Unlocking the extraordinary power of statistics can work wonders with data, writes PROFESSOR OLUSANYA ELISA OLUBUSOYE.

Protocols

I am deeply grateful to Jesus Christ, my saviour and all those who have played a part in my journey to becoming a Professor of Statistics. I am also thankful to those who have taken the time to attend this lecture. No king has ever crowned himself; others must do it for him. A king is born first, then appointed, and finally crowned by those who hold the power to do so.

This lecture is of great historical significance as it commemorates the 50th anniversary of the Department of Statistics, University of Ibadan, as a full-fledged academic Department. It is also being delivered 38 years and 18 years after the Department’s first and second inaugural lectures by two esteemed Professors: Professors Biyi Afonja and Timothy A. Bamiduro. Professor Afonja’s lecture focused on “Facts, Figures, and Falsehood Syndrome in Society” (Afonja, 1985) while Prof. Bamiduro presented “Statistics and the Search for Truth: A Biometrician’s Perspective” (Bamiduro, 2005). Indeed, Professor Bamiduro could not predict the next inaugural lecturer from the Department but attempted to predict the year when he said, “I pray that the third time will come very much earlier than 2025”. This prediction is fulfilled today, 21 December 2023, two years earlier than 2025. Today’s 537^th inaugural lecture is titled: “From Data to Wonder: Unlocking the Extraordinary Power of Statistics.” This allows me to share my background, academic achievements, expertise, and contributions with the public after my elevation to the pinnacle of an academic career. Earlier in 2014, I delivered the first Faculty of Science Lecture from the Department of Statistics during the tenure of Prof. I.P Farai as the Dean of Science.

My journey to academia was divinely inspired and a dream come true. I remember when the vision to become a university teacher first emerged. It was during the convocation ceremony of my elderbrother, Pastor Abel Olufemi Olubusoye, who was the first graduate in our family. This happened in 1984, and I was in my final year at Eyemote Comprehensive High School in Iyin Ekiti, the hometown of the late Major General Adeyinka Adebayo. The convocation ceremony held at the University of Lagos that year was a proud moment for our immediate and extended family, and everyone was eager to be part of the delegation. I was fortunate enough to be included, which allowed me to visit the university campus for the first time. The experience there cemented my desire to become a university teacher, even though I had no clear idea about how to become a Professor.

Mr Vice-Chancellor, Sir, contrary to my initial ambition of becoming an Accountant, I received a direct entry admission to study Statistics at the Department of Statistics, Faculty of Science, University of Ilorin, after my GCE A/L at the Federal School of Arts and Science, Sokoto, Sokoto State. It is significant to note that I did not study core science subjects such as Physics and Chemistryin secondary school. My admission was based on my performance in English, Mathematics, Economics, and Geography. In today’s academic climate, it would have been impossible for me to study Statistics, let alone become a Professor of Statistics at the University of Ibadan. My experience as a Statistician has shown me that Statistics is both a science and an art. This will become evident in the course of this lecture.

When studying Statistics, I often encountered confusion about explaining my course of study to those who made inquiries. Statistics was not a popular course at that time. Indeed, only a few universities offered it, and, in most cases, it was offered as a combined programme either with Mathematics or Computer Science or both. During my vacations, my uncles and relatives would ask me about my course at the university. When I mentioned that I was studying “Statistics,” the next question would be, “What kind of course is that?” It was easy for my colleagues studying Medicine to say “ise isegun oyinbo,” and for those studying Mathematics to say “ise isiro,” and for those studying Engineering, it was easy to say, “ise imo ero.” However, I found translating my course, Statistics, into the Yoruba language challenging (Olubusoye, 2014). Even though this confusion persists, it further underscores the need to eliminate the divide between language and science. Hence, during this lecture, you will see how my expertise transcends statistics and applies to other fields such as economics, energy, etc.

For this lecture, I will address three crucial points. Firstly, I will emphasise the significance of data and statistics in today’s tech-driven world. Possessing statistical thinking, reasoning, and literacy can lead to incredible accomplishments and wonders. Secondly, I will discuss how Statistics is celebrated globally but not given much importance in Nigeria, including our university. Lastly, I will showcase the power of Statistics through my statistical collaborations, research projects, and professional activities. I will also highlight my contributions to developing Statistics and Data Science skills.

Meaning and Contextual Usage of “Data”

The word “data” is incredibly versatile and can be used in various contexts, each with unique nuances. A quick search on Google yields about 20,960,000,000 (20 billion, 960 million) results as of 01.11.2023, 7.24 pm! In fact, “data” isprobably one of the most used words today, even among children. It is common to hear phrases such as “my data has finished”, “borrow me data”, etc. Airtel Nigeria recognised the importance of the word “data” and launched a new campaign slogan, “Data is Life”, to reposition its flagship product, SmartSpeedoo^[1]. This new campaign was a resounding success for Airtel, earning over one million views across digital platforms within the first week of its unveiling (Vanguard Newspaper, 1 December, 2016).

