The Futility of Marking Electoral Registers

One of the complaints that many people have about the conduct of elections in Singapore is that PAP polling agents are stationed at polling stations and conspicuously mark off voters’ attendance on their own copies of the electoral register during polling. This is completely legal and while indvidual voter’s votes are secret, the practice leads to an environment that reinforces the perception of pervasive surveillance in Singapore.

When e-registration of voter attendance was first announced a few years ago, I urged the Elections Department (ELD) to retain its practice of having Presiding Officers (POs) call out voters’ particulars while issuing ballot papers. This was so that polling agents could prevent impersonation or multiple voting (the same person voting more than once). In practice, however, the only effective checks against those malpractices are the Presiding Officers (POs) who check voters’ particulars and mark attendance in the official copy of the register. Polling agents are not allowed to check voters’ identity cards so they cannot really verify voters’ identities, and because voters may now receive their ballot paper at any of several tables, it is not practical for polling agents to keep track of which particular voters have already voted.

I have therefore changed my position in that while I still think it is important for transparency that POs audibly read out voters’ particulars when issuing ballot papers, there is no meaningful purpose in polling agents marking their own copies of the electoral register. For all the effort that PAP polling agents expend on marking their copies of the register, I’m very sure that Party branches just throw them away at the end of the election. Even if there were ever a dispute over the results of an election, the PAP’s marked copies of the electoral registers would have no legal standing. And given all the brouhaha over their polling agents not being able to hear properly, we can infer that the PAP’s records are chock-full of errors anyway.

As the Election Department(ELD)’s Guide for Polling Agents puts it, the role of polling agents is to “observe that polling at the polling station is carried out in accordance with the law”. They may mark voters’ attendance in their own registers but there is no legal obligation for elections officials to assist them in that task. Specifically, the ELD Guide states

5.6 Polling agents should pay close attention when the POs are reading out the particulars of the voters. They must not ask the PO to repeat the voters’ particulars or check their own copy of the register (obtained from their political party or candidate) against the PO’s copy, as this will disrupt the orderly conduct of poll.

Elections Department, “Guide for Polling Agents for General Election 2020”

This admonition against polling agents asking POs to repeat voters’ names and serial numbers has been in the Guide for Polling Agents since the first version of the Guide was released in 2011. Historically, enforcement has been lax, however, and it was common for PAP polling agents to ask POs to repeat themselves during past elections. This problem was aggravated this year because of the new polling station layout resulting from the switch to a centralized e-registration system and additional precautions taken for Covid-19.

Typical layout of polling station for 2020 General Election with labels for polling agents added to original drawing from ELD.

Unlike in past elections where polling agents were seated directly facing POs, polling agents are now seated further away from POs. In addition, POs were wearing face masks as a precaution against Covid-19. This made it more difficult for polling agents to hear the POs and led to much frustration on the part of POs who had to shout themselves hoarse trying to satisfy the PAP polling agents. Ironically, the POs were under no obligation to repeat themselves and should have just refused the PAP polling agents’ demands by referring to the ELD’s Guide for Polling Agents which specifically states that polling agents should not ask POs to repeat themselves. But the PAP polling agents were themselves performing a pointless task. Their marked copies of the registers will never be used, and their presence in white-and-white is sufficient to show their Party colours. There is no real need for them to go through the wayang of marking their electoral registers if the concern is just to prevent wrongful issuance of ballot papers.

As another polling agent has commented, PAP polling agents seem to be so obsessed with marking their electoral registers that they don’t pay attention to anything else. Indeed, I wonder what kind of training the PAP provides to its polling agents. From what I observed as a polling agent during this election, I got the distinct impression that at least some of their polling agents had not received any training or briefings at all before showing up and just being told to mark voters’ attendance in a name list. It is this focus on taking voters’ attendance that leads them to ask POs to repeat voters’ particulars for fear that they will be scolded by party bosses later on if they miss out anything. Because marking of the register is a tangible output, the PAP’s polling agents get fixated on it to the detriment of their more important task of keeping an overview on general proceedings in the polling station.

