University Subjects

MAST10011: Experimental Design and Data Analysis

MAST10011: Experimental Design and Data Analysis

University
University of Melbourne
Subject Link
View Subject

Subject Reviews

Shenz0r

11 years ago

Assessment
One written assignment of up to 10 pages due in the second half of semester (5%), eight to twelve homework quizzes (a combination of written and oneline) due at regular intervals during semester (10%), one 45-minute written computer laboratory test held during semester (5%), and a 3-hour written examination in the examination period (80%).

The 3-hour exam is marked out of 100 even though there are 120 marks. So there is a 20-mark buffer if you want to get the full 80% from the exam.
Comments
Stats is the subject which many people complain as dry, boring. It can be at times, but the subject is also not very difficult if you pay attention and everything. Half of this subject is literally just finding the right formula to use and plugging in numbers. The questions are quite formulaic too, so it's unlikely that you're going to be tripped up for most questions.

Davide lectures for the Semester 2 block and he is just a very good bloke. Although the tutorial questions he uploaded were on the rather easy side, he made our exam a bit more difficult than what Ray would normally write. Which was a bit unexpected, but oh well.

You'll have weekly quizzes every week which are taken off the multiple choice of past exam questions (apparently). You'll have 3 attempts for each quiz and they'll take your highest mark for each one. With the help of friends you should be able to full-mark all of the quizzes. They cover the things that you've learnt throughout the week, so they're pretty good for consolidating a week's worth of knowledge.

Tutorials and computer labs are optional, but I went to both all the time anyway. The computer lab pretty much gets you playing with Minitab, which you'll have to use for your assignment a lot. I guess they also do consolidate the material a little as well, so they do help with your understanding. Tutorials are very useful. I recommend registering for Sharon Gunn's tutes because she explains the concepts in a little bit more depth and makes everything very clear. The tutorials just consist of working through a bunch of questions, and it's recommended that you do the questions beforehand and then ask around during the class.

The assignment was a bit hard for us, initially. But if you collaborate and read up on the EDDA notes you won't find most of it tremendously difficult. Ours just consisted of 6 questions, and only one of them required Minitab. However, with the computer lab test, you do want to make sure that you've gone through all the computer lab questions because if you don't know how to read off the Minitab outputs, then you are kinda screwed. It's an open-book test so you can bring in your lecture notes and EDDA textbook. If you've revised everything then the computer lab test shouldn't be too bad as well.

Coming closer to exam time, you want to make sure you know the summary sheet at the end of the EDDA textbook like the back of your hand. The summary sheet is very helpful and pretty much has all the formulas and notes that you need. You also need to get a lot of practise with reading statistical tables and graphs, because you will be using them a lot.

The first quarter of this subject isn't quite interesting, it's just experimental design and lots of probability distributions. But when you get to estimation, hypothesis-testing, comparative inference and regression, you need to definitely know which formulas to use and when to use them. That's all you really need to know; if you can do that then you will not have too much trouble with this section. Generally a lot of people will struggle with this part of the subject (since it's the most maths-related) but all I felt that I was doing was plugging in numbers into the right formulae. If you were good at probability and distributions in VCE Methods, you might not struggle so much with estimation and all.

The exam for 2013 was more difficult in comparison to 2011-2012. We weren't expecting Davide to chuck us a hard exam or anything so I guess we should've prepared for the worse. It might be a goooooood idea to vaguely remember some formulas. In our exam, there was a question which told you to use a rank test but the summary sheet provided in the exam didn't have the rank test formula (whereas it did have it in the textbook).

All in all, this subject, although people complain about it as boring, is not tremendously difficult. Considering you have a 20-mark buffer in the exam, you should be able to H1 if you've put in the work. That means going through the tutorial questions, doing all the problem/revision sets, and doing all the past exams. You'll see that memory work isn't really that much because 90% of what you need to know is provided on the summary sheet. It's easy to complain about this subject's difficulty if you think it's uninteresting and dull, but if you do enough questions and ask for help, then chances are... you'll see that this subject isn't actually as hard as you'll make it out to be. Most of it is just plugging in numbers.
Lectopia Enabled
Yes, with lecture capture.
Lecturer(s)
Davide Ferrari. This guy is a top bloke. Like, he's just nice as and is always there if you need help.
Past Exams Available
Yes, all previous years can be found in the exam library. Your EDDA textbook is littered with past exam questions too. Answers for 2011 and 2012 exams were uploaded onto the LMS.
Rating
3.75 out of 5
Textbook Recommendation
You must buy the EDDA textbook. It has all the questions in it with solutions, and stats is all about trying out more questions. Most of the lecture notes are actually copied straight out of the textbook as well.
Workload
3 x one hour lectures per week, 1 x one hour practice class per week, 1 x one hour computer laboratory class per week
Year & Semester Of Completion
2013, Semester 2
Your Mark / Grade
H1 (98)

