Laptop for PhD in Epidemiology?

[sbeth]

I will be starting a PhD program in molecular epidemiology this coming fall, and am in need of a new computer. I forgot to look around the library during my visit to see what computers students there general used. I know that I will most likely be doing some computational work (computer modeling/statistics), and need a computer that will be able to support that. While I have enjoyed using apple products in the past, I am willing to switch to whatever laptop will be best. Any advice? 

[drivingthoughts]

@sbeth I don't know about your particular field, but when my wife was doing her graduate work, she did a lot of stats (SPSS, etc). Her one non-negotiable requirement was that whatever laptop she had must have a real & functional 10-key on the keyboard (she doesn't want to haul around a separate keyboard and mouse).  My wife still uses stats a whole lot in her field and is a master of excel - she uses a Dell Precision 17" because it has a giant screen and a real, full-size keyboard with number pad separate from the letters, and still costs less than the comparable Mac. I'm in the humanities and I love my Macbook so much, I've actually worn through some of the keys - but they don't come with a 10-key.

Good luck. 

[rising_star]

I also recommend having an actual numeric keypad. In addition, you'll probably want a PC because you'll want to be able to use SAS.

[time_consume_me]

ThinkPad P50

[Epi_2016]

If you want to use a Mac for mol epi, you'll have to either use a virtual machine program (more convenient, but I don't know what your comfort level or proficiency is with computers) or dual boot Windows to be able to use SAS, which will be absolutely necessary. I agree with a numeric keypad making things a lot easier, but I've gotten by with using something like this keypad. It's a not terribly bulky or inconvenient fix. So basically, a higher-end mac is totally doable if that's the brand you prefer, but you will have to make adjustments.

[themmases]

Macs are not uncommon in my program, so it's not impossible, but I'm not sure I'd recommend getting one. Maybe if you already own one you're happy with, wait a little bit and see if the barriers are acceptable to you.

SAS will be difficult to install on a Mac and many of my classmates have had delays at the start of the semester while they get everything they need installed. While you might not have to use SAS later if you don't want to, I found that a lot of my classes still only gave instructions for SAS programming of the material. Other programs I ended up using, including ArcGIS and SEER*Stat, were also difficult for my classmates to get working.

I'd second the recommendation that you need a numeric keypad. This is usually only available on bigger/wider models so keep an eye on the weight and battery life (so you can avoid carrying a charger). Most people in my epi program either carry a laptop that will allow them to run SAS/R/whatever, or don't carry anything (i.e. no one here is bringing a Chromebook just to take notes on).

The most common PC I see around my program is Asus and I have one too. Mine came in at 15.x", which was bigger than what I wanted, but fairly thin, light and affordable for its specs. The charger is a very reasonable size and weight. I have an hour commute on public transit so I bring it in when I know I will be coding in class and have a tablet for the other days.

[juilletmercredi]

Maybe I'm missing something here, but I don't get the recommendation for a numeric keypad. I did a lot of statistical analysis in public health (with large national datasets) on laptops without numeric keypads. If you're an epidemiologist using national/public data sets you aren't necessarily going to be doing a lot of data entry, and the actual analysis itself doesn't usually involve a lot of repetitive entry of numbers, at least in my experience. Plus, you can always get a USB keypad for cheap if you wanted. But I can't think of a single time when I was like "man I really wish I had a keypad." Maybe it's the molecular part?

I'd go with a Windows machine because using SAS in a virtual environment is a pain in the ass, in my opinion. I did this - I had a powerful Mac, 16 GB with a 1 TB Fusion drive. I dedicated half the RAM to my virtual machine. SAS still lagged.

Other than that the brand and type is kind of up to you. I schlepped my laptop everywhere and traveled a lot, so having a slim, portable machine was important to me. Right now I have an HP Spectre x360 for my personal laptop and I like it a lot, although I'm not sure I would want to use it as a work machine. My work laptop is a Lenovo ThinkPad X1 Carbon and it's an excellent work laptop. Super slim and portable, long battery life (I forget the charger all the time - I wouldn't say it actually has 10.9 hours unless you're not doing anything with it, but I would say around 6-8 is pretty accurate), quite durable (man I've dropped it so many times) and quite powerful. I think they start at around $1200.

ThinkPad has also introduced the P50, which is supposed to be for big data and other applications that require lots of power.