Wednesday, 24 December 2014

Academic Update AY2014/15 Sem 01

Initially I planned to skip Academic Update for this sem as I am really busy with my internship, but then, I saw a spike in blog visit so I will just do a short update for the past semester to satisfy your curiosity. =P

Did relatively badly compared to the previous semesters. The heck-care attitude is really getting onto me as my CAP is unlikely to drop below 4.5 now (Assuming my FYP can get an A-, I just need an average grade of 2.0 for my remaining 3 mods).

Computer Intensive Statistical Methods
Linear Model
Bayesian Statistics
Macroeconomic Analysis II


The module content have some overlap with ST3247 and ST4234, I would recommend people to take ST3247 first and then do both ST4231 and ST4234 concurrently. For  the first half of the semester, the content is manageable, however, things get more difficult in the second half. Compared to Prof Vik who taught me ST3247, Alexandre Hoang THIERY is weaker in explaining difficult concepts. However, he does upload very detailed tutorial solutions and R codes, studying them will greatly enhance your understanding.  

Midterm mean-median: 46.16-46 (/70)

ST4233 A compulsory module for stats students. Workload is extremely light with only weekly tutorial which consist of a few questions, one graded assignment and finals. The assignment is basically free marks for all, so everything will depends on the finals (which is not difficult, but I screwed it up). Although Zhou Wang spent most of his lecture proving theorems and lemmas, there are not much proving in his Finals.

Assignment mean-median: 96-96 (/100)

This module has an extremely light content, you will probably spend most of the time manipulating some pdf/pmf to get the desired form and identify the distribution. Catching Zhang Jin-Ting's accent can be quite challenging initially, but you will get used to it over time. Overall, he is a decent lecturer. 

 Midterm mean-median: 70-72 (/100)

It was a long time since i last touched macro as I took EC2102 during y1s2, and macro was not my strength to begin with, hence I did not do well for it. It dosnt help that VU Thanhhai is quite week in his explanation and his mid term was really terrible with lots of grammar error and there was a question with double negative.

Midterm mean-median: 21.83-23 (/100)

As usual, if anyone would like to have materials such as lecture notes or tutorials for any of the modules that I have taken before, just leave a comment below or email me ( with your nus email. Do NOTE that, I don't any secret study guide or access to pyp solutions. The file I share will contain only lecture notes, tutorials/ assignment and its solution and maybe some pyp found from the library portal. Hence asking me for the materials towards the end of the semester is kind of pointless. They are meant to be shared with people who would like to read ahead or have a glimpse before they bid for the module.

Friday, 11 July 2014

Review for year 1 Modules

Some of you may be wondering why I did not post any module review for modules I have taken in my first year. That’s because I blur blur and accidentally deleted them some time back. As the semester is about to start, I thought it might be helpful to post my review for some of these year 1 modules. Note that I have taken these modules 3 years ago so it might not accurately reflect the current syllabus or assessment criteria, but I doubt that there were much changes.


A lot of theorems which can be quite intimidating at first, so it is necessary to be familiar with all of them in order to use them during the time-packed examination.
Dr Ng Kah Loon is the best lecturer I have encountered so far and I voted him as the best lecturer for this semester. Great in delivering concepts. Pace is consistent and manageable if you are prepared. Helps to keep the lecture quiet so that it is more conducive for those that are keen to learn. And a sense of humour that keep the lecture engaging.

Assessment and workload
20% midterm. 10% lab quiz. 70% final. Lab quiz is for you to score so do prepare for it. His Mid-term is challenging, expect the mean to be around 22-24 out of 40. Finals on the other hand is much easier.


More in depth compared to A level calculus. Some proofing questions that JC students might be unfamiliar with, and will take some effort to get the hand of it.
Lecturer: Wang Fei
Passionate and particular in teaching the right concepts.  He also like to spend some times during the lecture to share how students should adopt the right attitude when doing math. Some might think that he is wasting time talking about it, but I personally find them quite useful. Pace is quite fast, hence it is advisable to do pre- and post-lecture reading up on your own.  
Assessment and workload
Homework - 15%
Lab assignment - 5%
Midterm - 20%
Finals - 60%

Personal experience
I made a comeback for this module and got an A. Basically I screwed up my mid-term due to some personal reason, scoring slightly below the mean score. And partly because I don’t know many people in the course, I am among the few that could not get full marks for all the homework and lab assignment as I have to do them on my own while most of the others just copy or discuss with each other. However, I put in a lot of effort after I got back my mid-term results and breezed through the finals while many others complain that the paper was ‘imba’. So with a lot of hard work, you can score for this module.


