographers are worth their weight in gold. That might be something worth exploring. I know our demographers have worked in health and in criminal justice policy areas, so it seems to be a pretty useful skillset.
Also in relation to having the experience to "run a project yourself", I don't think anyone comes out of a university course knowing that, so I wouldn't get too hung up about that as a criteria. I learnt more about doing research in the real world in six months of one job than I did doing a PhD. It's not somethg I'd specifically look for in a course.
posted by damonism at 3:48 AM on May 25, 2014 [1 favorite]
« Older Positive radio: any UK radio stations that aren't... | What the hell are my upstairs neighbors up to? Newer »
This thread is closed to new comments.
Experience doing real data analysis in R or Python would probably get you more interviews if you don't already have them, and I've found applied graduate stats courses to be thin on the ground (but it could just be my school; the only one with a real programming element I've seen/heard of is a Bayesian methods course).
posted by supercres at 2:02 PM on May 23, 2014