Arguably, the PAP may gain some votes from people who are reminded of the omnipresence of PAP-affiliated grassroots organisations when they see white-clad polling agents marking their names in a file, but this has to be set against votes lost from other people who dislike seeing the PAP trying to exert undue influence on voters. Over time, the votes lost from this practice may exceed the votes gained. The PAP will have to decide when that point is reached and how long they want to continue the practice.

I do not see ELD going back on e-registration which means that in future elections, polling agents will still be seated some distance away from POs and PAP polling agents will still get tempted to ask POs to repeat themselves. But after the experience of this election, polling agents from other parties are more likely to point to ELD’s own guidelines and remind elections officials that the PAP polling agents are not supposed to ask POs to repeat voters’ particulars. POs themselves may also become less willing to go along with the PAP polling agents’ demands.

A related issue is the provision of tables for polling agents. Tables were provided for polling agents in previous elections, but in the past, polling agents sat directly opposite POs and were much closer to voters so the tables acted as physical barriers between the polling agents and voters. In this year’s layout (see above), polling agents were no longer provided with tables. I do not know whether this was due to a general rethinking of polling station layouts after the introduction of e-registration or was a result of Covid-19 safe distancing requirements. Either way, it is a postive step. Providing tables for polling agents gives voters the impression that the PAP’s polling agents are acting in an official capacity when they mark their copies of the electoral registers. But they are not. They are merely observers. Tables are definitely not required for polling agents to do their job. If it wants to, the PAP can just buy clipboards for its polling agents in future elections.

The bigger question is why the PAP even bothers to mark voter attendance at all. Its polling agents’ habit of asking POs to repeat voters’ names and serial numbers just annoys both POs and voters. At some point, this practice may become a nett vote-loser for them. Marking the register does not help to catch electoral fraud and if I were a PAP polling agent, I would be asking my candidate why I am wasting my time and energy on a ritual that serves no useful purpose.

Badly behaved polling agents

I volunteered as a polling agent for the Workers’ Party in Marine Parade GRC in the recent election. It has taken me a few weeks to write this up unfortunately as I was busy with work commitments.

I have been involved in every General Election since 2006 as a civil servant (elections official) or observer. Election law in Singapore does not provide for independent, non-partisan observers so observers have to be appointed as polling agents or counting agents by candidates. While this does inject an element of partisanship into the role of polling and counting agents, I have always had a cordial relationship with observers from other parties in the past. Elections offficials have also always tried their best to be fair to all sides and to follow the procedures laid down by the Elections Department (ELD).

Kirsten Han has already written about PAP polling agents’ habit of interrupting proceedings by asking Presiding Officers (POs) to repeat the names and serial numbers of voters so that they (PAP polling agents) could mark voters’ attendance. PAP polling agents have done this in previous elections too, but it was worse this year because of the new polling station layout where polling agents are seated further away. Covid-19 precautions also made it harder to hear the PO’s because their voices were muffled by their face-masks. Unfortunately, one of the PAP polling agents at the Maha Bodhi school kept demanding that the PO repeat herself and was so insistent to the point that the PO shouted back at the polling agent.

Earlier in the day, this particular PAP polling agent had taken a table from a storage area to use to mark voter attendance in her copy of the electoral register. I objected to this on the grounds that giving the PAP polling agent a table created the appearance of favoritism. As seen in this official poster from ELD below, polling agents are provided with chairs but not tables. I did not see why a polling agent should be allowed to bring additional furniture into the polling station for her own convenience, especially since the table would give her the appearance of being in a position of authority. The other PAP polling agent at the polling station was able to mark his register without a table and I’m sure thousands of other polling agents around Singapore managed quite well without a table too. After a short discussion, the Senior Presiding Officer (SPO) accepted my objection and made the polling agent put the table back. This did not stop her from marking her register of course; she just rested her file on an extra chair instead of using a table.

Standard polling station layout in 2020 – Position allocated to polling agents is circled in red. Note that while polling agents are provided with chairs, there is no entitlement for tables.
Based on

This was not the most egregious behaviour I saw from the PAP polling agents that day. Around 11 am, two women dressed in white appeared and started talking to the two PAP polling agents. This raised my eyebrows because the women were not wearing any official passes and should not have been allowed to enter the polling station. Based on their dressing, however, I inferred that they were PAP polling agents. This was still a problem because each party was only allowed to have up to three polling agents at a time [1]. If the two women were also PAP polling agents, that meant that the PAP now had four polling agents in the polling station and had exceeded the legal limit.

