[GrayNess]

Currently I'm in a summer course for 2nd-year university statistics. The final exam is this week, I'm set for it so this thread is not about that. It's about something else: applying it and continuing with it.

Initially when I enrolled, everyone was telling me how stats is so boring, dull, etc... . I'm quite good at math so I thought it wouldn't be too hard. Now, I realize that I really like it. Whether or not other people do or don't is of no consequence to me.

But now I'm curious, for people who are in graduate school or are doctors or have some other form of higher education in any field, how often and how much do you use stats? I know some people on here have their PhD or PsyD and others are in graduate school or finishing university, so how much do you use it?

I'm thinking of minoring in it and adding that to my current double major biology and psychology, minor chemistry (I'm going to drop the minor chemistry no matter what). I've talked to my prof and she did say you need stats for graduate school and research (I've read many research and review papers so I'm aware it's needed) but just how much?

I'm not wanting to continue it because it's practical but rather because I enjoy it, to me it's fun to calculate whatever and interpret it for the situation. But ideally I'd like it to be of some use.

However, there is one bit I'm concerned about. I've heard that in upper years/levels, it can be tied in with computer science or programming. Despite me being a teenager and the whole idea of teenagers being little techies, I'm one of the worst ones when it comes to computers. We had a little bit of computer programming in high school and to say the least, I was horrible in it. Perhaps it was because the teacher was rather clueless on it but even now, when I see my friends who are great in computer programming typing whatever babble and I ask them to teach me it, the train goes and I'm not even at the station to put it lightly.

One main reason why I hopped onto the stats aside from the fact that I needed it for my major, is because when I learn something new, I want to know as much of it as I can and may just ignore whatever I was doing before (not always the case). I want to know that after I've read, memorized and understood the stuff, that people use the higher levels of it for reasons other than to fascinate themselves with their intellect.

So hopefully someone respond and this thread doesn't go to waste.

[Christina86]

Hey GrayNess,

I don't have any info about graduate or post-graduate necessity of statistics, since I'm still in my undergraduate degree,

But in terms of computer programming,

2nd year stats for me was all about formulas.
3rd year stats (if I had taken it) involves using a computer program, called SPSS.
http://www.spss.com/ It does data mining and does the statistical analysis and formulas for you, you just have to be able to actually plug in the right information and be able to read the results! That takes a while I've been told. I think you can get a free download of it to check out if you want.

[sunrise]

Hi GrayNess, I have a PhD in the biological sciences. I use statistics quite a bit in my current project, which is an epidemiological study. I have had to learn more statistics on my own than I covered in my undergraduate or graduate studies, particularly in the area of biostatistics. Depending on what sort of research you end up doing, you will use more or less statistics. People in social sciences research seem to use quite a bit, e.g. sociology, psychology, education. In all of those fields, if you go on to get an advanced degree, you will have to take graduate level statistics courses. And yes, you will need to use a computer for data analysis. You don't necessarily have to do computer programming and create your own stats programs, but there are various statistical software packages you will probably use.

It is probably very possible for you to take several statistics courses now, as an undergrad, beyond what a bio/psych major would typically take. You could just take these courses without getting a Minor in the field. Then you would have the knowledge but without all the requirements of the minor and hoops to jump through. Then you could really take the stats courses that are of most interest to you, whereas if you do a minor, there are probably set courses you must take to fulfill the requirements. Many departments tend to offer their own statistics courses, rather than all of them being offered by the math or statistics departments. So if you are interested in Education, you can take stats courses from the Education department, and so get stats that may be more "interesting" to you rather than just pure math. Same with psychology. Or whatever field you are interested in.

Statistics can be very practical. I wish I had taken more statistics, but not too much--there is really no way of knowing in advance what statistical techniques you will need in your future research. So I do my statistics learning now, rather than years ago in college.

[GrayNess]

Ah yes, our textbook refers to SPSS quite a bit, however, we use a different computer software that came with our book called minitab version 14. That one also takes a while to figure out but it does come with a rather thick book only for the computer software, however, it can be confusing at times mainly because I click on a drop-down menu and see some babble and out of the babble, I find the 1 little button I need without really knowing what on earth the other stuff means.

As long as it's using pre-made computer software programs then I'm better off than having to go from scratch.

