Observing Malaysian Social Media

Archive for January 2013

BFM Interview: Crowd-counting

BFM recently interviewed us regarding the crowd-counting methodology as part of their ‘The Week in Review’ show. You can listen to the podcast at the link below (opens in a new tab):


Timeline (minute:second)

  • 0:00 Crowd-counting (Ahmed Kamal – Politweet; Gabey Goh – Editor, Digital News Asia)
  • 6:20 Lese Majeste Laws of Thailand
  • 11:57 Graffiti Artists in Myanmar
  • 16:50 The Phenomenon of Cybertroopers (Tessa Houghton – Asst.Professor in Media & Communication, Nottingham University; Tun Faisal Ismail Aziz – Chairman, UMNO New Media Unit; Praba Ganesan – PKR Social Media Strategist)

Written by politweet

January 31, 2013 at 1:40 pm

Punggol East By-Election, January 26th 2013

A by-election for the Punggol East Single Member Constituency in Singapore was held on January 26th. It was won by Lee Li Lian (Workers’ Party) with a total of 16,045 votes.

The campaign started on Nomination Day on January 16th and ended on January 24th. Polling Day was on January 26th. We tracked mentions of candidate names and #punggoleast from January 20th onwards. Due to a lack of familiarity with Singaporean politics online, we did not track mentions of other hashtags or candidate campaign accounts (if any). Unless otherwise stated all statistics include tweets mentioning candidate names, Punggol East and #punggoleast.

The candidates for the by-election were:

  • Lee Li Lian (Workers’ Party, 16,045 votes)
  • Koh Poh Koon (People’s Action Party, 12,875 votes)
  • Kenneth Jeyaretnam (Reform Party, 353 votes)
  • Desmond Lim Bak Chuan (Singapore Democratic Alliance, 168 votes)

Twitter Statistics

The total number of users and tweets about the by-election are:

Users: 8,519

Tweets: 27,185

Searchterm Popularity


This chart shows the number of users and tweets mentioning each searchterm that we tracked between January 24th – January 28th. Note that Koh Poh Koon ranked lowest in tweets and users, but obtained 43.7% of the vote. Desmond Lim was the 2nd-most mentioned candidate but obtained the least number of votes.

Searchterm Users Tweets
#punggoleast 5383 16210
punggol east 3361 6903
Lee Li Lian 2395 3930
Desmond Lim 1389 2216
Kenneth Jeyaretnam 556 897
Koh Poh Koon 485 885

Tweets During Campaign Week


This graph shows the daily number of users and tweets about the by-election. The average number of users tweeting during the campaign was low, with an average of 381 users and a max of 591 users. There are over 20,000 active Twitter users in Singapore, so this reflects a low level of interest in the campaign.

Singapore’s election system has a Cooling-Off Day prior to Polling Day where most forms of campaigning are not allowed. That is the reason for the dip in tweet levels on January 25th.

Interest peaked at 16,625 mentions by 6,194 users on Polling Day.

Tweet Map


This map shows geo-located tweets from Singapore on January 26th. Blue dots indicate regular tweets (e.g. conversations, check-ins, opinions etc.) while Singapore flag markers indicate tweets talking about the by-election (e.g. candidates, #punggoleast). The number of Singaporean users tweeting about the by-election was small, but they were widespread throughout the country.

Tweets-per-minute, Polling Day


This graph shows the tweets-per-minute (TPM) about the by-election. Tweets peaked at 679 TPM from 579 users at 11 PM, which was shortly after results were confirmed.

Candidate Popularity


This graph shows the daily number of tweets mentioning each candidate during the campaign week.

Desmond Lim had a strong lead on January 21st due to his online rally, however interest steadily declined. By January 23rd Lee Li Lian was in the lead, before being overtaken by Koh Poh Koon on January 24th.

On Polling Day, mentions of Desmond Lim were high as speculation of him losing his deposit spread. Mentions increased further after he left the counting centre. Mentions of Kenneth Jeyaretnam were high as speculation of him losing the election spread.

Top Tweeple

The most talkative users about the by-election are listed below (username, tweets). Retweets are not included.

  1. @occupysg, 283
  2. @sgsilentmaj, 213
  3. @channelnewsasia, 137
  4. @todayonline, 134
  5. @yahoosg, 98
  6. @dk, 63
  7. @power98news, 58
  8. @izreloaded, 57
  9. @omysg, 52
  10. @singapolitics, 48

Top Tweets

The most retweeted tweets on Polling Day were:

1. Singapolitics, 627 RTs

2. Channelnewsasia, 257 RTs

3. Singapolitics, 250 RTs

4. Singapolitics, 236 RTs

5. Channelnewsasia, 231 RTs

*RT counts only include retweets made on Polling Day

Written by politweet

January 31, 2013 at 2:01 am

Facebook Census Update (Jan 2013): Pakatan Rakyat versus Barisan Nasional

In our December 2012 Census of Facebook Users in Malaysia, we included a comparison of likes for Pakatan Rakyat (PR) and Barisan Nasional (BN) for users aged above 21 years old. This did not explicitly include any politicians’ Pages.

