My fascination with numbers

Illustration: TBS

Since my childhood, I have been attracted to and fascinated by numbers. I used to love mathematics. My fascination and love for numbers were enhanced by Professor Shukhendu Shome, who was a teacher of mathematics and a colleague of my father at BM College, Barisal.

At that time, I was perhaps 10 or 11 years old. Whenever Professor Shome came to our house, he would play various mathematical tricks with me. For example, through some algebraic manipulations, he would prove that one equals two.

From the multiplication table, he would choose the series with the number 9. Then he would show that the additive figure of the multiplied numbers would also be 9. Say, 9 times 4 equals 36. Now if you add the digits 3 and 6, the total would be 9. Similarly, 9 times 6 equals 54— adding 5 and 4 also gives 9.

This is enough to spellbind a boy aged 10. Let me confess that after so many years, while I have been teaching Economics at Dhaka University, I amazed my students with similar tricks.

In college, I had Professor Shukhendu Shome as my mathematics teacher. He did mathematics at the speed of a tornado, with numbers racing from one end of the blackboard to the other. They would disappear before we could even take note of them. Professor Shome would whip his numbers just like a jockey whips a horse.

However, I was introduced to statistics when I entered the university. At the higher secondary level, I was a student of science. At the bachelor's level, I had economics as my honours subject— so naturally, I took mathematics and statistics as my subsidiary subjects.

As our statistics teachers, we have had such stalwart professors as Shaheed Professor Muniruzzaman, Professor Golam Mostafa, Professor Mahbub Ahmed and Professor Abdullah.

When in the 1970s, I went to McGill University, Canada, for higher studies, my fascination for statistics increased by many folds. For my PhD research and thesis, the topic I chose was building an econometric macro model for Bangladesh.

I cohabitated with numbers for the next three years when numbers, data, statistics and models became a major part of my life.

In the 1980s, while teaching at the Department of Economics, Dhaka University, I became more immersed in data and statistics, as I undertook different surveys and research.

Two things must be mentioned in this context. First, at that time, I was a frequent visitor to Mohiuddin and Sons, a bookshop in New Market. They used to keep various journals on statistics, government publications on surveys and statistical publications of the Bangladesh Bureau of Statistics (BBS).

Going through all these publications, I became more aware of numbers and more appreciative of statistics. I became familiar with the multidimensionality of figures and numbers. Second, during that decade, I served various advisory bodies of BBS. When I became an adviser to the Bangladesh Planning Commission during the second half of the 1980s, my collaboration with the BBS became more intensive.

During that time, along with analytical research, I also leaned towards practical policy research on various development problems in Bangladesh. So, I became linked with different research organisations both at home and abroad.

For research work, I had to travel to various parts of the country with my research teams. We had to collect field-level data through surveys to build a reliable and robust database. That work provided me with a renewed introduction to the realities of statistics.

But it was my teacher and colleague, the late Professor Musharraf Hossain, from whom I learnt the nitty gritty of practical research. I used to be amazed by his deep knowledge about poverty and malnutrition in rural Bangladesh as well as by his feeling of numbers. It seemed to me that numbers and data talked to him. From him, I have learnt to explain numbers, to draw conclusions from them and how to integrate statistics into policy recommendations.

My love affair with numbers reached its pinnacle when in the 1990s, I joined the Human Development Report Office at the United Nations Development Programme (UNDP). As the Deputy Director of the Office and as a member of the Report's research and writing team, headed by Dr Mahbubul Haq, my work involved analytical research, construction of the data structure and formulation of policies.

I was also engaged in building the data series and constructing a composite index— the Human Development Index (HDI) for 189 countries of the world. 

Through all these works, I came to realise that data and statistics are not only theoretical or practical issues, they are also highly sensitive political and diplomatic concerns.

Different countries looked at the data contained in the Human Development Report from their own political perspectives. They used to put pressure to change various numbers pertaining to their respective countries so that they appeared better in the HDI ranking.

In that respect, their main focus was on the HDI and in order to improve their position in the ranking, they used to inflate their national data artificially. Their sensitivities in these areas were extremely high.

It was not easy to juggle all these forces. But Dr Mahbubul Haq was a superb juggler and an exceptional number-magician. During the preparation phase of the HDRs, while heaps of numbers were presented before him, Mahbub would intently stare at them with fixed eyes.

Then suddenly he would put his finger on a number and would say either, 'This is not correct' or 'This number may cause political trouble.' The amazing thing was that he had always been right.

I understood that if you loved numbers, were committed to them and if you have been dealing with data for a long, you could acquire such skills. With regard to numbers, I received two types of training from two stalwarts— Professor Amartya Sen and Dr Mahbubul Haq.

Professor Sen used to emphasise the purity of numbers and in a way, he was totally uncompromising on the purity issue. On the other hand, Dr Mahbubul Haq was more flexible and would stress on the practical use of data. I feel that both are important and striking a balance between the two is desirable.  

Even after I retired from formal professional life, my friendship with numbers has not diminished at all. During the last five years, I have authored three National Human Development Reports (NHDRs) — those of South Africa, Nepal and Bangladesh. For Mozambique and Somalia, I have prepared the Poverty Eradication Programmes and for St Kitts and Nevis, I have drafted the country's 15-year perspective plan.

Through such work, I have become familiar with various statistical frameworks of different countries— their processes of data collection, data presentation and subsequent analysis.

I have developed a network with the statisticians of those countries, enriched myself through discussions and debates with them, and understood their respective views and positions. Through a comparative analysis, I have broadened my views on different issues. 

During my long professional career, in different contexts, with my fascination with statistics, I have come to realise certain aspects of statistics and numbers.

First, statistical frameworks must have a solid theoretical and analytical base. Data are not only numbers, but they also have theories behind them.

Second, the purity and honesty of data are significant for two reasons— one, to reach a correct conclusion, and two, to earn people's trust and confidence. The data, which are not deemed reliable by people, do not have utility either. Data must be disseminated among a wider population so that everyone has access to them. For this, there should be a 'data availability and transparency' law.

Third, data should be used in two ways— one, to raise public awareness, and two, for policy-making. In both contexts, the role of historical data series is important. In order to formulate policies, policymakers must have data literacy. And they must be aware of the importance of the necessary statistics.

Fourth, it is true that politicians use data for their interests. However, the process of data mobilisation and construction must be kept outside of politics. And the state statistical agency must be autonomous so that it can carry out its mandate independently. The work of such entities should also be decentralised.

Fifth, it is critical to collect disaggregated data as much as possible. Inequalities or disparities in a country are not reflected in overall data or in averages. Only disaggregated data can capture them. In this context, the gender inequalities and disparities are worth mentioning.

Sixth, data collection and preservation must take advantage of advances in information technologies. Attention should be given to necessary funding, training and infrastructure building.

Finally, a statistical framework and agency cannot be static, it has to be dynamic. It is essential to be familiar with new research, new knowledge and processes through exchanges of experiences. A statistical agency, therefore, needs to collaborate with universities, academic institutions and research organisations. 

In conclusion, numbers are power. And numbers also talk. But if that talk is true or false, depends on whether the data presented have been constructed correctly, honestly and objectively.

Data are often influenced by political considerations, pressures, or the desire to appease political structures. Such practices, of course, destroy the honest objectives behind statistics. Through my long-standing fascination with numbers, this is a hard truth that I have learnt over and over.

Sketch: TBS

Selim Jahan is the Former Director of the Human Development Report Office and Poverty Division, United Nations Development Programme, New York, US

Disclaimer: The views and opinions expressed in this article are those of the author and do not necessarily reflect the opinions and views of The Business Standard.