Despite my concerns, I decided to cut them some slack because it was not unreasonable to have an overlap of shifts for a short time during shift changes. It was only after they continued talking among themselves for some time that I realized that the first shift of PAP polling agents was training the second shift while inside the polling station. As neither the Assistant Returning Officer (ARO) nor SPO were present at that time, I advised the PAP polling agents that they had exceeded their limit on the number of polling agents and that they should finish their handover as soon as possible. I was shocked when the PAP polling agents snapped at me in response. It was only when I reminded them that they were breaking the law that the original two polling agents withdrew. Luckily, the SPO returned at this point and after I informed him that the two newcomers were not wearing any official passes, he instructed them to leave as well. After a few minutes, the original two PAP polling agents came back and stayed until 12 noon when the second batch took over (after having been issued the appropriate polling agent passes).

I must say that the first two PAP polling agents were very much the exception. None of the PAP polling agents or counting agents I have encountered in the past have gone so far. This is the first time that I have seen such arrogant and entitled behaviour from PAP polling agents. Starting from the polling agent who thought that she had could rearrange furniture in the polling station for her own convenience to the brazen attempt to sneak in two unregistered persons and conduct a training session inside the polling station while voting was in progress, I can only describe those two PAP polling agents as behaving as if they thought it was their grandfather’s polling station.

Other than the misbehaviour of the first two PAP polling agents, polling generally passed uneventfully that morning and early afternoon. Lines were not unreasonably long and while the second shift of PAP polling agents did occasionally ask the PO’s to repeat themselves, they were not disruptive and were much more apologetic about troubling the PO’s.

I do not think that the bad behaviour of the first two PAP polling agents affected the integrity of the election. However, their actions reinforced the stereotype of the PAP as being arrogant and not playing fair. Related to that is the PAP’s long-standing practice of marking the attendance of voters in their own registers. Whatever benefit they may have gained from that in the past, I don’t think that it is a nett positive for them any longer. But more on that in another post. Akan Datang…

See this post for more of my thoughts on polling agents and marking of registers.


  1. Under Regulation 3 of the Parliamentary Elections (Polling Stations) Regulations 2019, candidates are allowed to have one polling agent for every thousand voters assigned to a polling station. In the case of the Maha Bodhi School (B) polling station, that meant that each party was limited to having three polling agents at the same time.

Doing the same thing and expecting different results

Sigh. The past connects to the present. When it was first passed in 2014, the Protection from Harassment Act (POHA) was sold as being for the protection of vulnerable individuals from harassment. But not only did the Government try to make use of POHA to suppress criticism, the first Government agency to make use of it was Mindef. The Ministry of Defence ! Mindef can call upon whole regiments of armour, artillery, fighter jets and frigates and yet it could claim to be “harassed” by mere words ??? If I had still been an SAF reservist at that time, I would have been too embarassed to know where to hide my face. The Government eventually lost at the Court of Appeal, but the first mistake it made was in using Mindef as the test case to try to set the precedent. The idea of Mindef being “harassed” was just too far beyond belief for the Government’s attempt to abuse POHA to sit well with the public. If the Government had used a softer, cuddlier agency as the first case, it might have stood a chance of easing in its interpretation of POHA without triggering such a strong reaction.

Which brings us now to POFMA. Immediately after the Government lost its POHA case, a Ministry of Law spokesman said that it was considering a new law against online falsehoods. And now, in its very first use of POFMA, the Government issues a correction order against an opposition politician for statements which are substantially opinions rather than facts. So much for the scenarios of racial or religious violence, foreign interference in elections, or national security painted by the Government when it pushed the Act through. By no stretch of the imagination could Bowyer’s comments threaten the “public finances” of Singapore. Furthermore by targeting an opposition politician, the government just reinforces fears that POFMA is intended to be used to gag political opponents rather than in the national interest. Honestly, this was just stupid and will backfire on the Government. The Government expended a lot of political capital to push POFMA through. Why did it waste it on such a trivial case as this ?