Almost similar to H2 Econs but more in depth and some new stuff. Not recommended for those without Economics background as the content is really heavy.

Teaching staff:
Dr Cornie and another male lecturer. Dr Cornie's lecture just reap off everything from the textbook. Sometimes she seems a bit confuse over the slides. The male lecturer was much better, teaching stuffs that are beyond the textbook. 

Assessment and workload
Online tutorial (10%) - some technical screw up of the platform at the initial stage but was quickly rectified.
Mid-Term (30%) - MCQ. Easy. Everyone scored very high. If you are aiming for A+, u need to score full mark for this.
Finals (60%) - MCQ. Slightly harder but still quite simple. You still need to score close to full mark to get A+

Personal experience
Bell curve is extremely steep, majority of the students come in with econs background. Careless mistakes is not tolerated if you are aiming for A or A+.


Almost identical to JC stats with a bit new stuff towards the end. But the concept is still the same. Very dry module that makes me sleepy very lecture.

Teaching staff:
Wong Yean Ling speaks really really fast. Always on sick leave as she seems to be really ill most of the time... long winded but give good summary of the chapters

Assessment and workload
Mid-term - 35%, Finals - 60%, Assignment - 5%. Assignment is a free 5%, anyone could do it easily. Midterm is easy. Finals is tricky

Personal experience
Got an A-. Very dry and easy module, which is why I did not put in enough effort and didn’t get an A. Bell curve was terribly skewed as most people scored really high for it. Those with H2 math will be at an advantage.


General/special relativity. Quantum Mechanics. Particle Physics. No difficult math involved at all as it is only an exposure module. Abstract ideas that you can either accept it or you can’t.  Which is probably the reason why no matter how badly u do, the lecturer won’t fail you so that you can S/U it. Highly recommended for those that like to think a lot and has an interest in modern science and abstract ideas.

Teaching staff:
Prof Phil Chan is a passionate lecturer. However, he like to say 'please take note of it' at basically everything.
Tutors are very clear in clarifying doubts and explain concepts in another way to complement what was being taught in the lecture.

Assessment and workload
30%-Group Project (in groups of 2-3) Write a few thousand words report (either a story you made up or a book review). Choose your teammates carefully, avoid those who are intending to S/U if possible

20%x2 -2 tests one held during midterms (40 mcqs,1hr) and the other (45mcqs,1hr) before reading week. Both open book. Even though it is open book, you still need to be extremely familiar with the material as you will not have the luxury to flip through the lecture notes slowly to find your answers.
5%- star gazing (just go for 2 sessions. It's on certain Fri 7-10pm) Just need to go and sign attendance if you are not interested.

5% - forum discussions
10%- E-learning. Does not always apply. Not every semester will have e-learning.
5%- In class assignment. Free frag, TA essentially give you the answers.
5%- Tutorial Attendance


I did not have any background in computing, but was still able to follow the syllabus fine. The module teach a lot of basics C programming which are easy to learn. However the challenging part is applying these basics programming techniques to write programs which can be quite complicated and not straight forward. However, if you get yourself familiar with the programming language and is able to think logically, you will do fine in this module.

Assessment and workload
5 Lab Assignments - 5%
Mark is awarded by attempting. As long as you submit your assignment with reasonable effort, you will be awarded the 1% for each assignment. NOTE: Do not copy your program from your friends. They are able to detect copied work and I can assure you that you won’t like the outcome if you are caught.

Discussion Attendance - 5%
Free 5%. Just make sure you attend every tutorial and present some solutions to the problems and you will get all 5%

Practical Exam 1. - 10%
Easy. Median mark was around 8.8/10. Try to score well for this.

Practical Exam 2 - 25%
Challenging. Medium was around 60-70%. Still do-able. Practice on recursive

Mid-Term - 15%

Final - 40%   

Personal experience
I scored an A- for this module which is kind of disappointing. For some reasons, I could not do some of the not-so-difficult questions during the exam and some fatal careless mistakes which caused me my A. The good thing about this module is that it is not bell-curve based, so you do not need to worry if your course mate is extremely good in the module. As long as u manage to get certain marks, you will get the grades u deserved. Another thing is that the grading of this module is extremely transparent, you will more or less know your overall result and hence predict ur grade the minute u finish your final paper.


A more in depth version of the first part of MA1102R (no differentiation and integration). The content of the module is very light with only about 5 topics. However, there are a lot of proving in this module. Textbook is not a must-have, but will come in handy at times.

Assessment and workload
Homework Assignment - 10%
Like all other maths modules, everyone will be scoring near full mark for homework assignments, so make sure you do too.

Tutorial participation - 5%
Present the tutorial questions 2 times in class.