It is important to note that the term “data” is used in many contexts, including but not limited to Information Technology, Research, Business, Database Management, Telecommunications, Geography, Philosophy, and Statistics. In information technology, data refers to raw facts, figures, symbols, or signals that have not been processed. Thus, data can be collected, processed, and analysed to extract useful information. Data can refer to any form of digital information in computing, including texts, images, sounds, and videos. Data encompasses the observations and measurements collected during research in research and science. This can be qualitative or quantitative. Scientists and researchers collect data through experiments, surveys, observations, etc., and then analyse it to draw conclusions or verify hypotheses. In the corporate world, data may connote sales, operations, finances, customer behaviours, and more metrics, which are critical to decision-making. In the realm of databases, data consists of distinct pieces of information, usually formattedspecially and organised within a database. In telecommunications, data can refer to signals transmitted or received, often differentiating non-voice or video transmissions from regular voice signals. In geographical information systems (GIS), data often refers to spatial information or satellite images representing specific characteristics of areas on the Earth’s surface. Philosophically, data can be viewed as the given facts available for reasoning and discussion. These are just a few contexts where the term “data” is employed. Given its broad applicability, the precise meaning of “data” can often be inferred best from its contextual usage.

From the perspective of a statistician or data scientist like me, “data” is a complex word that represents the product of a statistical procedure, including design, collection, editing, processing and dissemination (Olubusoye & Akintande, 2022). Data could imply the collection of raw (unprocessed) observations or measurements. These observations can take various forms, such as numerical values, categorical labels, or textual descriptions. Data can represent facts and figures that describe specific aspects of a phenomenon. These facts and figures can be gathered through various methods, including surveys, experiments, observations, or sensors. Data can imply empirical information derived from the real world. It is collected systematically and objectively to provide a basis for analysis and inference.

These definitions emphasise the diverse nature of data and its role as the foundation for statistical analysis, modelling, and inference. Depending on the specific context and the goals of analysis, statisticians and data scientists may work with different types of data and apply various statistical techniques to extract meaningful insights.

Data Revolution

There has been a significant and transformative shift in how data is generated, collected, processed and utilised in the last decade. This phenomenon is termed “Data Revolution”. The terminology emerged for the first time in the report of the High-Level Panel (HLP)^[2] appointed by Ban Ki-moon, then the UN General Secretary, to advise on the global development agenda after the 2015 Millennium Development Agenda. Data is becoming available increasingly faster, with greater volume and scope. The concept of “Big Data” is closely connected to the data revolution. It simply exemplifies the pervasive nature of data in our world. The “Big Data” concept emerged from the growing realisation that traditional data processing methods and technologies could not handle the increasingly vast and complex datasets generated in various fields.

Several key factors contributed to developing and popularising the Big Data concept (Van Belle et al., 2016). Firstly, the exponential growth in computing power and data storage capabilities in the late 20th century laid the groundwork for dealing with large volumes of data. The advent of powerful, distributed computing systems and storage solutions made processing and analysing massive datasets feasible. Secondly, the widespread adoption of the internet and the digitisation of various aspects of life led to the generation of enormous amounts of data. This includes weblogs, social media interactions, online transactions, and more. Thirdly, scientific disciplines, such as genomics and particle physics, started producing datasets of unprecedented size and complexity. Projects like the Human Genome Project and the Large Hadron Collider generated vast amounts of data that required new approaches to analysis. Finally, Tech giants like Google, Facebook, and Amazon began showcasing how they leveraged Big Data to improve their services.

Statistics and Statistical Practice

Statistics is a term commonly used in two ways (Olubusoye, 2012). In everyday language, statistics refers to numerical data, such as scores or records. It is now common to hear the person in charge of the scoreboard being referred to as the statistician in televised quiz competitions. Statistics plays an essential role in our daily lives. It appears in newspapers and magazines, on TV and in general conversations. We come across them when discussing the cost of living, unemployment rates, medical advancements, weather forecasts, sports, politics and even the state lottery. We may not always be aware of it but we are constantly gathering, organising and analysing information and using it to make decisions that will affect our actions.

