Data-driven studies will lead to better guidelines


IT is important to get more people to contribute to the ongoing Malaysia Population and Housing Census (Census 2020), a 10-year survey by the Department of Statistics (DOSM) to improve data-driven policymaking, highlight public needs and efficiently channel national resources.

The census, which was originally scheduled to close in the middle of last year, has been restructured and its timeframe has been extended due to the extended movement control order put in place to contain the spread of Covid-19.

The pandemic has posed logistical challenges to the 9,500 censorship counters who make home visits across the country to conduct interviews. Although the digital e-census exists, Malaysians have traditionally preferred to answer face-to-face questions, which is why the e-census needs to be adopted even better. With door-to-door interviews temporarily suspended during the current lockdown, public adoption of the e-Census is urgently needed to complete data collection of the estimated 32.7 million population by July.

One way to get more people in Malaysia to participate in the e-Census and other data-driven research is to highlight the long-term benefits of participating in these studies.

Public consent is key to a good census, says Dr. Rachel Gong, Senior Research Associate at the Khazanah Research Institute (KRI). She explains that an improved public understanding of policy making and how the data-driven policy process can help identify the needs of a community that can make people more willing to contribute meaningful data.

This knowledge can also encourage the public to make more data available as it will enrich current studies and support further research that can improve living standards.

“There are many challenges to collecting this type of data, but I think it’s worth the investment and effort to get it right, especially for the census. But not just the census – when DOSM does its labor force surveys or household income surveys, I hope people understand why these are important and why they should take part, ”she explains.

Dr. Gong emphasizes that for good policy making, Malaysia must collect representative, reliable and valid data from the census and make this anonymized data publicly available without compromising individual privacy.

“It is important that everyone understands more about their personal data and how they handle personal data. If people are willing to give up their data for services like optimized driving directions in traffic jams, then there is no reason not to wholeheartedly join the census. “

Dr. Gong points out that there are different types of data that are collected and used for different purposes, including by the private sector, and that we as a society need to think more seriously about responsible data management in both the public and private sectors.

KRI, a think tank under Khazanah Nasional Berhad, has used the census data in most of its top reports, which include policy making recommendations.

“The most direct use of the census is to produce descriptive statistics on population demographics,” said Gregory Ho, research associate at KRI. He adds that the census data has also contributed to previous KRI projects such as those on social mobility.

Improving data accessibility

One of the biggest problems with government data in Malaysia is its lack of accessibility, and this is largely because there is no overarching law promoting the transparency of government data, says Ashraf Shaharudin, research fellow at KRI.

While Ashraf notes that DOSM has increased the volume of published data over the years, there is still room for improvement. He adds that one of Malaysia’s first steps is to enact a right to information law that will move the culture of government data from default secret to default open.

“This law should encompass digital technology; By requesting data release in a machine-readable format whenever possible and appropriate, and by setting up a website where users can submit and track requests for information. At the same time, the government should enact a data protection law that includes government data, ”he says, addressing the privacy concerns.

Protection of privacy in data management

The census is important in setting the context for what we know about a country’s population and housing stock, says Kenneth Simler, senior economist at the World Bank.

“Through the census, we know some basic facts – such as the male / female ratio, the urban / rural ratio or the proportion of the population in different age groups – that help frame concrete political analyzes,” explains Simler.

As such, the World Bank indirectly uses the census all along in its policy analysis, most recently in the report on Aging in Malaysia, A Silver Lining: Productive and Inclusive Aging for Malaysia. Another indirect use of the census for policy analysis is that the census provides the sampling framework for detailed surveys like the Household Income Survey (HIS) or the Labor Force Survey (LFS), says Simler, who adds that those sample surveys might not be substantiated with the census they make very misleading statements.

Simler notes that Malaysia’s statistical system, like Ashraf, is relatively well developed, but also recognizes that there is room for improvement.

“By far the biggest data-related challenge facing the World Bank in Malaysia is insufficient access to survey data. Inadequate access to survey microdata has severely limited the policy analysis and advice the World Bank can provide to the government, ”he says.

“Access was granted occasionally, but more often it was not granted. In advanced economies, and even in many low- and middle-income countries, access to microdata is much more open. “

With regard to data protection, Simler explains that greater data access can be compatible with data protection and confidentiality protection. For example, there are many best practices for anonymizing survey data that are both rigorous and easy to use and are used by national statistical authorities around the world to protect confidential information.

This can be done with more commitment and implementation of open data principles while protecting the privacy and confidentiality of individuals and companies, says Simler.

The future of data in Malaysia will require modernization, better use of data management technology and coordination between the many public and private entities that make up the national statistical system, he added.

If we invest in and focus on a cohesive data system, we will be able to create far-reaching and comprehensive policies that will bring the greatest benefit to the largest possible number of people.

Robust data ecosystem

High quality and accessible data will lead to better policy and strategy design. One of the biggest challenges for researchers is getting up-to-date information from government agencies, industry associations, community organizations and others in the ecosystem, says Prof. Mahendhiran S. Nair.

While Malaysia has made advances in data acquisition since 2014 due to the open data initiative, existing laws like the Official Secrets Act still hinder data sharing, says Mahehdhiran, Pro-Vice Chancellor (Research Engagement & Impact) at Sunway University.

“A lot of authorities are reluctant to share data because of the OSA, and many of them don’t know what is confidential and what is not. Therefore, many of them apply the blanket rule that prohibits sharing much of the data generated by their institutions.

“There is a need for a better system to ensure the security of sensitive information; However, this should not limit the dissemination of other information that is relevant to the public, industry, the research community and other stakeholders, ”he says.

The lack of standardization in the collection and reporting of data by the various authorities and institutions should be countered by improving technology and infrastructure.

It’s also important to retain data science talent, as their expertise is in high demand locally, and put in place appropriate regulations to ensure transparency and accountability among the authorities generating the data, Mahendhiran says.

“Big data will change the way we design policies, business strategies, and intervention programs. To make this possible, a robust and solid data ecosystem is crucial. “


Comments are closed.