Best quantitative method for deciding?

[bicsy]

So of course, I'm having a difficult time deciding between my programs. My qualitative experiences at my interview visits didn't make one school stand out above the rest, so I want to use a ranking system with different categories to help make my decision. 

For those who are doing/have done something similar, what method did you use? (e.g. weighted categorical rankings/scores, analytical hierarchy process)

Also, what categories are you prioritizing? (e.g. stipend, # of PI's, etc.)

[rising_star]

I don't know that there is any "best" quantitative method. As you note, people prioritize different things, which leads to a multitude of ways of ranking programs. You really need to decide for yourself what the most important factors are and then develop a personalizing ranking system from there. Personally, I put the most weight on the PI I would be working with (not in the lab sciences so there were no rotations), the stipend, departmental atmosphere (collegiality, whether I wanted to hang out with the grad students, etc.), fees and health insurance, and location (weather, cost of trips home [flights or driving]), in about that order.

I actually didn't care much about job placement or publications because, at the time, I wasn't all that concerned about an academic career. I was actually more concerned about getting dissertation research funding and the PI's success rate with those. I don't know if this is helpful though, since we're in different fields. 

[svent]

If you're gonna go that route, go by COL/stipend I guess. Depends if PhD or Masters too.

Actually I think it's also a good idea to consider going by where you feel more welcomed. Big difference between getting accepted in January and having them want you to come vs. being a backup choice and getting accepted 4/15.

For me after I got my first few acceptances, I probably spent a good week going back and forth, one day thinking I definitely want to go to school A, the next thinking I definitely want to go to school B, etc. The past couple weeks I've been pretty much dead set in my mind on a certain school. Of course I still have some other schools to hear back from.

[juilletmercredi]

I love statistics and data, but I really don't think that this is a decision you can make with an easy quantitative method. I think you'd put more time into designing the method and coding it than you would actually making the decision.

How many programs are you choosing between? You've got some time. I'd make a new decision every 2-3 days and sit with it. How does it make you feel?

[TakeruK]

I agree with juilletmercredi---I don't think there is a good quantitative metric and honestly, I don't think there should be! But it's a pretty personal decision.

My personal reasoning is that the decision has to be holistic for me. I can't have an quantitative metric because these metrics could lead to a school with extremes winning over something more balanced. That is, for many factors, to me, it's just "good enough" or "not good enough". For example, the financial factor is either "I can afford to live the life I want" or not. Beyond that, it's not a lot of money difference between offers, not enough to sway a decision by itself anyways.

Also, I think the most important factors are things that cannot be quantified. For me, the most important factors (equally important):
- Environment: Do I feel welcome there? Are people friendly? Are the faculty supportive of students? Do the department do things that specifically encourage or discourage competition?
- Potential advising style: Would I get along with the faculty member(s) I would want to work with?
- Access to necessary resources: Can I get the telescope time I need? Can I get funding to do what I want? Will I be able to travel and network? 
- Location (local): Does the city have the ethnic foods I want? Do I feel like I can be part of this city? Can I afford to live there? Is the weather nice? Can my spouse find work?
- Location (big picture): What's the time zone difference between me and my family? Is it easy for my spouse and I to visit home (e.g. how many flights, do we live near a hub airport etc.)
- Stipend: Is it enough to meet basic needs and have a bit leftover for savings?
 

I think these were the most important factors to me, and I didn't score them or anything. It was just yes/no and I went with the school that made me the most excited and the one where I have the most positive thoughts when I imagine myself being there. I did make one spreadsheet to help me decide though, but that's just for the stipend factor. I took the base stipend, subtracted out the major costs for a grad student that might vary from offer to offer, which is mostly just rent, student fees, and health insurance. Then I looked at what's leftover and make sure I can afford the other necessities that don't really change much in price from state to state** (e.g. groceries, phone, internet etc.)

** I know there are actually differences, but as a grad student, we mostly need to support ourselves or ourselves and another person. Grocery store prices vary across the nation but in the continental US, when I was making my spreadsheet, I didn't notice significant differences in estimated annual groceries bills due to changes in COL (i.e. less than $1000/year differences). I think COL is a good coarse indicator but I wouldn't look too much into it because I think a lot of things that do not affect grad students are factored into COL (e.g. feeding a larger family, owning property, etc.)

