GOOD NEWS, FAKE NEWS AND STATISTICS – NBS

0

Keynote Address By
DR. YEMI KALE
Statistician-General of the Federation / Chief Executive Officer
National Bureau of Statistics
At the
LAUNCH OF DUBAWA ONLINE VERIFICATION & RATING PLATFORM
PREMIUM TIMES CENTRE FOR INVESTIGATIVE JOURNALISM (PTCIJ)
ABUJA

20 FEBRUARY 2018

PROTOCOL

1. Introduction

I am privileged, indeed honoured, to join you here today at the launch of DUBAWA, an initiative of the Premium Times Centre for Investigative Journalism, to deliver this address I refer to as ‘good news, fake news and statistics’.

DUBAWA, I am told, takes its name from the Hausa word for “check”, and is an online platform dedicated to fact-checking the news media and public information, as well as supporting the journalism profession with tools to ensure veracity and accuracy in reporting. Indeed, this initiative and its noble objectives could not have come at a better time, given the amount of disinformation and misinformation that pervades our society in recent times. As we get closer to elections, such disinformation, misinformation and misinterpretation of data will, unfortunately, grow even stronger.

The proliferation of information and communications technology has become a double-edged sword in today’s society: bringing life-saving benefits at the same time
significantly raising the risk of international conflicts. Therefore, there is a responsibility on citizens and governments to evolve appropriate means and measures to address this challenge before irreversible damage is done. In my opinion, therefore, a mechanism that reduces the level of information asymmetry, like DUBAWA, is a significant step in this regard and I commend the promoters for taking this bold step.

I have been asked to “share knowledge on the need for data-driven projects and the usefulness of facts and data in all sectors of the country, including the media.” In my remarks, therefore, I will touch on the role data plays in this new information-driven world we find ourselves; what’s driving the demand for data, our experience at NBS with respect to improving policy decision-making through data, and the innovation that we have introduced in recent years to this end. I conclude with some thoughts on values which have guided our work at the NBS and would be relevant to this initiative. In doing this, I intend to demonstrate that the emergence of fact-checking, data analysis and polling organisations can be considered a natural evolution and consequence of recent institutional developments that have been made to improve the production and dissemination of data in the country.

2. There is growing demand for data on the Nigerian economy

Distinguished ladies and gentlemen, we find ourselves today in a global information society where the flow of data and information is constantly evolving. Data is playing a major role in shaping almost every aspect of human life and there is hardly any area of life data hasn’t been used to transform lives.

But what do I mean when I say data? When we hear data, we often immediately think of numbers and figures. But today the term data refers to so much more. Data can be in the form of images, words, figures, facts or ideas. It is very simply a collection of facts that have been translated into a form that provides information. The value of data is becoming more and more apparent as we continue to move towards an information-driven economy where data has become the new currency.

Data is vital as it provides clear, objective, and numerical evidence on all aspects of our lives and the state of our country, including the growth and characteristics of our population, economic performance, levels of health and wellbeing and the condition of our surrounding environment. It aids the decision-making process by enabling us to establish numerical benchmarks and to monitor and evaluate the progress of policies or programmes; which in turn ensures that our policy interventions are well designed, effective, and highlight any areas which require improvement. Data put information in the right perspective so that it is ready to be used in decision-making, therefore the significance of statistical information in supporting evidence-based decision making cannot be overemphasized. Without these, we cannot make well-informed decisions that will catalyse our socio-economic development and transform the future of our generation and the generations of the future. It is when we can collate, understand and interpret data correctly, as well as identify key areas in our society or our economy that require change, that the policy prescriptions and direction of our governments and businesses are more likely to respond to the real needs of our communities.

However, data in itself cannot be understood on its own; to get information from the data one must interpret it into meaningful information. While misinformation often comes in the form of promoting false data to the public, a fast-growing trend is the misinterpretation of data. As getting data has become easier, it is becoming easier to call out fake data. Accordingly, an increasing trend which DUBAWA will have to contend with is with respect to those that deliberately try to misinform by taking the true data but twisting its meaning to give a false conclusion.