AIM- Gone, Resurrected and Gone again

One of the more sordid political sagas that we have seen in Singapore over the last decade was the sale and lease-back by 14 PAP Town Councils of their IT system to a PAP-owned shell company, Action Information Management Pte Ltd. The justifications put forward by the PAP after the sale came to light were unsatisfactory to say the least, and while I do not believe the Party or any MPs profited financially from the deal, the PAP’s actions were symptomatic of their arrogance and culture of impunity.

One of the most basic decision tools taught in any corporate ethics class is the newspaper test – Before taking any course of action, ask yourself what the reaction would be if it were published on the front page of tomorrow’s newspaper ? Would you still do it if that action were public rather than secret ? AIM obviously failed the newspaper test but none of the 82 PAP MPs or hundreds of other appointed Town Councillors said anything. Admittedly, scholar-generals don’t come from private sector so maybe they have never gone through compliance training, but at least some of the rank-and-file MPs and Town Councillors do work for MNCs and should have been able to see the red flags flying furiously over AIM. But no one saw anything or said anything. Or perhaps even ordinary MPs and Town Councillors did not know about AIM. Which would itself be a damning indictment of the level of governance of PAP Town Councils.

I’ll skip over the whitewashing of the PAP’s actions by MND and go straight to the striking-off of AIM four years after the scandal first broke in 2012. Striking off is basically a means of closing a company by declaring to ACRA that the company is no longer in business and has no more assets or liabilities. The legal formalities were complied with and AIM officially struck off the company register in July 2016.

I was stunned, therefore, when Andrew Loh reported on November 2 that AIM had been restored to the the company register. While it is prefectly legal for companies to be restored after being struck off, this was extremely surprising because when AIM’s directors, all former PAP MPs, requested for AIM to be struck off in 2016, they must have already decided that there was no further use for AIM, and also made a declaration to ACRA that the company was no longer in business. What would make them want to revive the company after two years ?

Searching for AIM on ACRA’s website gave two hits, one for a restoration notice and one for the original striking off.

The first link showed that AIM had been restored on 28 Nov 2018 but ACRA’s business entity search tool still showed AIM’s status as struck off. Which was it ? Was AIM actually restored or not ? The answer came on 15 November as I was digging around ACRA’s website and discovered that the listing of companies deregistered in 2018 had changed. Whereas the original list had 71 entries, the new list only has 42 entries and AIM is not on the list !

[See Addendum 2 below]

Somehow or other, ACRA must have uploaded an incorrect listing of restored companies and nobody noticed for almost a year. That then leads to the question of whether ACRA discovered the mistake on their own or whether they were informed of it by some people who follow Andrew’s posts very closely. Either way, it is quite embarassing for an Agency whose job it is to make sure that business entities file their reports accurately and on time.

The bright side of this is that AIM appears to still be dead but it is disconcerting that ACRA could make mistakes like this. I have submitted an inquiry to ACRA asking why there were two versions of the 2018 Restorations document but haven’t received an answer yet.


Another anomaly with regard to AIM is that ACRA’s website does not mention when the First Gazette Notification for its striking off was published. Ordinarily, ACRA will publish a First Gazette Notification that a company will be struck off and the striking off will take effect after 60 days if no objections are received. Out of the 1,495 companies that were struck off on the same day (4 July 2016) as AIM, all but one were included in the May 2016 First Gazette Notification. The one exception was AIM. Admittedly, I am only referring to the summary documents published on ACRA’s website. Legally, only the official Government Gazette counts and it’s quite likely that the first notice of AIM’s striking off was published in the Government Gazette in May 2016 but ACRA just missed it out when it compiled the list for publication on its own website. Unfortunately, the Government Gazette is paywalled and is extremely expensive so I haven’t been able to check yet. [Could someone with an eGazette subscription help me to check what were the actual dates of the First and Final Gazette Notifications for the striking off of 199103607Z ACTION INFORMATION MANAGEMENT PTE LTD ?]

