Our Database of Users in Malaysia

Are you interested in doing political research using Twitter data?

We have been collecting tweets related to Malaysian politics since 2010. Since then we have built a database of over 560,000 Twitter users based in Malaysia that have been active since 2021.

From May 2021 we have been collecting tweets from these users to learn what users in Malaysia tweet about besides politics, not just on a national level but by state and urban/rural areas. We have also mapped out the follower/following relationships between our users.

Work with us to leverage this data for use in your research or campaign. With this data you can:

• Find Malaysia-related content using common words or global brands
• Look at monthly trending content to better understand the Malaysian market
• Identify Twitter users who are popular with users in Malaysia based on criteria such as matching keywords or followers of specified accounts

We only provide data on a consulting basis based on the nature of your campaign. We do not provide any personal tweets or location-based history for individual users.

Our database enables us to monitor common keywords and filter tweets to get local Malaysian content.

The network graphs below show how users tweeting about a topic are connected by retweets and replies. By filtering the network against our database, the non-Malaysian accounts (highlighted in red) disappear. This makes it easy to see prominent influencers.

Here are some examples of filtering global tweets to find local content:

Without filtering, this was the top tweet in mentions of Mcdonalds between Apr-May 2019.

After filtering tweets against our database we can easily see top Malaysian content.

Learn more about Malaysian Twitter demographics, most-followed users by audience language and state in this report.

We have developed a generalised sentiment analysis system for Malaysian tweets which we make use of for our research and twitter monitoring services. In this example we show sentiment analysis for tweets mentioning then-Prime Minister Najib Razak in urban and rural constituencies.

This is effective for high volume conversations where there is little divergence of opinion. For deeper analysis we would recommend user opinion analyses.

We specialise in studying and evaluating conversations on Twitter to discover and measure opinions on political and social issues.

Our approach reveals insights that can’t be found in surveys or automated sentiment analysis.

What we do is examine individual user timelines to establish the contributing factors to their final expressed opinion on an issue.

We perform balanced sampling based on the distribution of users by state. This helps reduce the impact of high-population states on the final result, giving us a better understanding of popular opinion on an issue.

Some examples of our work include:

1. An Analysis of Opinions on Tun Mahathir, Trends in Political Interest on Facebook and Political Support by Malay Youth in Malaysia
2. Analysis of Opinions on Lim Guan Eng’s Corruption Charges by Twitter Users in Malaysia
3. Analysis of Support for the TPPA by Twitter Users in Malaysia

You can find more reports on our blog.

We have done detailed research on how Malaysians voted in previous elections using a combination of electoral results by polling lane, previous electoral rolls and social media.

Our analyses and statistics make it possible to measure support by race and age group for Federal constituencies (Parliament seats) in Malaysia.

By performing analyses between elections, we can measure the ‘swing’ or shift in support.

Some examples of our work include:

• Analysis of Federal Election Results & The Swing in Support by Race, Age Group and Level of Urban Development in Melaka in Malaysia’s 14th General Election
• Election Forecast for Pakatan Harapan in Peninsular Malaysia (GE14)
• Evaluating Voter Support in Kuala Kangsar and Sungai Besar Using General Election Results and Twitter
• Analysing Pakatan Rakyat’s Performance with Malay Voters in Peninsular Malaysia (GE13)

You can find more reports on our blog.