Such misuse of data can also not be ignored. Some misuse statistics because they are neither willing nor prepared to face the reality and truth exposed by data. This is not a problem of statistics, but rather a problem of the ill-motive of the users of the data. Just as a knife can be utilized to save lives if it is used by a surgeon, it could very well be a harmful instrument in the hands of a criminal. Thus, statistics are ‘innocent’, and, hence do not deserve any wrong labelling. Blames must go instead, to the ones who abuse or misuse it, particularly those who do so knowingly or selectively.

3. The proliferation of data has changed the way we do things

What did life look like before the emergence of this information age? For many of us here, in our early days, we would get our news either from transistor radios if you were an artisan or a trader in the market, or from the radio in your car if you were in the middle class and owned one, or watch foreign news on DSTV if you could afford a satellite dish. If none of this was available, you could always join the F.R.N. – not to be confused with the Federal Republic of Nigeria, but the “Free Readers Association of Nigeria”. As an FRN member, you could read the dailies at most local vendors for free. High-quality, trustworthy and reliable news articles from top newspaper houses of the day – The Guardian, Daily Times, Sketch, Tribune, Concord, Tempo or magazines like Tell or Newswatch. The news was available to virtually anyone, even in local languages in some cases. The kind of news you got through the phone was often from relatives or neighbours, who probably have access to the same or similar sources of news that you do: Voice of Nigeria, BBC/VOA, NTA or the state-owned broadcasting corporations etc. And since the
universe of news outlets and sources were more or less limited, it is safe to say most people generally had access to the same largely reliable information at the same time—with a few days delay in some remote places. This was very much the situation of things before the emergence of this information age. Information asymmetry – which I use here to mean a situation where some portion of the public believed one thing and another portion believed another because they receive their information from different sources -was generally not the case. Let’s consider this period as the period of ‘good news’.

Obviously, today, things are no longer the same. Times have changed.

In this room, one can safely say that no two people will have exactly the same news applications on their phones, even when they are co-workers or siblings. The majority will likely have Facebook or Twitter, or Instagram, or Youtube and possibly some additional media sources on their phones. In today’s information age, we are all living in each other’s worlds even without leaving our own cubicles. News – and information more broadly – is now made available straight from distant lands into our hands. We can track the news of a burning skyscraper in Dubai, at the same time watch a live soccer match in London, or follow the discovery of new butterfly species in the Amazon rainforest. And yet, not only do we now have access to more news, but we also have access to different news. Each day, we are being formed and informed by different experiences and sometimes different realities. What we know to be true has become imperilled by overt and covert efforts to reshape narratives in support of private agendas which are more often than not against the public interest.

This new era, which we call the fake news era, is symbolized by the spread of false dubious information masked as “news” and even embellished with false data or twisted interpretation of data, enabled by various ideologies and cantankerous intentions and accelerated by modern technologies. Some say it is the new way future wars will be fought, in cyberspace, where bullets and bombs will become bytes and megabytes of data, and triggers will be keyboard strokes. Our shared reality about the world, events, facts and truth are turned upside down and re-shaped into an alternate reality that primarily seeks to promote bias and fear and sow seeds of discord in the society. Information asymmetry exists because those who create fake news know it is so, but those who consume it may not.

For this reason, entities that support the pursuit of truth, and facts-based journalism become the only remedy to ensuring that society does not disintegrate or implode under this circumstance. But for these entities like DUBAWA to be successful, there needs to be political and ideological neutrality as well as sufficiently trustworthy data that is reliable and accessible not only to such entities but to which they can refer their users. This is the role that data and NBS plays by hosting publicly-available reference data and its clear interpretation on nearly every facet of Nigeria’s socio-economic life so that even when fake news inevitably exists, fact-checkers and citizens can readily verify these by themselves.