[svent]

There are significant COL differences. If you're renting a crappy little room in California, you can get your own place in the midwest. Utilities are more expensive in CA. You might need to commute to save money on rent in CA, which makes gas and car maintenance very expensive. In the midwest, you might even get free gas from grocery points since you can probably afford to live close to campus. Food is much more expensive in CA (maybe $8.99 for a lb of ground beef instead of $6.99, or $6.49 for a lb of chicken instead of $3.49). Now if you're comparing schools in Arizona vs. Michigan, it might not matter a whole lot.

I don't see why people don't understand that, say, a 10% increase in income makes the biggest difference at low income levels. Sure, getting a raise from $100k to $110k is nice, but it's a luxury. Increasing your income from $20k to $22k is HUGE. Never underestimate the impact living in squalor for your entire 20s may have on your mental health. I know this is a forum about grad school, and you're expected to live very frugally, and money isn't everything, but it's very reasonable to think about your mental health and your quality of life, especially since PhD dropout rates are pretty high and there's no guarantee of finishing (or even if you do, of landing your "dream job"). A 1% increase in income may not be worth changing your decision over, but 10% is quite significant.

[TakeruK]

6 minutes ago, svent said:

There are significant COL differences. If you're renting a crappy little room in California, you can get your own place in the midwest. Utilities are more expensive in CA. You might need to commute to save money on rent in CA, which makes gas and car maintenance very expensive. In the midwest, you might even get free gas from grocery points since you can probably afford to live close to campus. Food is much more expensive in CA (maybe $8.99 for a lb of ground beef instead of $6.99, or $6.49 for a lb of chicken instead of $3.49). Now if you're comparing schools in Arizona vs. Michigan, it might not matter a whole lot.

I don't see why people don't understand that, say, a 10% increase in income makes the biggest difference at low income levels. Sure, getting a raise from $100k to $110k is nice, but it's a luxury. Increasing your income from $20k to $22k is HUGE. Never underestimate the impact living in squalor for your entire 20s may have on your mental health. I know this is a forum about grad school, and you're expected to live very frugally, and money isn't everything, but it's very reasonable to think about your mental health and your quality of life, especially since PhD dropout rates are pretty high and there's no guarantee of finishing (or even if you do, of landing your "dream job"). A 1% increase in income may not be worth changing your decision over, but 10% is quite significant.

Sorry if this was not clear---I do completely agree with you that COL differences are important. What I meant is that I do not think the metric of "Take your stipend and divide by COL" is a good metric, because COL factors in a lot of things we may not care about. Also, I think a lot of the COL are calibrated for something like a family of four, which have different rent costs than what you may want. (i.e. if you are interested/willing to share a house with some people, it's a lot different than renting or owning a house or condo).

Therefore, I find it's much more useful to take your stipend and subtract out the actual cost of rent you'll typically pay. If you are going to be living on campus subsidized housing then you should be using that rent metric, not the typical cost of rent in the area and not the COL. For example, right now, spouse and I are paying half as much as market rate for our 2 bedroom apartment because we are living off campus on subsidized housing by the school. The difference is over $10,000 per year saved. It would be a bad idea for students interested in my school to be using COL estimates to determine if they can afford it because the COL estimate will say it's not affordable but the school specifically have programs that help make things affordable. However, school subsidized housing does come with some restrictions (e.g. no pets) so those who prefer to not have these restrictions would need to determine how much the market rent would be. That is, this is why I would say it's more important to find out your estimated rent costs based on the life you want to live instead of a COL number that rolls everything into one and may not represent your actual needs.

For food costs, I guess it really depends on what and how much you eat. I've lived in several places now and my grocery costs does not fluctuate very much. But maybe I have been comparing places in high cost of living areas so it's been expensive everywhere. I live in CA and I do not pay the prices you are listing for ground beef and chicken (although I know you are just providing an example). But my point is that if you are buying groceries for 4 people, COL is a lot more important than just buying for 1 or 2. e.g. I typically spend $45-$50 per week on groceries to feed 2 people for a week. In the last place I lived (much lower COL), we spent $40/week typically. $10/week is $520 per year, which is not nothing, but also within the uncertainties of estimation anyways. (i.e. when I estimated my costs for each city, I expected to be wrong by about $1000-$1500 due to the fact that these are estimates).