Addendum 2

So ACRA got back to me, and guess what ? There is now a third version of the list of companies restored in 2018 on the website. There are 70 entries in the latest version compared to 71 in the first version. The only difference between the two versions is that the latest version does not include AIM.

VersionFile Name/LinkCreatedNumber of entriesIncludes AIM
12018-restoration.pdf8 Jan 2019 17:2971Yes
22018-restoration-notification-for-s344e-and-s344f.pdf6 Nov 2019 14:5142No
32018-restoration-notification.pdf19 Nov 2019 17:2670No

The first version was taken down from ACRA’s website around 15 November and replaced with the second version. The second version was itself taken down around November 25 and replaced with the third version. The difference between the second and third versions is that the second version excluded 28 companies which had been restored in 2018 but then struck off again this year. In response to my inquiry, ACRA replied that AIM was “struck off on 04/07/2016. It has no records of restoration”.

ACRA was extremely unlucky to have not only hit a “politically exposed company” by mistake in the beginning but to have also made another mistake when it tried to correct its first error. All I can say is that AIM is just bad for the reputations of everyone who touches it.

British Colonialism and Rule of Law

One of the things we keep on hearing as the government commemorates the bicentennial of the colonisation of Singapore by Raffles is that it was thanks to the British that we have “Rule of Law” in Singapore. Well, yes, Singapore ranks highly in global surveys of perceptions of Rule of Law, and Singapore was a British colony, but Fiji, Kenya and even India were all British colonies and none of them are known for having strong Rule of Law. So, was it really the British who made the Rule of Law stick in Singapore ?

I decided to test this proposition by looking at the World Justice Project’s Rule of Law Index and comparing the scores of former British colonies with that of other countries which had never been colonised by the British. Out of 126 countries in the index, 42 were former British colonies. And the average overall score of the former British colonies was exactly the same as that for all other countries (details below) ! In other words, having been colonised by the British makes no difference as to whether a country has strong rule of law today.

Continue reading “British Colonialism and Rule of Law”

How old is this Bridge ?

Sandwiched between an industrial park, a cluster of Chinese temples and two giant Newater storage tanks is a short stretch of road now known as Tampines Avenue. This road is a remnant of the original Tampines Road before the old road was diverted for construction of Paya Lebar Airport. The area has the feel of a small Malaysian town and is a little corner of Singapore that time forgot. About the only reason most people would go there today is a restaurant, Goodyear Seafood Village that used to serve good Pontian style Bak Kut Teh 5 or 6 years ago but whose standard has dropped substantially since then. Their zichar dishes are still okay though.

What motivates this post, though, is the bridge and the “1889” marking shown above. I’m wondering if that could actually mean that the bridge was built in the year 1889. Tampines Road is a pretty old road, and it appears in maps as early as 1873. I’ve attached two screenshots comparing the 1873, 1945 and 1975 maps from Drainage is not shown in the 1873 maps, but the 1945 and 1975 maps show that Tampines Road crossed a stream roughly at the location of the present bridge as far back as 1945.  Of course, the stream has become a concrete drain today but could it be that the bridge has existed since 1889 ? Or first built in 1889 but rebuilt or renovated in the 130 years since then ? If it really was built in 1889, that would make it as old as the Read Bridge at Clarke Quay.

Maps from 1873 and 1945 showing Tampines Road before construction of Paya Lebar Airport

Somebody has suggested to me that the 1889 does not refer to the year of construction but could be a serial number or some sort of other marking. That’s certainly possible. The bridge including the markings can be seen in Google Street View ( I took those photos five years ago and for now, there is no sign of imminent redevelopment, but I’m hoping that I can solve this mystery before the bridge and everything around it gets demolished.

Afternote (21 Jan 2019):

I was eventually referred to someone in LTA and got this answer, “We have checked the mentioned bridge along Tampines Avenue and noted that there is no bridge structure on site but only a pipe culvert. Also, we do not have records of when the said pipe culvert was built.”

Oh well….

Postscript 14 Sep 2020

Doesn’t matter how old it is or whether it is a bridge or a culvert, it’s gone now.