4. What’s driving the demand for data?

I believe that the growing demand for data on Nigeria can be attributed to two main forces: an externally-driven force, or exogenous demand; and a domestically-driven force, or endogenous demand. With respect to the external/exogenous demand, these include demand driven by the interest of international investors to reap higher returns in Nigeria, given historically low-interest rates in developed markets. Another prominent source has been as a result of data-reporting obligations for the sustainable development goals (SDGs). In these cases, data on the Nigerian economy is required for investment decision-making, resource allocation, programme design, and even allocation of voting powers (e g. at the UN, World Bank or IMF). International pressures promoting good governance have also led to increased demand for local data. Indeed, we see this reflected in the institutional partnerships that Premium Times has been able to put together behind the DUBAWA Initiative that we are launching today.

At the same time, Nigerian citizens are also demanding good governance, and policymakers are increasingly using evidence-based policymaking approaches at Federal and sub-national levels. At least 20 States now have state statistical agencies or have enacted legislation to establish one. We saw this clearly during the last general elections, in which various campaigns made extensive use – in both good and bad ways – of data. Depending on their motive, one side took all the good data and rejected the bad while another took all the bad and rejected the good. Similarly, when a particular mall was about to be built in Abuja, the promoters came to NBS to inquire about household consumption data for residents within a circumference of 20km of the area. During the construction of the Abuja-Kaduna rail, some foreigners also sought data from NBS on passenger, freight and transport patterns in the Kaduna axis. This is what I call the endogenous demand for data, that is the demand for data coming from citizens, policymakers and business leaders.

As a nation, we are faced with many development challenges in health, education, security, employment, governance and so on, and measuring the success or otherwise of development interventions and initiatives can only be substantiated by accurate and reliable data interpreted correctly, which is arguably as important as, if not more than, the programmes or strategy itself. Thus, the combination of both endogenous and exogenous factors has created a scenario where there are immense attention and interest in Nigeria, as one of the major destinations for growth and investment opportunities. With this has come increased demand for information and the significant efforts over the last few years to improve the quality, timeliness and reliability of statistics. Using the analogy of “bad money driving away good money”, we cannot allow “bad information” in form of fake data, or fake news to take root and drive away “good data”. This is why at NBS we have made great efforts in recent years to improve data quality as well as our systems processes. If DUBAWA is to be successful and become a source of reliable information
and data, relying on good data produced and accurately interpreted is a must.

5. NBS has improved its process and operations to ensure data availability and data quality

In the last five years, the NBS has been actively participating in the major economic policy decision making organs. This ordinarily should not be remarkable, but it is based on the increasing recognition by key policymakers that policy failures and the phenomenon of abandoned projects will persist if they are not based on accurate and reliable data for policy formulation, project identification, or monitoring and evaluation.

Permit me at this point to do a quick fact-check of my own by illustrating with one instance when we have had to deal with an issue similar to what DUBAWA is meant to address, and the importance of building public trust and confidence. Based largely on ignorance, historical cynicism and distrust of ‘anything government’, some suggested back in 2014, that the revision of official statistics was undertaken to make the government of the day look good. Let us check this claim. In fact, it was the culmination of years of effort to improve the quality and reliability of our national statistical system and the resulting statistics that originate from it.

Take the unemployment methodology which attempted to modify the existing unemployment methodology calculations. The primary goal was to provide more information to domestic policymakers, and on a secondary note, to align our methodology closer to ILO guidelines for better international comparability. Our
previous threshold for the definition of unemployment was based on a Nigerian definition of full-time employment, which was an individual working 40 hours a week, typically 8-4, or 9-5. This implied that if a person worked 39 hours a week, the individual was classified as unemployed. Just to highlight why this was such an issue: in the UK context, this meant that the entire population of full-time workers in the UK working between 35 and 39 hours may have been considered as unemployed in Nigeria since an individual is considered fully employed in the UK if they work 35 hours or more. As a result, following consultation and recommendations from a wide array of stakeholders which included Premium Times representatives, by the way, we lowered the time-based threshold for determining who is unemployed, from 40hrs a week, to 20hrs a week. We also further disaggregated already published unemployment figures, by separating those
working 0-20hrs (now termed as ‘unemployed’) from those working 20-40hrs (now termed as ‘underemployed’). Based on this disaggregation, it became clearer that Nigeria’s unemployment situation was really an issue of ‘underemployment’, in which case many job seekers tend to try their hands on low-skill, low-wage, non-full-time jobs, just to make ends meet in the absence of any unemployment benefits or social safety net system. This tells policymakers the magnitude of the problem, that although underemployment is higher, unemployment is actually growing faster than underemployment. This was not fake news at all. The truth is that if we had simply continued to rely on the previous 40hour threshold, policymakers would be none the wiser about the dynamics of the situation. Based on this information, disaggregated by gender, age groups, educational qualifications, state of residence etc, we believe policymakers can make more effective interventions. The same thing occurred with the rebasing and re-benchmarking of the GDP which aimed at using the more recent structure of the Nigerian economy to replace one that was over two decades old and was still considering Betamax videos and Nitel as the only telecommunications provider. That old structure did not even consider the emerging entertainment industry as existing.