In scientific discussions, however, statistics often refers to a discipline, a field of knowledge. The term “statistical science” could describe this discipline (Chatterjee, 2005), although it may create a wrong impression. Statistics is not a branch of science like Physics, Chemistry, Botany,or Geology. It does not have a specific domain as its field of study. It is domain-neutral like pure Mathematics, and its concepts and techniques, just like those of Mathematics, are used in other disciplines, including the natural and social sciences. My utmost desire in this lecture is for everyone to be statistically literate, not just statisticians. In everyday situations, being statistically literate can help individuals make better decisions to improve their quality of life. From the scientific perspective, using statistical methods and principles can accelerate social and economic transformation and drive the national economy. To sum it up, I wish to draw inspiration from the conversation between Paul the Apostle and King Agrippa in Acts 26:28-29.

Then Agrippa said to Paul, “Do you think that in such a short time you can persuade me to be a Christian?” Paul replied, “Short time or long—I pray to God that not only you but all who are listening to me today may become what I am, except for these chains.

Paul wished that all who were listening to him might become Christians. Similarly, all listening to me today may become statistically literate, except for the chains that bind Paul (Olubusoye, 2014).

Mr Vice-Chancellor, Sir, at this juncture, I will like to briefly discuss the best statistical practice for individuals who frequently use statistics in their respective fields. Statistical practice involves using statistical methods and techniques to collect, analyse, interpret, draw inferences and present data; it should avoid the following pitfalls.

• Manipulation to meet pre-conceived goals.
• Trivialisation of its principles to minimise the possibility of a rejection of the anticipated thesis.
• Indiscreet, frivolous, ill-advised modelling to justify fixed views.
• Pretentious applications involve highly technical, complicated modelling instead of a less demanding, appropriate, and adequate alternative.
• Oversimplification due to simplistic but inappropriate and inadequate modelling as an alternative to more complicated but truer modelling.

The best statistical practice involves the choice and application of principles, processes, determinations and inferences that statisticians and data analysts adopt to ensure that the findings are reliable, relevant, and appropriate to the immediate and general situation of the data.

Statistical practice can be broken down into several steps. Firstly, the process begins with careful planning of data collection. This step involves defining the research question, selecting appropriate data sources, and designing surveys or experiments yielding high-quality data. The methods used for data collection include surveys, experiments, observations, and data scraping from databases or the internet.

Data pre-processing or data screening follows after the data collection stage; it involves cleaning and transformation. Raw data often contain errors, missing values, and outliers. Data cleaning is, therefore, essential in identifying and addressing these issues to ensure that the dataset is accurate and complete. Additionally, data may need to be transformed or standardised to meet the assumptions of statistical models. This can be achieved through normalisation, log transformation, or standardisation. Exploratory Data Analysis (EDA) is also crucial in understanding the data. This is done through visualisations using charts, graphs, and plots to explore the data and identify patterns, trends, or anomalies. Moreover, summary statistics such as mean, median, variance, and percentiles are calculated to describe the central tendency and variability of the data.

At the core of statistical practice are modelling and inference. Statistical modelling involves hypothesis formulation to test relationships or differences in the data. It also involves selecting appropriate statistical models or techniques based on the nature of the data and the research question. Model parameters are then estimated using data and statistical algorithms. Additionally, model goodness-of-fit is evaluated, and its predictive power is assessed through cross-validation. Statistical inference involves hypothesis testing to determine whether observed effects or differences are statistically significant. It also involves calculating confidence intervals to estimate the range within which a population parameter is likely to fall and interpreting p-values to assess the strength of evidence against null hypotheses. The interpretation of the results of their analyses should be in the context of the research question, and the practical significance of statistical findings is explained. The best statistical practice adheres to ethical guidelines, ensuring the responsible and ethical use of data. This includes obtaining informed consent, protecting privacy, and addressing issues related to bias and fairness. Overall, statistical practice is a systematic and rigorous process requiring technical expertise, critical thinking, and effective communication. It plays a vital role in various fields, from scientific research and healthcare to business and policymaking, by providing evidence-based insights and informed decision-making.

Despite this, it suffers greatly from abuses by users in various disciplines. As identified above, statistical abuses, also known as misuses or pitfalls, are common errors or questionable practices that can lead to misleading or incorrect conclusions when working with data and statistics. These abuses can undermine the validity and integrity of statistical analyses. The common abuses include P-hacking, which involves conducting multiple statistical tests on the same dataset until a significant result is found, without adjusting for the increased risk of Type I errors. Most researchers often engage in data dredging, which is the practice of selectively reporting results that support a particular hypothesis while ignoring or omitting contradictory findings. Other abuses include sampling bias, misinterpretation of correlations and associations as causation, failure to account for confounding variables that may influence both the independent and dependent variables in a study, formulating hypotheses after analysing the data and observing the results, which can create the illusion of a priori hypotheses (Hypothesising After Results are Known (HARKing)), data fabrication and falsification to support a desired outcome or hypothesis, and misleading visualisations by altering scales, omitting relevant data or using inappropriate visualisation techniques. Even journal editors are not spared from these common abuses, as they engage in publication bias. This occurs when studies with statistically significant results are more likely to be published, while studies with non-significant or negative results are less likely to be published. Researchers, analysts, and practitioners must know these statistical abuses and adhere to the best data analysis and reporting practices. Ensuring transparency, robustness, and ethical conduct in statistical practice is crucial for maintaining the integrity of research and decision-making processes.