This course is a general stats course, although there also are biostats, psychstats, etc..., the general stats one was offered in the summer and it was at a nice days and times. However, many of the questions use data from published sources and such for psychology, biology and other random ones.

Right now I've been considering psychology and biology research for quite a while, probably something in neuroscience or neuropsychology but in a few years, who knows, it may be only psychology.

In our abnormal psychology course, almost all of our lectures had epidemiology in it, and a few were only epidemiology. So I'm also a bit interested in that. I'm curious about your study, I realize you may not be able to tell the details of it and I'm fine with that but I'm more curious from a statistical perspective. In our course we always assumed or showed that the data may be approx. normal distribution (also has uniform, standard normal, etc...), but for a real study, do you assume or try to show it's normal or do you go for non-normal? Basically all we got up to was calculations for 2 samples (mostly independent, a bit of dependent) for mean, sample size, variance, etc... .

But you're right, I could just take the stats courses and don't do a minor in, that is a possibility also. I'm trying to take many psych and bio courses so I figured it'd be best to do a stats course offered by the stats department so I can generalize it later to either psych or bio. I'm also quite a big fan of math so I would like it to be as much pure math as possible but also learning the various theoroms and such.

[sunrise]

Originally Posted by GrayNess

I'm curious about your study, I realize you may not be able to tell the details of it and I'm fine with that but I'm more curious from a statistical perspective.

It's a case control study. We have matched sets of cases and controls. Cases have the outcome and controls do not. We look among the research subjects for various exposures and apply statistical calculations to see if cases have/had a certain exposure more frequently than controls (exposure therefore equals a risk factor) or less frequently than controls (exposure therefore equals a protective factor). The statistical measures we use are the odds ratio and the confidence interval.

[GrayNess]

Originally Posted by sunrise

It's a case control study. We have matched sets of cases and controls. Cases have the outcome and controls do not. We look among the research subjects for various exposures and apply statistical calculations to see if cases have/had a certain exposure more frequently than controls (exposure therefore equals a risk factor) or less frequently than controls (exposure therefore equals a protective factor). The statistical measures we use are the odds ratio and the confidence interval.

Confidence intervals we have covered, although I'm sure there is plenty more of it. I'm not sure if we covered the odds ratio. I don't remember and don't see that in the part of the book we covered (another course which is the continuation uses the same book and all so we may learn more of it in there). But that course is offered in the normal course time.

[Gayle7]

I Looooove stats! Go figure, since I'm not a techie nor a math whiz, either. I'm in a grad program, and after my fifth stats class I still love it. However, I'm glad I didn't pursue a quantitative psych grad program because, like you, I have other interests like psychopharmacology, neuropsychology and forensics. When I chose my academic path, the biggest deciding factor was - what do I want to do with this when it's over? Taking the stats path really meant research, and while I love it, I knew I didn't want that to be my primary job role. You can take the classes which will support you in doing your own thesis or dissertation project, or allow you to get the most out of reading others' research with a critical eye without focusing on stats so much that it dilutes your attention to other areas like biopsych. Good luck and happy SPSS'ing!

[GrayNess]

Originally Posted by Gayle7

I Looooove stats! Go figure, since I'm not a techie nor a math whiz, either. I'm in a grad program, and after my fifth stats class I still love it. However, I'm glad I didn't pursue a quantitative psych grad program because, like you, I have other interests like psychopharmacology, neuropsychology and forensics. When I chose my academic path, the biggest deciding factor was - what do I want to do with this when it's over? Taking the stats path really meant research, and while I love it, I knew I didn't want that to be my primary job role. You can take the classes which will support you in doing your own thesis or dissertation project, or allow you to get the most out of reading others' research with a critical eye without focusing on stats so much that it dilutes your attention to other areas like biopsych. Good luck and happy SPSS'ing!

What do the more advanced stats courses cover? My prof mentioned (before the course ended, still waiting for the mark), that calculus can be involved. I find calculus to be pretty fun, although I'm not a fan of using it simply for derivations. So, does the more advanced stats courses need linear algebra, calculus (including multivariable), etc...? Also, do the advanced ones let you use a pre-made computer software or do you have to make one yourself?

I'm fine with either doing clinical work or research but not teaching.