But at that time, and even till today, several media outlets and influencers portrayed this as government efforts to tamper with official statistics. But routine methodology revisions are common to every serious country’s statistical authority. The Guardian (UK) in 2013 reported a similar incident in the UK when the ONS published results that showed over half a million jobs created but political opponents suggested this was government fiddling with employment figures. But the ONS maintained it had clear procedures for counting jobs created. Even in the USA, the current USA president claimed all the data produced was fake news but since becoming president now tweets the ‘remarkable’ progress the economy has witnessed using the same data by the same agency he suggested was false a couple of years earlier before he was elected. All of a sudden the data is now correct. Sounds similar to our circumstances in Nigeria, doesn’t it? In any case, as many of you are very much aware, NBS has always published data that is sometimes considered as “favourable” and “unfavourable”, so it is really difficult to support the notion that NBS somehow is being manipulated to produce one type of data to make any interest group look good or bad.

6. Innovations in data collection and dissemination at NBS

On our part, despite poor funding, we continue to address the binding constraints to the production and dissemination of reliable data by leveraging on two important pillars: collaboration and technology. First, through increased collaboration, openness and engagement with critical stakeholders, including other government agencies, business associations, academia and the media, NBS has developed a reputation of being able to undertake nationally representative data collection exercises with higher quality, speed and reliability. We have been forced to take this route also because funds are limited. To avoid duplication of outputs across agencies, we are harmonizing statistical activities across all sectors. Previously, each ministry/sector would want to undertake its own sectoral survey, contrary to the statistics law, independent of the statistics office and without consideration for how its data relates to other sectors and would apply whatever method suited them in a complicated and disorganised manner. But increasingly, the NBS is working with these agencies to jointly undertake such activities and to enforce harmonised processes and methodologies. It saves costs, ensures there are no conflicts in the eventual results and the overall system works in a coherent and coordinated way. In addition, we are able to deepen sector knowledge and skills for our staff, as well as the staff in the collaborating agencies. It also provides a unique opportunity for advocacy as we often involve representatives from these groups to observe and participate in the fieldwork and data processing aspects. I am sure some of the journalists here including Premium Times can confirm the times they have been on our survey teams at one time or another as observers.

We have also been able to improve the quality of our activities at every point of the data production process by taking advantage of the rapid diffusion of information and communications technology in Nigeria in recent years: from data collection to processing to results dissemination and even advocacy. For example, we publish on our website at the start of each year a data release calendar to intimate our users of the estimated release dates of our statistical products. This facilitates their own planning and stimulates demand for the products. To a large extent, we have met and sometimes surpassed the targets set by the calendar.

In addition, we have successfully tested, and now deploy more widely, computer-assisted personal interview (CAPI) devices to improve the speed, accuracy and reliability of field data collection. This initiative has allowed us to save costs considerably over the longer term. It also enables us to scale-up data collection more quickly and effectively as the need arises. In addition, it has also reduced errors and made truancy difficult. Since deploying the CAPI systems which can be uniquely identified and matched with a specific enumerator, we are able to monitor who entered what data, when and where through the GPS features. We can also track how long it took to complete the survey interview. The result has been better quality field data and faster data processing.