We found that:

  • PR has more likes than BN in every age group.
  • PR has more male and female likes than BN in every age group. However the gap is narrow for female likes, which means BN can overtake PR in likes from potential women voters.
  • BN likes have grown at a faster rate than PR in November, narrowing the gap between both coalitions.
  • In November- BN’s highest growth was in men aged 21-30 years, while PR’s highest growth was in women aged 21-30 years

On January 1st 2013, we ran another census for PR and BN. We found that:

  • PR has more likes than BN in every age group.
  • PR has more male and female likes than BN in every age group.
  • BN likes grew at a faster rate in December compared to November, for both men and women.
  • BN gained more likes than PR in every age category for both men and women.
  • PR likes grew at a faster rate in December compared to November, but only for men. For women, PR’s growth rate reduced from 14.98% to 3.29%.
  • For women aged 21-30 years, PR growth was 14560 likes in November. In December, PR growth was only 1520 likes. This reduced growth rate applies to all age categories of women.
  • In December – BN’s highest growth was in men aged 21-30 years, while PR’s highest growth was in men aged 21-30 years
  • The gap for female likes has reduced further, with BN only 2160 likes away from overtaking PR for women aged 21-30 years.
  • PR’s fan-base has become more male-dominant, with the male/female ratio shifting from 209:100 to 217:100.
  • BN’s fan-base has become less male-dominant, with the male/female ratio shifting from 154:100 to 152:100.

Questions to think about:

  1. What did PR do in November to gain such a high increase in likes from women?
  2. What did PR do in December to increase its growth rate in likes from men, but greatly reduce its growth rate in likes from women?
  3. What did BN do in December to increase its growth rate in likes from both men and women?


Barisan Nasional

Total likes: 428,540

Male/female ratio: 152:100

Pakatan Rakyat

Total likes: 606,820

Male/female ratio: 217:100







Facebook’s advertising tools were used to collect data on the number of likes for each topic. The results are an estimate by Facebook based on the search criteria we used. We found the margin of error to be +/- 0.1%, though this can vary based on the topic.

Written by politweet

January 29, 2013 at 1:06 pm

Himpunan Kebangkitan Rakyat #KL112 Crowd Estimate

Himpunan Kebangkitan Rakyat (People’s Uprising Rally / #KL112) was a protest rally held at Stadium Merdeka, Kuala Lumpur on January 12th, 2012. It was organised by Pakatan Rakyat and various NGOs.

At its peak the crowd is estimated to have been between 63,976 – 78,193 people. This number is not expected to be revised, unless some new evidence surfaces to suggest a different crowd size.


Photos and videos that showed the size of the crowd were evaluated. These were used to determine the extents of the area covered by the crowd and the crowd density. The area was then refined by removing obstacles and estimating the percentage of space used. A crowd density of 4 – 12 sq.ft. per person was used.

Building the polygons

After initial evaluation of the photos, one was chosen as a reference to build the polygons from. The polygons were then adjusted after evaluating photos and videos. The steps are shown in the diagram below:


  1. The original photo
  2. Satellite photo, perspective-adjusted
  3. The original photo, perspective-adjusted to match the satellite photo
  4. The completed polygons on the map


The polygons in the map below shows the extents of the crowd, divided into different zones. The total space covered by the polygons is 456,627.05 sq. ft. The space taken up by the crowd is 275,532.03 sq. ft. These measurements do not include the stadium seating.


The estimated crowd in each zone is:

A = 20,122 people

B = 1,188 people

C = 8,652 people

D = 2,474 people

E = 7,428 people

F = 1,221 people

Stadium seating = 30,000 people

Total = 71,085 people

Margin of error is +/- 10%.

The estimated size of the crowd is 63,976 – 78,193 people.

Read the rest of this entry »

Census of Facebook Users in Malaysia, Dec 2012

View this document on Scribd

The following is the executive summary, included in the Scribd document above.

Executive Summary

This document describes the population of Facebook users in Malaysia based on public information collected from Facebook.  The goal is to provide a point of reference for social media marketing and a report on Malaysian interest in politics.

Personal characteristics and interests in topics such as Pakatan Rakyat (PR), Barisan Nasional(BN), coalition leaders, Bersih, 1Malaysia and both mainstream media (MSM) and online alternative media publications were measured by age group and gender.

Users interested in a topic are considered the fan-base for that topic. Research focused on potential voters (users aged 21 years and above).


Malaysia has a total population of 13.5 million Facebook users. The population is male-dominant – 53% are male and 47% are female.

Out of the total population, 9 million are potential voters – 54% are male and 46% are female.

Results of the analysis show that men are more interested in politics than women, but the overall population has little interest in politics. Only 8.8% of total potential voters expressed interest in PR and BN parties.

The research drew attention to a possible relationship between the male-female ratios of each topic’s fan-base:

  • Female bias was present in the fan-base for Malay-language MSM, Chinese-language MSM.
  • Female bias was strongest in the fan-base for 1Malaysia.
  • Male bias was present in the fan-base for English-language MSM and Barisan Nasional parties.
  • Male bias was strongest in the fan-base for Pakatan Rakyat parties, Bersih and online alternative media.

For female-dominated topics, there is 1Malaysia and Malay-language MSM. The 1Malaysia brand is associated with BN and Malay-language MSM tends to give more coverage to BN parties.

For male-dominated topics, there is PR, Bersih and online alternative media. Bersih is a coalition of non-governmental organisations (NGO) seeking electoral reform. Bersih is strongly supported by PR. Online alternative media tends to give more coverage to PR parties compared to mainstream media.

PR’s fan-base is 1.5 times larger than BN’s fan-base. However the gap between BN and PR is small when it comes to women and PR’s fan-base is more male dominant compared to BN. This means that BN can overtake PR when it comes to potential women voters.


Population growth in November showed that both BN and PR’s fan-base were becoming less male-dominant. However BN started with a more favourable position in terms of gender balance.

Both 1Malaysia and Malay-language MSM have a female-dominant fan-base, which is a good market for BN to draw users from.

PR does not have a female-dominant source to draw from. Bersih and online alternative media are too male-dominant. The main source left to draw from are women who currently do not show an interest in politics, which is a challenge faced by both coalitions.

This places BN in a better position than PR to increase its share of interest from potential women voters.

Written by politweet

January 16, 2013 at 12:54 am