Real-World Wonders

Notwithstanding the abuses, Statistics has been instrumental in uncovering extraordinary insights and achievements across various fields. In the medical field, Statistics has played a pivotal role in genomics and personalised medicine. The analysis of massive genomic datasets has led to breakthroughs in understanding genetic diseases, identifying targeted therapies, and predicting disease risk. Statistical designs in clinical trials are critical for evaluating the safety and efficacy of new treatments and drugs. Landmark trials that have shaped medical practice and improved patient outcomes include the Framingham Heart Study (Dawber, 1951), the Diabetes Control and Complications Trial (DCCT) (Group, 1993), the National Lung Screening Trial (NLST) (Aberle, 2011), and the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) (Lieberman, 2005).

Statistical models play a crucial role in analysing climate data, projecting future climate scenarios, and assessing the impact of climate change on ecosystems and human society. These models integrate historical climate data, physical principles, and statistical techniques to predict future climate conditions. The Intergovernmental Panel on Climate Change (IPCC) is a prominent authority on climate science. Its assessment reports are comprehensive summaries of the state of climate science and the impacts of climate change. These reports heavily rely on statistical models and analyses to support their findings and recommendations. The IPCC has released several Assessment Reports, the most recent being the IPCC Sixth Assessment Report (AR6). These reports provide extensive information on climate modelling, projections, and impact assessments (IPCC, 2021) and (IPCC, 2018).

Statistics forms the foundation of machine learning (ML) and artificial intelligence (AI) by providing mathematical and probabilistic frameworks for understanding, modelling, and predicting data. Probability theory, a branch of statistics, is essential for modelling uncertainty and making predictions in AI and ML. Concepts like Bayes’ theorem enable probabilistic reasoning and inference. Statistical models, such as regression, classification, and clustering, are fundamental tools in ML. These models are built using statistical algorithms and techniques. Statistical principles guide data cleaning, normalisation, and transformation techniques. These steps are essential for preparing data for ML algorithms. Statistical principles are leveraged to solve real-world problems in image recognition (Bishop, 2006), natural language understanding (Jurafsky & Martin, 2020), and autonomous vehicle technology. Big Data analytics have driven business optimisation, innovation, and better industry decision-making. By leveraging massive datasets, organisations can gain a competitive edge, enhance customer experiences, and improve operational efficiency.

In a nutshell, the real-world wonders we have explored above vividly, though the tip of the iceberg; illustrate the extraordinary power of statistics. From unravelling the mysteries of our DNA to predicting the future of our climate, statistics serves as the compass guiding us through the labyrinth of data. It transforms mere information into actionable insights, revolutionising industries and shaping the course of human progress. Through statistics, we have harnessed the potential to make life-saving medical breakthroughs, understand and mitigate climate change’s impacts, and enable machines to understand language and images. We have also bridged the gap between data and decision, uncertainty and clarity, and the present and future.

The true wonder of statistics lies not only in its transformative capabilities but also in its universal accessibility. It empowers individuals, organisations, and nations to make informed choices, allocate resources wisely, and confidently address complex challenges. It is the bridge between curiosity and discovery, between questions and answers. As we stand at the threshold of an increasingly data-driven world, let us appreciate the profound role of Statistics in our lives. Let us embrace its power to illuminate the path forward, to make sense of complexity, and to shape a future that is not only data-rich but is also enlightened by the wisdom of numbers. In the grand tapestry of human achievement, statistics is the golden thread that weaves through the fabric of knowledge, progress, and innovation. It is the compass, the torch, and the guide to unlocking the extraordinary wonders of our world. So, as we embark on our journey beyond this lecture, may we carry with us a newfound appreciation for the boundless potential of statistics – a force that empowers us to turn data into dreams and to transform the ordinary into the extraordinary.

The Celebration of Statistics and Nigeria

Mr Vice-Chancellor, Sir, it is widely recognised that statistical science plays a crucial role in furthering humanity’s progress.

At the global level, World Statistics Day (WSD)^[3] is celebrated every five years on October 20th, with the most recent celebration on October 20th, 2020. This global event, organised by the United Nations, draws attention to the significant role of Statistics in shaping the world and promoting sustainable development. It highlights the importance of obtaining high-quality, reliable data for making informed decisions based on evidence.