With respect to data processing, we have consolidated a number of our surveys to reduce duplication and, over time, will de-emphasise surveys altogether while strengthening the system of administrative statistics. This has been a key strategy to ensure greater data quality and reduce costs of data production. As we all know, surveys and censuses are very expensive. The revisions done to the GDP and unemployment statistics that I mentioned earlier were aimed at improving the quality of our data outputs, ensuring they are reflective of current socio-economic realities while still remaining comparable over time so we can evaluate our progress. This informed our theme of ‘Measuring Better’ during the 2014 GDP rebasing exercise.

Data dissemination has seen a number of innovations as well, driven largely by efforts to save costs and maximise the use of technology. All our regular and special publications are provided free, online on our website [nigerianstat.gov.ng], while the data is uploaded on the data portal, and ultimately archived in the National Data Archives (NADA). Our refurbished and fully-functional website provides numerous resources to serve as a reliable stream of data for your verification and validation efforts. In addition to our mailing list, we regularly provide our key reports to key government officials, while using social media to reach a wider, more youthful, tech-savvy citizenry. Through these efforts, we directly address the concerns of users before it becomes another form of fake news. It fosters transparency, makes users aware of available data, and enable them to utilise such data to make their decisions. We have also started publishing infographics so that the most interesting aspects of our reports are made easily available to those not able to go through tables in our reports. Lastly, we have developed mobile apps available for iOS and Android platforms.

Another current practice is to invite the press to ask questions about our data and how to correctly interpret it. Furthermore, we invite private observers to join our data collection exercises to monitor the process in line with our philosophy that the best way to engage with someone that doesn’t trust you is to open up completely to and not hide anything, so they can see for herself. This was the process we adopted during the GDP rebasing exercise and it was a successful strategy in renewing public confidence in our processes. For us, the opportunity to get feedback from our users is an important culture that we try to feed into the data production process, and a key benefit of the emergence of new technologies. Indeed, it is the feedback we get from our data users that helps us to plan our data release calendar for the subsequent year. These are some of the most visible innovation efforts at the NBS to address the misuse of statistics, fight fake news and provide the necessary data to support right policies in Nigeria.

I have taken the time here to lay out these points in order to establish that data operations at NBS are thorough and continue to improve, and DUBAWA and other users can always utilize the vast data resources at NBS to support its fact-checking, verification and validation mission but it must also build its reputation based on core values and principles which I now touch on as I conclude.

7. Conclusion – values for success

Our experience at NBS demonstrates that society rewards organisations that are identified as trustworthy and credible. This goodwill is the most important asset for an initiative like DUBAWA, and I urge the promoters to jealously guard this asset. In this respect, I would suggest certain institutional values – which I call the D.U.B.A.W.A values – that have helped us at NBS and which I believe will also be invaluable to you.

One, D for Diligence. Your fact-checkers should be diligent and devoted to finding out the truth in order to successfully combat misinformation and fake news. Otherwise, it might end up making a bad situation worse if they also get it wrong.

Two, U for Uncompromising. It is important that your fact-checkers remain uncompromising and unbiased, not only in what you choose to fact-check but also being transparent in the evidence you provide in verification. Providing a balanced assessment of the issue strengthens the platform’s credibility in the eyes of users. Remember: public trust is difficult to gain but easy to lose.

Three, B for Being bold and brave. I believe Premium Times already has some of the most fearless journalists in the country today, and one hopes the platform will also benefit from such talent.

Four, A for being Assertive. It is important that fact-checkers assert their independence, are confident in their work and able to defend their sources and methods regardless of their subjects whether individual or government, corporate or political.

Five, W for Wariness. This allows the journalist to be critical and sceptical and not always take things at face value; to be curious and inquisitive enough to uncover the truth behind claims.

And the last A is for Accuracy. This is the primary goal of the platform, for which all other values listed above exist. It should never be sacrificed no matter the cost.

Finally, ladies and gentlemen, I would like to use this opportunity to invite you all to join the emerging data revolution in Nigeria by supporting the NBS, State Statistical Agencies as well as this new initiative, DUBAWA which really is a commendable act of public service being undertaken by a private party.

On behalf of the Bureau, I congratulate the Management of Premium Times Centre for Investigative Journalism, reassure of NBS support and wish you good success in your operations. Thank you.