Fixing the Mazda 2 AUX jack

So my 2010 Mazda 2 started to act up on me a few months ago – At first I had to jiggle the Aux cable a few times to be able to switch to Aux  input, but after a whiIe I had to use one hand to hold the plug in at just the right angle and another hand to press the Aux button on the control. A bit difficult to do while driving…

As it turns out, this is actually a common problem in Mazda vehicles, caused by their use of two pins to detect the insertion of the jack. The pins are supposed to be shorted out when the jack is inserted but after a while the pins wear down or get loose and the stereo is not able to sense that an AUX cable has been inserted. According to this YouTube video, the solution is to short out the two pins permanently, so here goes…

First, we have to open up the console. Press in and pull up the front of the console:

This will reveal the cables connecting to the lighter and the auxiliary (AUX) port. Carefully pull out the connector for the Aux port and note the wires connecting to the Aux connector. On my car, there were five wires: pink, black, white, red and green. The middle black, white and red wires presumably carry the audio signal while the pink and green wires are used to detect the insertion of the plug.

According to the video mentioned above, pins 2 and 6 have to be shorted, but note that that video refers to a US (?) Mazda 3 and other forum posts mention different pin numbers for other Mazda models. So, first thing to do is to check which pins to short. I just used a short length of wire to short out the pink and green wires and confirm that I could now select AUX using the control button.

VERY IMPORTANT ! The method I used is a destructive method so be very, very sure that you have the correct wires before cutting them.

I cut the pink (#2) and green (#6) wires and shorted them together. In retrospect, I think the green (#6) wire was actually ground and shorting wire #2 (pink) to ground is probably the signal to indicate insertion of the Aux plug. It would have been smarter to cut the wires 1 cm away from the connector and use a crimp connector to short the pink and green wires together so that it would have been easier to reverse the process, but that’s 20/20 hindsight. This is what I actually did: I just cut the pink and green wires, stripped a little bit of the ends and wrapped them together.

After putting everything back together again, I am now able to switch between radio and AUX using the button and even though I cut the green (ground ?) wire there doesn’t seem to any static or other interference so far.

Watching the watchers

Originally published under the title “An eye for an eye” in the Computer Times supplement of The Straits Times on Aug 13, 2003

In 1991, the Law Reform Committee of the Singapore Academy of Law proposed that Singapore adopt a Data Protection law to complement the Computer Misuse Act.

Over the next decade, many countries such as Canada, Australia and Hong Kong introduced new laws or strengthened existing laws on Privacy and Data Protection.

Singapore, in contrast, chose not to adopt any omnibus Privacy or Data Protection laws even though it enacted a far-reaching Computer Misuse Act.

This reflects the fact that, in Singapore, the balance between individual and group is tilted firmly in favour of the group. This was clearly seen in the Government’s response to the severe acute respiratory syndrome (Sars) outbreak.

Surveillance cameras were installed in the homes of persons under quarantine. Confirmed and suspected Sars patients were compelled to reveal details of their movements and contacts under amendments to the Infectious Diseases Act.

During parliamentary debate on the amendments, only one MP, Mr Chiam See Tong, voiced concerns: The possibility that information collected by Health Ministry contact tracers could be misused.

He supported the Bill, on balance, and in view of the urgency of the situation. But Mr Chiam’s concern over possible abuse of personal information is certainly justified.

Several police officers have been convicted for using the Ministry of Home Affairs database to illegally obtain personal information about other people.

Sars contact tracers are covered under the Official Secrets Act (OSA) but the OSA has primarily been used to guard the Government’s secrets not individuals’.

Other laws such as the Income Tax Act and Statistics Act specifically prohibit the disclosure of information obtained under the respective Acts.

The Census Act also provides that information collected during a census cannot be used as evidence for prosecution of offences under other Acts. It is unfortunate that no similar safeguards were built into the Infectious Diseases Act.

One positive development in Privacy protection in Singapore has been the incorporation into law of ‘fair information principles’ as they apply to the Consumer Credit Bureau.

The bureau is required to ensure that all the data it collects is used only under specified purposes. They are: accurate and up-to-date; accessible to the subject; kept secure; and destroyed after a specified time

The right of access to our own credit reports is a landmark in Singapore privacy law. Up until now, individuals have not had a right of access to personal information held on them by government departments or businesses.