African Statistics Day (ASD) is an annual event celebrated on November 18 at the continental level. The day was adopted in May 1990 by the Sixteenth Meeting of the United Nations Economic Commission for Africa Conference of African Ministers responsible for Planning and Economic Development. The African Union and the United Nations Economic Commission for Africa (UNECA) organised the event to promote the value of Statistics in informing policies and development agendas across the African continent. Accurate data is important to address Africa’s unique challenges and opportunities. In addition, African Statistics Week (ASW) is held in November, coinciding with African Statistics Day. During this week, various activities and events are organised to raise awareness about the importance of statistics in Africa. ASW allows statistical organisations, policymakers, researchers, and civil society to engage in discussions and collaborations to improve regional data collection, analysis, and utilisation.

The 2023 edition^[4]was held under the theme “Modernising data ecosystems to accelerate the implementation of the African Continental Free Trade Area (AfCFTA): the role of official statistics and Big Data in the economic transformation and sustainable development of Africa”. The theme aligns with the African Union theme of the year 2023, “Acceleration of the AfCFTA Implementation”, and encapsulates the call to modernise data systems on the continent to produce and use high-quality official statistics and seize the opportunities that big data presents.

Many regions, such as the European Union (Eurostat) and the Asia-Pacific region (United Nations Economic and Social Commission for Asia and the Pacific – ESCAP), have regional statistical offices or organisations. These bodies focus on regional data collection, harmonisation, and the promotion of statistical development within their respective areas. Regional economic communities, such as the African Union and the Association of Southeast Asian Nations (ASEAN), prioritise statistical data for policy formulation and decision-making within their regions. They often collaborate on data collection and reporting initiatives and support capacity-building initiatives to enhance statistical capabilities at the regional and sub-regional levels. These programs aim to improve data collection, analysis, and dissemination practices.

Nevertheless, despite what we see at the global and continental levels concerning the use and celebration of statistics, it remains uncertain whether Nigeria, our university, or my listeners place the same significance on statistics. While the rest of the world and Africa recognise the value of statistics, we have yet to fully appreciate its power in Nigeria. When will we begin to acknowledge, embrace and celebrate statistics in our nation?(Olubusoye, 2014) Isn’t it time for Nigeria to have Nigerian Statistics Day (NSD)? Isn’t it time to unlock the extraordinary power of statistics in Nigeria, improve the quality of life, and bring development to every nook and cranny of the country?   As far back as 1985, (Afonja, 1985) lamented as follows:

 “What has always been a puzzle to me and remains so is whether successive governments really sufficiently appreciate the seriousness of the situation beyond mere public utterances of the importance of statistics. Good enough data hardly exists. Where they exist they are so disjointed, disorganised or even buried unanalysed so that they are virtually useless to anybody”.

When it comes to setting agenda, there is a notable distinction between the global level and Nigeria, and it lies in the use of data and statistics. For example, the defunct Millennium Development Goals (MDGs) had eight key points, 21 measurable targets, and 60 indicators to track progress towards the agenda. Similarly, its successor, the Sustainable Development Goals (SDGs) consists of 17 goals, 169 targets, and 247 indicators. Data and statistics play a crucial role in developing implementation and monitoring strategies serving as management tools and a report card for measuring progress and ensuring accountability. A robust indicator system is vital for identifying and measuring problems and achieving meaningful transformation. To do so, it’s essential to establish quantifiable targets, develop performance indicators, and create a framework for data collection, management, processing, and reporting.

How has Statistics been promoted and celebrated at the University of Ibadan? There has been concern about the limited space for the Statistics Department at the University of Ibadan, despite the Department moving to the Math/Statistics Building Complex about 40 years ago. The Department has been in an occupied territory for years and has experienced expansion-related challenges. This situation is similar to the relationship between Israel and Palestine, with the Department living with the “landlord”. In 2010, the then Dean of Science, now the Vice-Chancellor, intervened to negotiate more space for the Department in the building complex but this did not work out. However, a solution resembling a two-state solution was found. Eventually, the construction of a new building commenced during Prof. Adewole’s administration and lasted for more than five years. However, just as the department began moving to the new building, rockets were fired, and the building was partially destroyed by fire. Coincidentally, this happened under the Vice-Chancellor, our mediator and sympathiser, thus facilitating the speedy rehabilitation. We are hugely indebted to the Vice-Chancellor for intervening in our situation. He is a good friend of Statistics!

