On the debate of schools geography distribution
One thing that I noticed about the debate this week between Chee Soon Juan from the SDP and Vivian Balakrishan (PAP) was the claim that our schools favour the rich by nature. This is due to the congregation of the best schools in the central parts of Singapore, such as Bukit Timah where property values are higher. As schools worked by balloting due to geographical proximity, the argument is that rich people could afford to buy property in central areas, and hence get better education.
It struck me as a strange claim to make, as the schools Dr. Chee mostly quoted were secondary schools which worked by PSLE scores, rather than balloting by geographical proximity. I didn’t really put much thought into it until I saw a whole bunch of posts online and discussions where I felt many people seem to have this impression that our school system, especially of balloting, was favoured towards the privileged.
I do think there is some bias towards people who are more financially secure, as in all countries, but we need to be clear of our facts. Certainly, I don’t want our young studying in these “elite” schools to be plagued by guilt due to a misconception (from a rgs/rjc girl on social media?). I also think we need to grasp a handle of what’s really unfair rather than just accept an assumption that all our schools are located unfairly. When we think that there is a problem but is unclear of the root cause, any proposed solution will not work anyways.
Looking at the data behind our primary schools
So with regards to balloting for primary schools, lets see if schools are really unfair 🙂 Here’s what the data says.
I pulled out all the data of all primary schools from data gov. Strangely enough, I missed out 2 top primary schools, Catholic high and St Nicks Girls school from that data set. I think it’s a mistake in that dataset but let’s proceed without them.
So the claim is that the best primary schools are congregated in rich areas i.e Bukit Timah. So I had to define what is a good/bad primary school. I couldn’t really find an official ranking of the primary schools by the government. After all, “All schools are good schools”, am i right?
However, I found this ranking done by this parents forum (https://blog.learnable.asia/singapore-primary-school-rankings/), where they ranked all the primary schools in Singapore, from 1 to 185. I think we can use this as a proxy of how the good schools are, as after all, we are dealing with public perception of what the best schools are.
So, after some painful manual matching, I grouped the schools by percentile. The top 25th percentile, and so forth. Here’s how it looks.
Green – Top 25%, Yellow 25-50, orange 50-75, Red Last 75-100
From what it seems, There doesn’t really seem to be a favoured distribution of the top 25th percentile of schools. It also seems like the North East seems to be unfairly favoured. A simple explanation though, is that the North East is a much newer estate with newer schools, so it’s obvious that the schools there haven’t had time to build their reputation as good schools.
To make it clearer, let’s look at the top 25th% percentile of schools, without the rest.
Again, seems to be evenly distributed. I decided to cut down to the top 10th percentile to narrow the list. Here is the list of the top 10th percentile of primary schools btw.
- Nan Hua Primary School
- Nanyang Primary School
- PEI HWA PRESBYTERIAN PRIMARY SCHOOL
- RED SWASTIKA SCHOOL
- ROSYTH SCHOOL
- TEMASEK PRIMARY SCHOOL
- Methodist Girls’ School
- RULANG PRIMARY SCHOOL
- RIVERSIDE PRIMARY SCHOOL
- WEST SPRING PRIMARY SCHOOL
- JUNYUAN PRIMARY SCHOOL
- PUNGGOL GREEN PRIMARY SCHOOL
- EUNOS PRIMARY SCHOOL
- PEI CHUN PUBLIC SCHOOL
- SHUQUN PRIMARY SCHOOL
- WESTWOOD PRIMARY SCHOOL
- AI TONG SCHOOL
- Fairfield Methodist School
So here it is. In fact, I guess we can say that there appears to be an unfair distribution to the west, rather than central areas. Since primary school balloting works by 1km, 2km radiuses, I decided to draw 1km/2km circle around these points and see what areas are covered.
From here, there is a small cluster of good schools in Bukit Timah, but it mainly also covers areas like Clementi and Buona Vista, which has plenty of public housing. I don’t think that this is very strong evidence of an obvious bias towards the central areas, and won’t certainly make that claim.
On the other hand, we’re not juxtaposing this against the geographical density of these locations. A fairer metric would be to judge how many young parents are able to ballot for these schools, and the range of property prices around these areas. However, I couldn’t really find reliable data on that. Also, if we have a more reliable metric rather than the “public impression”, i would love to have access to that but I think public impression is the best we can go for.
What is obvious is that if you want your kids to go to a “top 10th percentile” primary school though, don’t live in Yishun.
And of course, we could selectively slice the data to fit whatever narrative we want. But I chose to fix to a top 10th percentile even before deciding on any hypothesis, since I don’t really want to bias my findings to any direction.
On making conclusions based on good data
So I think this is a good start, and I don’t think we need to be worrying or panicking about the “unfair distribution” of schools by geography. I think there are many other things that impact the equality of opportunity, but we should be relatively sure that geography of our best schools is not at the top of the pecking list.
And I’ll just like to remind everyone that while things might be obvious at 1st glance, maybe we can be a little more careful before jumping to conclusions. Maybe do some research and keep an open critical mind. I’m also a little afraid that we jump to conclusions from statements that are based on unproven data.
If anyone wants me to dig further, maybe for secondary schools, then let me know. Or if you have better datasets to share so I can dig further. And I’ll share the links of my raw data at the bottom of the article.
Key Points
- No obvious bias of the top 10th percentile of primary schools to the central areas (i.e Bukit Timah)
- We should probably move on from geography as as a factor for equality of education opportunity, and direct discussion and efforts to more salient factors
- Yishun is cursed
If you spot any errors (I don’t really check my work), feel free to reach out to yourdailytao@gmail.com
All raw data here:
https://docs.google.com/spreadsheets/d/1R7osmpxiFqU-ogGn7HRIvCTgEWNhqNFusgJkIJmvJTo/edit?usp=sharing
