Help me get a C+, so I can never touch this godforsaken subject again!

November 2, 2007 10:49 PM

Stats Help Filter: Arrrgh! Help me pass my Intro Stats course with a C+

I made a 66% on the first half of the midterm and I think I can't have made more than a 50% on the second half of the midterm, and that's with luck and probably curve correction. Therefore the remaining 50% of my mark has to be something spectacular. Please, share the resources that helped you make the grade?

It's intro to Stats Psychology, if that helps, with a focus on things like Z scores, Regression, Correlation and Probability.

If I can't pass this course it'll scupper my GPA and doom me to math free majors like History, which while I've very good at, are not my cup of tea.

posted by Phalene to education (17 comments total)

Go talk to your professor. Go take advantage of the TA help sessions and the math tutoring that your university offers. Yes yes. (Do these things even if you get great, helpful advice here.)
posted by LobsterMitten at 10:54 PM on November 2, 2007

Extra info: Between a 66% and a 68% is a C+. That'll keep my GPA in honours and make the class count for use as credits towards my major.
posted by Phalene at 10:55 PM on November 2, 2007

We don't have the syllabus or the grading sheets, so giving us information on what sort of grade you need is not helpful.

Have you read the book? I mean REALLY read the book?

Start by sitting down with your midterm that you bombed, and figure out which topics you don't understand. Then, go to the relevant chapters, and red them, sentence by sentence, taking notes along the way. This should feel like it takes forever, because it does. It will, however, help you to really understand it.

Don't just read the examples, work them out along with the book. Then do all of the problems at the end of the chapter, especially those with solutions in the back of the book. Don't cheat and look at the answers before you do the problem. If you get the wrong answer, go back, work through the problem again, and figure out where you went wrong.

If you get stuck on one of these problems, ask a friend in the class, a TA, or the teacher during office hours. There's no shame in it - most teachers are glad to see that you're making the extra effort to really understand teh material.
posted by chrisamiller at 11:02 PM on November 2, 2007

When I tutored math students at a community college, I found there was one factor that vastly overshadowed all the rest on how much improvement someone showed. Dedicated study time. The ones who improved the most were those who showed up for tutoring regularly, studied on their own, and spent no less than three hours out of class studying for every hour in class.

You are not a "statistics person", obviously. You will spend a lot of time for relatively little results, but it's one class to finish before you can write it off barring any usage of statistics in research psychology.
posted by Saydur at 11:16 PM on November 2, 2007

Have you considered that maybe it would be best if you took it again?

Boy oh boy, do I sympathize. Stats and I have a long, turbulent history. The only reason I did well in my intro to stats course was because the last half was looking stuff up in tables. Mostly we were doing hypothesis testing, so all you had to do was figure out what type it was and then go to the back of the book.

I just don't have a mind for stats. If you put sometime inside an expectation, my brain shuts down. And it's not like I'm a light weight either: multivariable calc, vector calc, group theory, discrete, numerical methods, diff eq., fourier transforms, laplace transforms, discrete fourier transforms, z-transforms... I loved all of these things.

Eventually, however, stats won. I couldn't escape it and I got so depressed about it I had to withdraw from my major! But first I put myself through a semester of hell trying to figure out mean squared convergence/differentiation/integration, random processes, auto correlation functions, and power spectral density.

I kind of forget where I was going with this post. Point is, sit down and really try to figure this shit out now, because it's never going away.
posted by sbutler at 11:18 PM on November 2, 2007

Seconding chrisamiller - I was able to breeze through my previous math courses, but the grad course in probability theory I'm taking this year has been a challenge. I met with the professor and he recommended:

(1) Read the entire chapter before attempting any of the homework problems - instead of rushing through the homework at 2AM before the due date

(2) Use the examples as practice problems - cover up the explanation/proof, and try to work through it by yourself.

I've ended up having to take a lot more time out of the week to study, but it works well, and actually, I'm starting to enjoy sitting down and pretending to be a mathematician. You should try it too!
posted by pravit at 11:55 PM on November 2, 2007

I agree with all the people above who say stats is an unbelievable pain in the ass. I've ended up getting pretty cosy with some statistical concepts, but what helped me was taking machine learning classes where I could really see exactly how much power statistics gave me. This gave me the motivation to study a lot of proofs and derivations that I have to admit I'd kind of dismissed as sometimes a bit unnecessary and boring and fuzzy.