However, the university must prioritise Statistics in addition to addressing expansion-related issues. To achieve its goal of becoming a world-class institution for academic excellence that caters to societal needs, university management must actively promote Statistics. This can be done by funding and creating an enabling environment for the Department of Statistics to provide free and mandatory statistical assistance to the university community in a fashion similar to the Multidisciplinary Central Research Laboratory (MCLR). This service will improve the quality of research conducted and offer graduate students valuable learning experiences as they work as consultants on real-world problems. Furthermore, community members participating in sponsored research will benefit from in-depth assistance from statisticians, which could lead to more successful grant proposals. Virginia Tech (VT), USA, has set a great example by providing free statistical assistance through the VT Laboratory for Interdisciplinary Statistical Analysis (LISA). LISA was established in the Department of Statistics through funds provided by various units within the university, including the Office of the Vice President of Research, the College of Science, the Graduate School, the Office of the Provost, and all seven other colleges within the institution (Agriculture and Life Sciences, Architecture and Urban Studies, Engineering, Liberal Arts and Human Sciences, Natural Resources and Environment, the Pamplin College of Business, and the Virginia-Maryland Regional College of Veterinary Medicine) (Olubusoye, Alaba, Folorunso, & Akintande, 2022).

The University of Ibadan Laboratory for Interdisciplinary Statistical Analysis (UI-LISA)

Establishment, Mission and Growth of LISA

In 2015, I led my colleagues to establish the University of Ibadan Laboratory for Interdisciplinary Statistical Analysis (UI-LISA), now an esteemed unit within the Department of Statistics (Olubusoye, Alaba, Folorunso, & Akintande, 2022). The establishment of this lab was part of a larger initiative called LISA 2020, which aimed to establish twenty statistical laboratories (Stat Labs) in developing countries by 2020. This visionary idea was from Prof. Eric Vance of Virginia Tech in Blacksburg, USA, in 2012. The first LISA Fellow, Olawale Awe, a Ph.D. supervisee of Dr Abosede Adepoju, was selected in May 2013 and underwent intensive practical training under the supervision of Prof Eric Vance. Upon completing the training, Olawale Awe returned to Nigeria to establish the first Stat Lab, the Laboratory for Interdisciplinary Statistical Analysis and Collaboration (LISAC), at the Obafemi Awolowo University, his home university. The second Stat Lab, SUALISA, was established in Tanzania at the Sekoine University of Agriculture in December 2014(Vance & Magayane, 2014). UI-LISA became a proud member of the LISA 2020 Network in June 2015. Today, with 35 full-member LISA 2020 Global Network Stat Labs^[5], 6 Transitional Stat Labs and 7 Proposed Stat Labs spread across ten countries, including Nigeria, Brazil, Pakistan, Ghana, South Africa, etc., the original vision has been achieved and exceeded. A well-documented LISA 2020 Global Network account is available in (Awe et al., 2022).

Three challenges underline the creation of UI-LISA in the department. The first is maintaining a judicious balance between theory and practice. At the department’s inception, much emphasis was placed on traditional statistics, with overwhelming importance placed on theory as the driver of statistical concepts without the essential practical skills needed for using statistics to solve real-life problems. The curriculum design failed to show where mathematics ends and where statistics begins. The second problem is connected with a statistical collaboration laboratory’s concept, setup, and operation. Even though statistical computing labs exist, they are not the same as statistical collaboration laboratories. Instead, statistical collaboration laboratories provide a learning platform for soft and technical statistical skills that can be readily transferred and deployed to work with domain experts in science, social science, medicine, education, agriculture, etc., to create solutions to research, business, and policy goals. According to (Vance & Pruitt, 2022), “Those with a strong background in theory and methods of statistics can learn applied and collaborative skills … Through mentorship from an experienced individual or via a formal program, a statistician can become a collaborative statistician or data scientist; collaborate with domain experts on problems in research, business, or policy …  An individual can also learn collaboration skills through self-study of Vance and Smith’s ASCCR (Attitude‑Structure‑Content‑Communication‑Relationship) Framework for Collaboration.”The third but notable problem was the poor motivation or lack of incentive for interdisciplinary or collaborative research in the university system. The issue of single, co-authorship and multiple-authorship has been debated in the literature. The university promotion guidelines reward single authors compared with multiple authors. Modern-day realities put collaborative research at the epicentre of societal problem-solving (Olubusoye, et al., 2022).