Mid-term - 20%
Relatively easy paper. Be careful of careless mistakes. Median score is 71 for my semester. The format of the paper is exactly the same as the previous sem, even the questions are similar. I could have done much better had I known this fact.

Final - 65%
A lot of proving questions. Personally I feel that the paper is not do-able at all, I do not know how to do at least half the paper and wrote rubbish in it, praying that it might earn me some sympathy marks.   

Personal experience
The most terrible experience I had ever had for any math modules. For the first time in NUS I could not do half the paper and left the exam hall feeling really terrible. But in the end I think the bell-curve saved me and I still manage to get a B+.


Content was not very heavy. Focus a lot on understanding and manipulating some economic framework graphically. Textbook is a MUST. If you are unwilling to buy the textbook, there is a free eBook ver. online available for download (google it yourself).

Assessment and workload

Tutorial Participation - 10%
No idea how is the point awarded. But just be as active as you possibly can. If you are really shy, attempt to present your solution at least twice.

Mid-Term - 30%
MCQ. Slightly challenging. Median was 21 and the 75th percentile was 24 out of 30. Know your graphs well.

Finals - 60%
Closed book. Know your graphs well such as how to derive some graph from another and practice on solving simultaneous equations will be useful.

Personal experience
Got an A for this module. Know your graphs well and understand them and you will do fine in this module. If you know some math, it will be an added advantage.


Astronomical objects and events. Pretty interesting most of the time, particularly if you like astronomy. Prof Cindy had a funny accent, and she can be quite funny sometimes.  Textbook not required.

Assessment and workload
Tutorial Attendance + Assignments - 20%
Almost free 20%. Attend every tutorial and do the assignments (simple calculations, just be careful of careless mistakes) and you should expect more than 18%

Term test 1 - 20%
MCQ - closed book. Beware of stupid questions such as the date of next astronomical events which does not question your understanding of the concepts you have learned. For the semester I took, the mid-term was harder than usual as the highest mark was only 28/30. The lecturer claim that this was because she had run out of easy questions. Focus on memorization skills

Term Test 2 - 20%
MCQ - Open book. Test more on concepts. Some questions requires you to know where to find the answer in the lecture notes unless you memorized every single little details which is kind of impossible. Term test 2 might be more challenging to some, but because I suck in memorization, I think I fare better for this test compared to the first one.

Group project + individual Report (20% + 20%)
Do an experiment related to astronomy with a group of around 4 people (google for the experiment). The prof will show you a lot of past project done by students from the previous semesters. It is okay to do experiments that other people had done before, or will be doing, because trust me, after so many semesters, I believe the prof have seen every possible experiment out there. The IMPORTANT thing is that your experiment must have a lot of variables (at least 3) for you to play around with. For the individual report, you need to demonstrate your understanding of the experimental results and explain why certain trend or anomaly exist. As long as your explanation make sense from physics point of view, then it will be fine.

Personal experience
Got an A+ for this module. Read up and understand the concepts from the lecture notes. Even though it is an open book test for term test 2, prepare a cheat sheet with important dates or information so that you can find them quickly during the test. I spent a lot of effort doing the individual report, (more than a week including reading up and researching on the topic I did) while I have some team mate who had not even started the report less than a week before the deadline, I even heard a random stranger boasting that he completed his report within 5 hrs. Spend some effort on the report and it will definitely help u with your grades especially when a lot of people are taking this module with a heck-care altitude.

Tuesday, 1 July 2014

Online Survey? You are doing it WRONG!

I am sure many of you have conducted surveys at some point in your life. Many students love to use survey results in their projects to substantiate their stand or to identify trends pertaining to certain issue.  In a properly designed and sampled survey, this is certainly viable. The problem is, conducting a proper survey requires so much resources, it is almost impossible and impractical for any students to do so. Most students choose the easy way out by conducting online survey due to its convenience. The end result is a brunch of numerical values fitted forcefully and incorrectly into statistics. I have even seen an intern from a certain national body recruiting online survey responses through Facebook. Since then, I have serious doubt on the validity of the reports and findings produced by that particular organization.  

The ability of a small sample in a survey to reflect the opinions of the larger target population does not happen haphazardly. It requires proper probabilistic sampling method. Otherwise, even simple statistics such as taking the average of the response is wrong as it implicitly assumed that everyone in the target population have equal probability of responding to the survey, which is certainly untrue given the way how most students conduct their surveys. 

Self-Selected Sample 

With the advent of forums, blogs and social medias, one of the most commonly used methods for students to conduct such surveys is through these online platforms. I mean, why wouldn't they? With the help of the countless free survey tools available online, it is extremely easy for anyone to set up a questionnaire and distribute it online to gather responses. Most of these tools are easy to use, often producing beautiful graphs and charts with the click of a button and most importantly, they are convenient. 