And I also agree that a 10% increase is huge at $20k vs $22k level. Sadly, where I live, both of these numbers are not affordable, in my opinion. This is why the "stipend" factor, in my own decision making algorithm, is either "meets needs" or "does not meet needs". For the schools that did not meet needs, I asked if there was a way to apply for more money and when they said sorry, there isn't, I thanked them and declined the offer. I completely agree with you that it's very mentally straining to always worry about money---I know from experience. This is why the stipend was one of my most important factors and to me, the definition of "meets needs" is basically 1) I can pay all of my bills, 2) I never have to worry if I will be able to pay all of my bills and 3) If I want to go and buy a treat (e.g. a coffee) this afternoon, I can do so without worrying whether I would still have enough left for rent or groceries. Sure, I will still have to think about financial stability for things like saving for a home, for starting a family etc. but I need to not have to worry about day-to-day or month-to-month financial needs.

In my opinion, beyond this minimum, anything else is just a luxury because realistically, most schools do not offer much more than a few thousand dollars above this minimum. Where I live, I think the minimum stipend for surviving is about $28,000 per year and we're currently paid $31,000 per year. However, if there was another school that offered $33,000 per year in an equivalent cost of living area, I would not favour the second school more in my metric, since both schools meets the need for me. I would pretty much call the two schools a tie on the stipend metric and use other metrics to decide. This is because the other factors, which also affect my productivity and future earning potential is more valuable to me than the immediate $10,000 ($2,000/year for a 5 year degree). But this is a personal decision I made for myself---I am not saying that everyone must adopt this mentality. 

On the other hand, if the school offered only $27,000 where $28,000 is needed, then that $2000 suddenly becomes very important. I would certainly rank a school offering $29,000 a lot higher than a school offering $27,000.

[svent]

Those numbers are what I saw when I was in CA (different cities in CA, so I assume the numbers are pretty standard -- though I'm sure you can do better if you buy in bulk). I don't know how you can feed 2 people on $50/week unless you're eating Ramen every day. Maybe if you weigh 100 lbs and don't ever eat out (meaning stuff like going to Subway or some other place on campus for lunch).

[TakeruK]

1 hour ago, svent said:

Those numbers are what I saw when I was in CA (different cities in CA, so I assume the numbers are pretty standard -- though I'm sure you can do better if you buy in bulk). I don't know how you can feed 2 people on $50/week unless you're eating Ramen every day. Maybe if you weigh 100 lbs and don't ever eat out (meaning stuff like going to Subway or some other place on campus for lunch).

I don't include eating out in grocery costs because "grocery", to me, only includes home cooked meals. I cook meals for the two of us 5 days a week and we eat out 2 days a week. We used to eat out a lot less but now life is a lot busier and we have more disposable income, so we eat out more to save time and energy. Anyways, on average, cooking costs about $5/person/day (or just over $1.50 per meal but breakfast is a lot less and dinner is often more), so 5 days of cooking * 2 people is about $50/week. 

Those meat prices you saw are really expensive! We don't eat/buy a lot of meat and I do buy in bulk to save money. I buy about 6 pounds of ground beef every 4-6 months from Costco and it costs about $20, so almost 1/3 of the price you saw. In the last year, I have purchased maybe 12 pounds of chicken breast in total (about $24 at Costco). Typically, if I only buy 1 pound of ground beef at a time, it costs between $5/lb to $7/lb depending on how lean the meat is. But even if you don't go to Costco, the idea is that grocery stores have rotating sales so you only buy the "loss leaders" each week plus essentials. I've noticed that this year, some prospective students were interested in things like grocery store prices and they were noting things like prices on storefronts and flyers, but these prices fluctuate a lot over the year. For example, the week they were here, asparagus was $2/lb. But I've seen it go to $5/lb on some weeks. 

I don't include eating out because the cost can range a lot. Sometimes if we don't want to cook, we just pick up sandwiches ($4 each) at a nearby shop and sometimes we go to an actual restaurant ($15-$20/person). But yes, you are right---I should amend my "minimal" grocery number to be something more realistic. It would be more like $70-$80 for two if I am cooking all the meals (and in reality, we probably spend closer to $100-$110 including the eating out).

The example numbers I gave are to just compare grocery store prices, not to actually say "you can survive on this much". Also, my spouse and I both work and as I said above, we do make more than the minimum needed to survive, so we are choosing to spend the money on nice things, like eating out and buying nicer ingredients at the store.

But I feel like we are drifting a little off topic (sorry to others reading this)! I love talking about my budget / cooking / saving money but really the point of my post was to say that to me, stipend is only a deciding factor on my school decision if it was not enough to live comfortably. And, I am thinking factors like rent (which can vary by hundreds of dollars per month from place to place) are much more important to think about than factors like grocery store prices (which might vary by about ten dollars per month from place to place).