Perhaps along with your study time, you could talk to the professor about ways that this course can actually help you in your life. Then the next time you're struggling a correlation coefficient you can connect it back to a paper you read and didn't believe and think about all the shitty, deceptive statistics they're doing to make the numbers they published possible. If nothing else, understanding stats deeply will make you savvier than probably 98% of psychology students. Your professor, being a stats prof, will almost certainly have a few dozen choice examples for you to look at that may be persuasive.

And, yeah, on sbutler's last point, I spent most of my education starting in middle school until about my third year of college thinking that this, for the love of god, would be the last stats class I ever took. Finally I woke up and realized that as long as I'm taking classes that have applications to the real world - to any facet of the real world - I'll probably be running up against some goddamn statistics whether I want to or not. This is a truth you might want to get used to.
posted by crinklebat at 12:17 AM on November 3, 2007

Get your textbook and start by reading just the words in between the notation. Once I had understood the text look at the notation to see how that fitted in with the words and then do the examples and practice questions...work out exactly where you've gone wrong...if you can see where you've gone wrong but don't understand why ask somebody to explain it to you...
posted by koahiatamadl at 1:30 AM on November 3, 2007

Start the homework early. Make sure you do all of it, even if it's not being collected. Even consider doing extra problems that haven't been assigned! Make sure you are actually getting the right answers, if possible. (Are there answers to the odd-numbered exercises in the back of the book?) Make sure you really understand what you are doing and why.

Study with other people. Make sure you understand what they are doing and they understand what you are doing. Often, teaching someone else (i.e., helping someone else understand what's going on) can be helpful.

When you have questions about what's going on, ask them! Ask them in class, ask them during office hours, ask them by e-mail, but ask them! Don't worry (within reason, if you've been going to class and paying attention) about your questions being 'stupid'; the likelyhood is that someone else has the same question too. By asking questions---especially in class---you are actually helping the instructor do his or her job better.

Go back over the first exams and figure out where you went wrong. Re-do the exams from scratch and try to get them right. It is fairly likely that an understanding of the early stuff will help you to understand the later stuff. (Fair warning: I don't know the content of your statistics class, but this is true in almost all the other math courses I have taught.)

How much time are you spending? A good rule of thumb is try for 2 hours per week outside of class for every hour in class. And this doesn't mean watching TV while the stat book is open, or staring blankly at the pages, or reading the words but not really taking them in. This is two hours of carefully reading the text, going over your notes from class and making sure you understand them, doing homework problems, and talking to other people about the material.

In terms of grades, I don't know what kind of school you're at or how flexible the professor feels about grading, but if a student managed to ace the last exam and/or the final after tanking the first two exams, I would be really impressed and likely bump the student's grade. (However, you should understand that this rarely happens. As in, I've never had this happen in a class that I have taught, although I have heard of one student doing this in another class---statistics, as it happens.)
posted by leahwrenn at 6:18 AM on November 3, 2007

Of course go over your old tests, read the book, and do all of the problems you can get your hands on. In this situation, I would recommend ten hours of work outside of class because you're behind: these concepts build on one another, and you need to devote some time to relearning the basics.

Make an outline of the topics the teacher covered on the test. Then look at your syllabus, the homework problems, and your class notes. What patterns do you see? Did you do better on problems that were covered more thoroughly in class, or that you did more homework on? Did you do better on problems on the test that related to problems you worked out with friends, or worked out on your own? Could you have guessed, based on the homework and your notes and the syllabus, what was going to be on the test?

You can bet a significant number of students in the class have access to exams given previously by this same professor. Unless the professor has said "Materials from previous years are off limits," try to get your hands on some old exams and do them as practice. Then analyze as above.

When you complete a homework assignment, make a photocopy of it. Sometimes you won't get all homeworks graded and back to you by the test. You can go over them with a friend if you don't know how you did yet.