The main problem lies in the fact that the respondents are self-selected. These volunteer samples are likely to have stronger opinion on the survey topic, which compelled them to do the survey in the first place. In the case of conducting survey through social media such as Facebook by asking your friends to fill up the survey, it may also lead to over-representation of people who belongs to a very similar demographic group. 

As an illustrative example, in 1993, shortly after Bill Clinton was elected as the president of the United States, a TV station in Sacramento, California conducted a TV poll seeking viewers’ opinion to the question: “Do you support the president’s economic plan?” Coincidently, the result of a properly conducted survey asking the same question was published around that time, with the following results: 

Self-Selected Sample (TV Poll)
Random Sample
Not Sure

As shown in the table, the results are contradicting. In the TV poll, majority of the viewers are not supportive of the president’s economic plan while in the proper survey, 3 in 4 respondents are supportive of it. In addition, none of the respondents in the TV poll was ‘Not Sure’, highlighting the tendency of people who responded have a stronger opinion.

In summary, survey which relies on self-selected sample is a complete waste of time. The result is meaningless, or worse, misleading. The only accurate information you can derive from such survey is probably the count of number of people who bothered to respond. 


Even for some reasons, you managed to get a proper sample, e.g. in a very simple survey where your target population consists of only your class. There are many other factors which need serious consideration, but are often ignored. One such pitfall is the wording of questions used in the survey.
Building questionnaire may seem like a simple task at first, but this couldn't be further from the truth. The way you phrase a question and the order in which respondents respond to them can have unintentional consequences on the resulting responds.
In an experiment conducted by Loftus and Palmer, a video footage of an automobile accident was shown to 2 groups of college students. The first group was then asked the question: “About how fast were the cars going when they contacted each other?” While the second group was asked “About how fast were the cars going when they collided with each other?” The average responses of the 2 groups were 51.2km/h and 65.7km/h respectively. Simply changing one word in the question led to a very different result.
The ordering of questions can also skew the responses in the survey. For example, if a survey asked, “How often do you dine at KFC?” and then asked “Name the top 5 fast-food restaurant you think teenager visit most frequently.” It is quite likely that the respondents will include KFC in the latter question.
These two examples are definitely not an exhaustive list of the potential pitfall one can encounter when designing questionnaires. There are many other considerations researchers should be aware of in order to create a good survey. However, I have no desire to turn this blog post into an entire chapter, so I will skip them. For those who are interested, you may read more about them here

Characteristic of a proper survey 

So how do we conduct a proper survey? While each survey should be customized according to the objective of the study, a good survey necessarily share the following characteristics:

1.       Well defined target population
Who are the people or what are the objects you are trying to make inference on? It also allows us to identify potential sampling frame. 

2.       A sampling frame and sampling method
Sampling frames are lists intended to identify all elements from the target population. However, a perfect sampling frame which contains all elements in the target population can be prohibitively expensive or even non-existent. Therefore, we often have to make do with frames that are imperfect which may not contain all elements in the target population or include ineligible elements.
After obtaining the sampling frame, we have to have a probabilistic method of selecting the sample from the frame in order for the survey result to be statistically valid. The simplest approach is to use Simple Random Sampling, where each element in the frame is assigned a probability of selection and you select a sample of predetermined size based on this probability profile.

3.       Questions should be carefully phrased to avoid bias. Pilot study to ensure question yield the desired information
As mentioned previously, wordings can cause bias to the response collected. Therefore, care must be taken to ensure the questions in the survey have minimal bias. If need be, pilot study should be conducted to evaluate the questionnaire’s effectiveness in eliciting the desired information. 

4.       Properly trained surveyor
If the survey requires surveyor to collect responds from respondents through face-to-face interview or telephone interview, adequate training should be provided in order to minimize influence of the surveyor on the respondents. 

5.       Ensuring a good response rate
In a sample survey of humans, it is almost impossible to obtain 100% response rate. Some respondents could not be reached while others simply refused to be interviewed. People who tend to respond to a survey could be very different from people who don’t. This difference can bias the survey result. Therefore, measures should be taken to minimize non-response. 

After reading this post, I hope you will realize that designing a proper survey requires a lot of efforts and resources. The way you have been conducting surveys are most probably wrong and the results are nothing but a pile of meaningless numbers. As a student, you may be bounded by the syllabus to churn out ‘surveys’, but as a responsible user and consumer of data, this stupid practice of conducting survey with no statistical basis should be dumped into the garbage bin where it belongs.