[pterosaur]

I'm always surprised by people thinking $25/week is crazy. Through high school, my parents budgeted $100/week for groceries for our family of 4. We shopped sales, bought house brands, and always cooked meals from scratch. I average about $28/week on food for myself now, including the occasional meal out.

[svent]

Food is very expensive. Even if you cook everything, healthy food ain't cheap. Especially if you like to workout and prioritize eating non-trivial amounts of protein. Sure it may be possible if you eat nothing but dollar menu items at McDonald's. But then the medical bills will catch up to you. There is no free lunch.

[rising_star]

On 3/2/2016 at 9:15 PM, svent said:

Food is very expensive. Even if you cook everything, healthy food ain't cheap. Especially if you like to workout and prioritize eating non-trivial amounts of protein. Sure it may be possible if you eat nothing but dollar menu items at McDonald's. But then the medical bills will catch up to you. There is no free lunch.

I don't think that's a fair characterization. When I lived in a place with good access to fresh produce, $10 would get me an entire reusable grocery bag full of fresh fruits and vegetables (often more than I could eat in a week). By shopping at Costco and the bulk foods section of stores, I was able to supplement with healthy grains (quinoa, whole wheat pasta) and canned and dried beans. In the winter, I'd often buy more frozen vegetables (store brand goes on sale for $1/lb regularly). I ate well by cooking for myself (making my own breakfast, lunch, and dinner with the help of the slow cooker, lots of varieties of soups for lunch, and a lot of oatmeal and eggs for breakfast), not buying meat very often (only when on sale as a loss leader!), and being willing to eat the same thing with slight variations on a regular basis. I could count on two hands the number of times I went out for "dollar menu items" in grad school. It just wasn't a thing. If I wanted a quick meal, I always kept some Lean Cuisine or Healthy Choice frozen dinners on hand, just as a back up. 

[svent]

You ignored the part about protein. Even if you go for cheap $3.49/lb chicken, that adds up. There's a reason poor people have more health problems caused by poor diets. Healthy food is expensive, a bag of candy or a happy meal is dirt cheap.

[rising_star]

1 hour ago, svent said:

You ignored the part about protein. Even if you go for cheap $3.49/lb chicken, that adds up. There's a reason poor people have more health problems caused by poor diets. Healthy food is expensive, a bag of candy or a happy meal is dirt cheap.

I didn't ignore the part about protein! I said I bought and used a lot of dried and canned beans, which are a huge source of protein. I'm mostly vegetarian so buying meat isn't a big part of my budget or lifestyle. Even so, $3.49/lb is the price of organic chicken on sale, which isn't what I'd label cheap chicken. Where I live, chicken leg quarters (so drumstick and thigh) regularly go on sale for $0.79/lb. If you were to buy a package of those with 3-4 leg quarters, you'd have enough meat for at least 6 meals for about $3.50. 

[svent]

Chicken breasts are where to go for an efficient source of protein. Are you really defending $100/month as a reasonable food budget? What do you do if you're a 220 lb bodybuilder eating 200+ grams of protein per day? I'm not talking about the food labels that base everything on a diet eating 50g/day, which is absurdly low.

Yesterday I went to the store and spent about $25 on food that'll last me about 4 meals plus a few snacks. $400 is a reasonable food budget, not $100. Sure, it can be totally doable to go under $400, but I sure hope you guys don't actually have a straight face on when you say eating on $100 is no big deal.

[rising_star]

@svent, you and I are clearly different people. No one asked about the diet for a 220 lb bodybuilder, so I assumed we were talking about people eating an average diet. You keep saying that the only way to survive on a $100/month grocery budget is to eat unhealthy food and I'm point out that that isn't the case. I did it for a couple of years so I do know that it's possible. I'm also not 220 lbs or a bodybuilder, so there's no way I need 200+ grams of protein a day. That said, I'm not saying that anyone has to limit themselves to $100/month, just pointing out that it is possible to do so and eat healthily at the same time. I'm not really sure why you're so adamant that this is not the case. Even now that I'm working full time, I spend less than $400/month on food, eating out, and drinks with friends. 

[svent]

A bodybuilder's diet is the opposite extreme of a $100/month diet (they probably spend more like $1000/month). What I can say for sure, having grocery shopped in many different areas of the country, is that meat and produce are quite expensive, but when I get an occasional bag of candy or chips, it is VERY cheap. Same thing with fast food (which I don't eat myself). This is why poor people don't eat right.