Ironically, while we have no right to view our own medical records, health care providers are not legally obliged to protect the confidentiality of our medical records.

In the Simon Shorvon scandal, pharmacies at two hospitals released the names of patients taking a certain drug to Dr Shorvon even though he had no legitimate use for that information. Had the pharmacies put more robust patient confidentiality rules in place, the scandal might well not have taken place.

The National Trust Council has finalised its Data Protection Code but because it is voluntary and intended primarily to promote e-commerce rather than to protect individuals, it is of very limited value to people seeking to protect their privacy.

Some recent events illustrate the difference in the Government’s response to surveillance technology when it is used by individuals rather than by the Government itself.

In the first incident, a Raffles Junior College (RJC) student used a personal digital assistant to videotape his teacher berating another student and tearing up the student’s notes. This video eventually made its way onto the Internet and resulted in a reprimand for the camera-wielding pupil.

The college’s principal spoke of ‘betrayal’. Indeed, government ministers lost no time in admonishing the student for the irresponsible use of technology.

Barely two months prior to the incident, the Government had been busy installing video cameras in the homes of people placed under Sars Quarantine Orders.

The Police have also announced that they will be setting up a network of 30 video cameras in Little India, Boat Quay and Newton Hawker Center to deter trouble-makers.

It seems odd for the principal of RJC to assert a right to privacy in the classroom when, by design, a classroom is filled with students who are supposed to be taking notes. The ‘privacy’ that the principal is talking about is really the freedom from accountability.

In the words of author David Brin: ‘Whenever a conflict appears between privacy and accountability, people demand the former for themselves and the latter for everybody else.’

In his book, The Transparent Society, Brin recognises that advances in technology are inevitable and that they result in ubiquitious surveillance.

In a controversial leap, he then argues that this may be a good thing and that we should welcome it but only if the surveillance is two-way.

If governments and corporations have the power to shine a torchlight onto the lives of individuals, citizens – all individuals -must also have the power to shine the torchlight back at the mighty and the strong.

The question facing society, then, is not whether it should ban cameras (a moot point as it’s not negotiable), but rather who should control the cameras.

Disconcerting as his arguments may be to privacy fundamentalists, Brin poses a key question: is the loss of privacy that technology brings acceptable – even desirable – if that same technology leads to greater accountability for those in power

In Singapore, what would happen if the Land Transport Authority (LTA) takes a different approach to privacy in Electronic Road Pricing and asks its CEO and board of directors to install satellite-based in-vehicle units (IU) in their cars, and to publish their real-time locations on a publicly-accessible website – just as the authority can monitor the movements of other motorists.

Changes in technology may be inevitable. But it is people who write the rules that govern how they use technology and where to delineate the boundaries of privacy.

To paraphrase Acting Minister for Manpower Dr Ng Eng Hen, individuals must be responsible in their use of technology – and the government even more so.

Ngiam Shih Tung is an engineer in a multi-national aviation company.

Sampling checks in GE 2015

The Elections Department released early indications of polling results for the first time during the 2015 General Elections by publicly announcing the result of sampling checks within 2 hours of the close of polls. Sampling checks have been carried out since 2001 or earlier, but the results were not made known to all candidates or the public until this year. The sampling check is carried out by Counting Assistants drawing a sample of 100 ballots from each counting table after opening the ballot boxes and mixing the ballot papers together. As each counting table corresponds to one polling district and there are about 2,000-3,000 voters per polling station, the 100-ballot sample per polling district corresponds to 3-4% of votes cast and is a large enough sample to make a good estimate of the final polling result

As shown in Table 1, the margin of error varies from ±1.2% points for a large GRC, Pasir Ris-Punggol with 66 polling districts, to ±4.4 % points for Potong Pasir, the smallest Single Member Constituency (SMC) with only 5 polling districts.  For simplicity, no adjustments are made to account for the variation in number of voters in each polling district, and the error margin is calculated for a 50:50 vote split.  The Straits Times has reported that the error margin for sampling counts is ±4% but this is a worst case and only applies to Potong Pasir which is the smallest SMC by far. On average, the uncertainty in the sampling check is within 1.3% points for GRCs and 3.3%  points for SMCs.