Finally, and most importantly:
Schaum's: Probability, and Elements of Statistics II. Buy them both. They will help you get back up-to-speed. Use them a lot before the week of the next test, and they will help. I heart Schaum's.
posted by Eringatang at 9:08 AM on November 3, 2007

Ten hours per week, I mean
posted by Eringatang at 9:09 AM on November 3, 2007

I was a stats TA in grad school, and I had to take several courses as a sociology undergrad (probably not that different from your psych class).

I'm a visual learner, so charts and graphs made more sense to me than paragraphs of text and equations. Try drawing the concepts in the book. If it uses an example with 40 people as your sample size, then draw 40 stick figures. 10 of them have condition X? Circle 10 of them. And so forth. I do best with concrete, tangible examples, not abstract equations. My guess is you're getting tripped up on OMG it's MATH!, when it's really not. The central question in regression is "in light of the information I have about the population, how likely is the result I just observed?" You use math to get there, but if you understand that concept, the math becomes just a tool. And really, SPSS or similar programs do it all for you anyway. You just have to know what you're looking at when you get the output.

Ask your TA or professor for help. The stats professor I worked for understood the abstract concepts so well that he had trouble finding concrete examples. If you don't understand what they're saying, rephrase it in a different way that is easy for you to understand. (Oh, you mean like as a person's smoking increases, the probability of cancer also rises?) Form a study group with other students. Find a room in the library with a whiteboard so you can draw the concepts and discuss.
posted by desjardins at 9:51 AM on November 3, 2007

Let me disagree with crinklebat, pravit, and sbutler. I do not think they are talking about the same subject as the one you are taking.

This is intro stats for an undergraduate psychology sequence, yes?

Assuming so, phalene is almost certainly not integrating anything, not dealing with autocorrelation, not dealing with random walks, and not dealing with proofs. Standard intro stats classes in a social science context simply don't cover these topics -- they are basic how-to-do-it and what-does-it-mean courses. So lay off with the "You're right to think this is really hard," because it shouldn't be. In all likelihood (no pun intended), the most math that phalene is being called on to perform is messy high school algebra.

So anyway, if you're bombing, something is amiss.
posted by ROU_Xenophobe at 10:01 AM on November 3, 2007

The advice part:

One nontrivial possibility is that your instructor sucks. What are your university's deadlines for dropping and withdrawal? If you can still get out of the class with no record or a simple W, check to see if someone else is offering the course next semester, and check around the local scuttlebutt to see if that person has a better teaching reputation.

Second, there is no substitute for lots of work. In particular, you cannot read a math-ish book in the same way that you would a psych text or a novel for a lit class. You must, must, must sit down and pore over it, slowly and in detail. Then do it again. Do this before class, each and every goddam session, so that you have some idea of what was a complete mystery to you. This will equip you to ask useful questions in class.

Third, work through every example on your own paper. Do additional problems, especially if your textbook has some answers in the back of the book so you can check your work.

Fourth, keep up. Do your problem sets early enough that when something gives you fits, you have time to stop by the professor's or TA's office and ask questions about it.

And use whatever tutoring and office hours and so on are available to you.
posted by ROU_Xenophobe at 10:13 AM on November 3, 2007

One nontrivial possibility is that your instructor sucks.

So true.

I've taken a couple of basic stats classes that I really liked. I'd emailed back and forth with a couple classmates and they claimed it helped - so if you have specific questions I'd be happy to give it a shot. Math is very cumulative, so my guess is that you missed just one or two basic concepts at the beginning. If you get those cleared up, the rest will not be that hard.
posted by selfmedicating at 7:26 AM on November 4, 2007

Oh, also: definitely go to office hours and talk to the prof. A friend who did very well in grad school once told me that he though he went up a full letter grade in every class where he went to office hours. I'd say come right out and say that you want to pass with a C+ so as not to tank your GPA, but feel like you are missing something, and want advice to help guide your studying. From his perspective, does it really matter if he gives you a C or C+? If you really want help, work hard, and actually improve between now and the final, the prof may be willing to bump you up an extra couple points, in case you don't quite make a C+. But it helps to lay the groundwork now.
posted by selfmedicating at 7:47 AM on November 4, 2007

B-!!!! BOOYAH!
posted by Phalene at 8:26 PM on January 9, 2008