I'm not 220 lbs either, nor am I a bodybuilder, but I do workout a lot, and I also eat a fair amount, including well over the ridiculous expectation of 50g/day of protein. I believe staying active (and consequently, eating more) is important for your mental health, which is extremely important in grad school. If your income is so bad that you're considering only spending $100/month on food, the place to save money is likely rent.

[rising_star]

49 minutes ago, svent said:

If your income is so bad that you're considering only spending $100/month on food, the place to save money is likely rent.

Did I say that my income was so bad that I could only spend $100/month on food? No. I actually budgeted for more than that but, after adding up my monthly bills, I realized I was spending $90-115/month on groceries except when shopping for special holiday meals.

[TakeruK]

Obviously, what each person needs (or wants) to eat and what each person considers a good food budget is a personal choice. I completely agree with you that unhealthy food is way cheaper than healthy food and that it's worth considering how what you eat affects your mental health when deciding on your personal budget.

But this is all very personal choices and I feel like you are saying if people don't eat like you, then they are somehow screwing up their body and happiness. I'm glad you found what works for you and it's good that you know what you need to budget for food so that you can maintain the life you want and be healthy and happy! 

However, that doesn't mean the same numbers that work for you would work for everyone else. As I said, different people have different food needs. Not everyone wants to work out a lot or eat a lot of protein to be happy. And, in some places, the cost of rent dwarfs the cost of food, so when considering whether or not your stipend is enough, rent is often the primary factor. 

[ashiepoo72]

I think it's very easy to have a decent diet on $100-150 a month. I buy dry beans, lentils and brown rice in bulk (spend $20 on a giant bag one month, have enough for 4-6). Go to Costco and spend $20 on 6-8 lbs of ground turkey, split and freeze it--this will last my daughter and I at least a month, usually 2. I buy tuna in bulk, but try not to eat a ton of it each week to mitigate the whole Mercury thing. Huge packs of chicken are stupid cheap at Costco too. There are plenty of websites that teach you how to freeze vegetables for cooking, so you can buy a huge pack of spinach, mushrooms or whatever and do that. Also, canned veggies in a pinch--yeah, they don't taste great, but you can get them super cheap. Baking your own bread is super inexpensive, really easy and much healthier. I generally keep tortillas and cheese in the house for quick, cheap and easy breakfasts. It's ridiculously easy to make your own tomato or white sauce, so there is a pasta dinner (oh, pasta in bulk is also cheap as hell). Am I eating like a wealthy organic food enthusiast? No. But my diet now is way better than when I had more disposable income and ate out all the time.

And if I'm super broke, yes, I totally eat ramen or macaroni and cheese and I don't feel ashamed about it. 

How and what people eat is a very personal choice.

OP--sorry your post got derailed and that I felt the need to contribute to its derailing.  If you care to see it, I actually made a spreadsheet that weighted different factors when I was making my decision, so you can PM me if you'd like a copy. You can change the weight of the factors depending on what you think is most important. I should say I ended up declining the program that topped my list after going to recruitment and getting a better funding offer/impression from somewhere else.

 

[bicsy]

Thanks everyone for your comments. I didn't expect this post to lead to discussion about food costs, but I definitely gained something useful from all your replies! 

I've decided to just write everything that I think is important about each program using pen and paper, along with any extra comments I have, and judge based on what I've written down. The first few replies about using a holistic approach without really letting any programs 'win' with a couple of extremes really made sense. 

I'd love for this post to stay alive so people can continue to add their own approaches to making their decisions, because it really is a daunting process. 

[knp]

When I was choosing between three schools for undergrad, I ended up doing something similar. I was torn between these last three, so I made a very extensive spreadsheet with two or three different "weighting" levels that I could attach to each factor. I hoped this effort would clarify my thinking and provide a rational answer for what I should do. So I set to work. At the end of the day, when I'd finished putting in all the factors I could think of, one school had pulled way ahead. I stared at the spreadsheet. Really, that one had pulled ahead? I stared some more. But I liked that one, which had come in second, better. "I could tinker with the weighting to make that one come out ahead," I thought. "Hm. Screw it!" And then I picked the one I actually liked best. And it was great.

[pterosaur]

I just finished all my visits, during which I took a lot of notes (about potential advisors and the programs in general) and made pros and cons lists. Now that I have all the information in hand, I'm planning to make a list of factors that are important to me, give them weights based on importance, then rank the schools in each category and see where I come out. Will I end up using the numerical end result to decide where to go? Probably not directly, but it'll help me think through what's important to me and what each program's strengths and weaknesses are.