Did you find this review helpful?

kaybee94

11 years ago

Assessment
3 hour end of semester exam 80%, weekly online quizzes 10%, assignment 5%, computer test 1 2%, computer test 2 3%
Comments
Other than maths for biomedicine this is the other subject that cops a lot of criticism in 1st year biomedicine course. The subject really isn’t that bad if you do concentrate and be attentive. I do admit that Ray has a tendency to stutter and stumble a lot and a lot of the lectures you sit there thinking what is he going on about but the most important thing is to sit down after lectures or maybe during the weekend and go through it again. Statistics is one of those things where you do need to look at it about 4 or 5 times before it becomes clear in your head. Ray does take the entire semester’s worth of lectures and I found him to be quite a nice lecturer and seems very approachable in person too. Also he tends to have his microphone volume either turned on low or it’s not close enough to his mouth so maybe sitting somewhere near the front will help. Especially with a subject like this sitting at the back allowed me to zone out a lot more and I wouldn’t know what’s going on by the end of the lecture.
Topics we cover include experimental design, data analysis, probability and its applications, probability distributions, estimation, hypothesis testing, comparative inference and regression & correlation. First four topics tend to be fairly easy but this builds on to the crux of the course which is estimation, hypothesis testing and comparative inference. If you don’t understand this then there’s a big chance you might fail. Regression and correlation are kind of a standalone topic much like experimental design and it’s not too difficult.
I recommend trying to find Sharon Gunn’s tutorials since she has the most experience in statistics and she’s very nice too. She’ll explain things clearer than Ray most of the time. Tutorials are mainly just discussing a few problems Ray has written up and maybe doing some text problems. The computer labs usually follow right after the tutorial and these use the Minitab program. It’s to get you used to using the program since the 2 computer tests and assignment are based on analysing data and answering questions using the computer program and interpreting output. Tutorials and labs aren’t compulsory. Tutorial solutions posted up on Friday afternoon every week and computer lab sessions are posted up on LMS when semester commences.
Computer tests and assignment aren’t meant to be hard but usually a lot of people struggled with it so make sure you collaborate with friends. With practice exams Ray has given us about 3 of them and you can find heaps on library site with no solutions but I wouldn’t recommend it. The revision sets in the back of the reader are old exams. The multiple choice of the old exams are actually put onto the weekly quizzes every Tuesday so it’s good practice. For the weekly quizzes you get 3 attempts with your best mark being your final but should be an easy 10%. Try to do the problem sets of all the chapters since they are either old exam questions or similar style with numbers tweaked around etc. Hopefully by the end of the semester you’ll see that this subject really is not all that difficult. Memory work is minimal since we get given all the formulas on the summary notes provided in the exam. The subject does get a bit dry but it doesn’t mean that it is difficult. Although around 95% of people struggle with the subject there is no reason to since you have plenty of resources available to you. I’ll reiterate that some concepts will be difficult to grasp and it requires you to look over it several times (especially all the power curve stuff and sample size determination etc.). Subject isn’t exactly enjoyable but workload isn’t really stressful either. You have 3 days to complete the weekly quiz and computer assignments vary from 1-2 weeks.
Lectopia Enabled
Yes with screen capture. However Ray tends to not show the projector screen on the lecture capture as an incentive to actually attend or reward those that do go.
Lecturer(s)
Ray Watson
Past Exams Available
Three past papers with solutions were given. Revision sets in the back of Ray's reader are mostly old exams.
Rating
3.5/5
Textbook Recommendation
Biostatistics for the Biological and Health Sciences by Triola & Triola. You definitely do not need this book. My heart goes out to those who did accidentally purchase it. All you need is the reader which is about 300 pages thick from co-op. If you really don't wanna fork out the cash Ray eventually uploads all the chapter text as he's doing the course and everything is on LMS. Recommend saving your valuable time by just purchasing it.
Workload
3 x 1 hour lectures per week, 1 x 1 hour tutorial per week, 1 x 1 hour computer lab session per week following the tutorial
Year & Semester Of Completion
2013 Semester 1
Your Mark / Grade
H1

Did you find this review helpful?

Study Honours at the no.1 university in Australia

Open to students from all universities, Honours in Biomedical and Health Sciences builds on your bachelor’s degree in science or health and enables you to explore your interests in research. If you’re interested in pursuing a PhD or becoming a qualified health professional, then Honours is an ideal pathway.

Find out more