Constant monitoring and follow-up by Prof. Eric Vance helped to sustain UI-LISA and position it to continuously serve as an engine for development through training and collaboration. There were two commissioned visits to UI-LISA to help strengthen its operations and assess its activities. I also made an exchange visit to two prestigious universities in the United States, Virginia Tech and Purdue University, to see first-hand how statistical laboratories train students to communicate and collaborate with non-statisticians and how to manage labs to help clients apply statistics in their research projects. The unforgettable visit by Mr Ian Crandell, a doctoral student from VT to UI-LISA, helped UI-LISA develop a programme similar to LISA at VT and aligned with the LISA 2020 programme. During his visit, he significantly contributed to setting up UI-LISA as a fully functional statistical collaboration centre. The second visit commissioned by the founder of LISA 2020 was by Ms Monica Johnston, an expert in statistics education and assessment with expertise and experience working with female statisticians and small business owners in the United States. She conducted an assessment of UI-LISA and identified some strengths of UI-LISA, including potential opportunities to make UI-LISA even stronger and more sustainable. Table 1 below summarises the report of her assessment. She also shared some ideas for writing grant-winning proposals to sustain UI-LISA.

The overall mission of UI-LISA is “Building Statistics and Data Science Capacity in Nigeria.” The goals are to (1) complement statistical training in the department by developing the non-technical skills of our students to be able to relate with domain experts in other fields; (2) make statistics graduates to be employable in industry or be self-employed through practical experience; (3) promote statistical literacy within and outside the university community; and, (4) use statistics to transform our society by identifying and solving local problems. Consistent with the LISA Network mission, UI-LISA puts UI students at the centre of its mission statement to build their capacity in data science and extend the same to statistics students in higher education institutions in Nigeria. Consequently, the lab programmes and activities are designed to support this mission and goals. Thus, UI-LISA has seven (7) mission-driven programmes, namely LISA drop-in assistance, Collaborative Training, Short Courses, Student Training (formal and 1-1 tutoring), Industrial Training (IT) for students completing their academic degree, One Hour with a Statistician and the Mobile Statistical Clinic. Two programmes serve the students, while the other five serve the entire university community, including statistics and non-statistics students, academic and non-academic staff from other departments and interested persons from neighbouring universities.

Table 1:Summary of Monica Johnston’s Assessment of UI-LISA Programmesin January 2017

Source:(Olubusoye, et al., 2022)

Statistical Training Activities at UI-LISA

UI-LISA has provided statistical advice to 65 research projects, including 18 PhD theses, 46 masters’ dissertations, and 1 NGO project. In collaboration with CPEEL, UI-LISA participated in a university-wide energy audit project commissioned by the Vice-Chancellor of UI in 2018. Indeed, the most productive collaboration UI-LISA has had is with CPEEL (https://cpeel.ui.edu.ng/). The collaboration has yielded two journal articles on renewable energy in Africa (Olanrewaju, et al., 2019), (Olubusoye, et al., 2021), Akintande, et al., 2020).

Over the past years, UI-LISA has been collaborating with the Industrial Training Coordinating Centre (ITCC) in UI to build the capacity of the IT students through hands-on and practical skills beyond the formal training provided for the undergraduate statistics programme in the NUC BMAS document. The initiative commenced in 2015 with the posting of two male students to UI-LISA by ITCC, and since then, the number has been increasing steadily. The reputation of the internship training has spread beyond the shores of UI. Applications for internship training have been received from prospective interns from neighbouring universities, polytechnics and other tertiary institutions. These institutions include the Federal University of Technology, Akure (FUTA), in Ondo State; Federal University of Agriculture, Abeokuta (FUNAAB) in Ogun State; Obafemi Awolowo University (OAU), Ile-Ife, in Osun State; and ThePolytechnic, (Ibadan Poly) and Federal School of Statistics (FSS), both in Ibadan, Oyo State. Also, graduate students who felt deficient in practical statistical skills from both quantitative and non-quantitative disciplines have participated in UI-LISA internship training. Specifically, UI-LISA has trained graduates who completed the National Youth Service Corp (NYSC) and those running a Postgraduate programme with the Centre for Petroleum Energy Economics and Law (CPEEL).

Table 2 shows the distribution of intern trainees at UI-LISA. The record shows that 55 interns, comprising 37 (67%) male and 18 (33%) female trainees, were enlisted between 2015 and 2020. The interns participated in all lab activities, including walk-in statistical clinic sessions where statistical advice was provided to university community members. They are directly involved with the day-to-day activities of the lab, such as attending to clients and participating in the design of survey tools and survey fieldwork. They undergo specialised hands-on training in statistical tools and software, particularly R, statistical consulting practices, statistical collaboration, statistical communication and professional conduct and ethics for practising statisticians. The interns undertaking their IT programme with UI-LISA are closely monitored and thoroughly supervised by their academic supervisor andITCC supervisors respectively; this is in contrast with the plight of their counterparts posted to government organisations. After training, the interns on our programme complete an online survey to collect feedback about their experience and skills learnt.