Table 1 – Error margins for selected constituencies and SMC and GRC averages
Number of polling districts Sample size Margin of error at 50% vote share
Pasir Ris-Punggol GRC 66 6,600 1.2%
Aljunied GRC 50 5,000 1.4%
East Coast GRC 32 3,200 1.7%
Bukit Panjang SMC 11 1,100 3.0%
Potong Pasir SMC 5 500 4.4%
GRC Average 45 4,500 1.5%
SMC Average 9 900 3.3%
Error margin is calculated as half the width of the two-sided 95% confidence interval (Normal approximation) at 50% vote share. All polling districts within a constituency are assumed to have an equal number of voters.

In a first-past-the-post election, however, what counts is not the actual vote share but crossing the 50% threshold. If we ignore three-cornered fights and make use of the normal approximation again, we can calculate a “victory threshold” or minimum sampling check result for which a candidate can be 95% certain of victory.  This is shown in Table 2 for selected constituencies.  Note that the victory threshold is slightly lower than would be obtained if we simply added the error margin from Table 1 to 50% because the victory threshold is a one-sided rather than a two-sided test. If you’re not statistically-inclined, don’t worry about it – that effect is small in this case. In Aljunied, despite the appearance of a nail-biting finish, the 52:48 sampling check result was above the victory threshold for the Workers’ Party and the PAP in fact had only a 0.2% chance of winning in that constituency once the sampling check result was known [See note 1].  In Punggol East, however, the sampling check result of 51% for PAP was lower than the victory threshold of 52.4% and the PAP could only be 75% certain that they would win there. Put another way, Lee Li Lian still had a 25% chance of winning even after seeing the sampling check results. In all the other constituencies, the winning party’s sampling check result significantly exceeded the victory threshold so the final result was not in doubt once the sampling checks were completed.

Table 2 – Victory threshold for selected constituencies.
No. of polling districts Victory threshold Actual PAP share in sampling check Probability of PAP win given sampling check result
Pasir Ris-Punggol (6-member GRC) 66 51.0% 73% 100%
Aljunied (5-member GRC) 50 51.2% 48% 0.2%
East Coast (4-member GRC) 32 51.5% 61% 100%
Sengkang West (SMC) 13 52.3% 63% 100%
Punggol East (SMC) 12 52.4% 51% 75%
Hougang (SMC) 9 52.7% 42% 0.0%
Potong Pasir (SMC) 5 53.7% 68% 100%
1. Victory threshold is defined as the minimum sample count result for which a candidate can be 95% certain of receiving over 50% of the actual vote (one-sided test).
2. Probability of PAP win is the probability that the actual PAP vote share is > 50%, given the observed sampling check result.

The error between the sampling check and the actual result gets smaller as the sample size increases. Hence GRCs will have smaller error margins than SMCs and 6-member GRCs will have smaller error margins than 3-member GRCs. This is seen in Figure 1 where the difference between the sampling check and the actual result was only 0.1% points in the 6-member Pasir Ris-Punggol GRC while the largest difference of 2.6% points was observed in MacPherson SMC.  The observed differences between sampling check and actual vote counts were all within the expected 95% error margins except for one constituency, but one out of 29 is about right, statistically. We also did not take rounding errors into account, which slightly widen the margin of error.


The sample check is a form of a quick count, which is used in developing democracies where there are concerns with regards to the compilation of electoral results by the central government. In Singapore’s case, there is no obvious need for a quick count as the entire counting process can be observed by candidates’ counting agents and elections results have always been announced within a few hours. Nonetheless, given that the Elections Department has chosen to conduct sampling checks, the decision to publicly reveal sampling check results is a welcome one.


[1] Some additional uncertainty is caused by the sampling check results only being reported as whole number percentages. The true PAP sampling check result from 5,000 samples could have been as high as 48.5% rather than the reported 48%.  However, this would still have given the PAP only a 1.8% chance of victory once the sampling check results were known.


Sampling check results  and actual vote counts are tabulated in Sample check vs actual v3 (Excel format).

A version of this note was previously published on The Online Citizen,