Table 2:Number of Industrial Trainees at UI-LISA from 2015 to 2020

Source:(Olubusoye, et al., 2022)

UI-LISA Collaboration with Transforming Evidence to Action (TEA)

Transforming Evidence to Action (TEA) is a novel programme of the Laboratory for Interdisciplinary Statistical Analysis (LISA) at the University of Colorado Boulder and the statistics and data science collaboration laboratories (“stat labs”) of the LISA 2020 Network (www.lisa2020.org). The programme entails stat labs creating institutional statistical analysis and data science capacity to enable and accelerate local solutions for local development challenges and is funded by the U.S. Agency for International Development (USAID). The logic of TEA is that if we can create and sustain stat labs comprising well-trained statisticians and data scientists who can move between theory and practice to apply statistics to solve problems and make decisions, and if they can collaborate with development actors empowered to produce data and take action on development issues, then the members of the stat labs could transform data into evidence to answer their collaborators’ research questions or help them make data-driven policy decisions that will lead to developmentimpacts and benefits to society(Olubusoye, et al., 2021). In this study, we discuss a case study in TEA to aid in formulating a national electoral policy to strengthen democracy in Nigeria. The study was executed jointly by the University of Ibadan Laboratory for Interdisciplinary Statistical Analysis and the Independent National Electoral Commission (INEC)—the electoral umpire in Nigeria—with full support from the headquarters of the LISA 2020 Network (Olubusoye, et al., 2020). 

UI-LISA Collaboration with the Independent National Electoral Commission(INEC)

The lab has just completed a project on Enhancing Election Participation in Nigeria in collaboration with The Electoral Institute (TEI), a research and documentation unit of the Independent National Electoral Commission (INEC). The LISA provides funding at the University of Colorado Boulder in cooperation with the US Agency for International Development (USAID) Accelerating Local Potential (ALP) Programme (Cooperative Agreement Number: 7200AA18CA00022). The project has produced three policy briefs and three research working papers. Additionally, the lab has completed a sanitation project funded by the USAID TEACH Fund in collaboration with the Ministry of Environment and Natural Resources, Oyo State, Nigeria.

The chequered history of elections in Nigeria spans the years of military rule (1966–1979 and 1983–1999) and civilian regimes (from 1999 to date). The pattern in the turnout rate since the return of civilian democratic rule in 1999 has revealed voters’ despondency and apathy toward the democratic process. The presidential elections that usually should attract the biggest voter turnout dropped from 69 percent in 2003 to its lowest rate of 29 percent in 2023. Despite the voting age population increasing from 53 million in 1999 to 97 million in 2019, the number of voters in the presidential elections dropped from 30 million in 1999 to 28 million in 2019. This tragic pattern of voters’ turnout has continued to pose serious concerns not only to the electoral body INEC but to all stakeholders committed to democracy in Nigeria. If this downward trajectory in electoral participation persists, it will constitute a major threat to good governance, growth, and development.

The low participation rate is a hydra-headed problem because of myriad events connected to the voting process, including the continuous voter registration (CVR) exercise, management of the voter register, and the accreditation/voting exercise. Addressing the non-participation in elections holistically requires finding answers to several key policy questions. A seven-member team from UI-LISA collaborated with The Electoral Institute (TEI), a sub-organization within INEC (INEC-TEI) from May 2019 to March 2020 to investigate the factors responsible for voters’ apathy and begin to answer the policy questions below.

1. Policy questions relating to the quality of the voter register:

1. Is the database of registered voters clean?  Has it been updated to exclude dead and non-Nigerian registrants or voters who have left Nigeria and to include all citizens who are 18 years or older?
2. How accurate is the information captured in the database of registered voters for items such as age, contact address, etc.?

2. Policy questions relating to the conduct of the CVR exercise:

1. What are the barriers to CVR participation by eligible Nigerians?
2. For what reasons do eligible citizens fail to register for elections?
3. What are the socio-economic characteristics of the successful registrants compared to the failed registrants?

3. Policy questions relating to the conduct of the voting exercise:

1. To what extent were the INEC guidelines for accreditation and voting complied with during the 2019 presidential election?
2. What are the socio-economic characteristics of the voters in the 2019 presidential elections compared with non-voters?

4. Policy questions related to the measurement of the turnout rate and quality of census data:

1. How reliable is the Voting Age Population (VAP) as the denominator of voter turnout for measuring participation rate, considering that VAP is an intercensal projection based on the national population census last conducted in 2006?
2. Does the relationship between VAP figures and total registered voters (RV) exhibit features that question the quality of the VAP projections or the voters register over the years?

In collaboration with INEC-TEI, the UI-LISA team conducted a nationally representative sample survey covering six states selected based on the criteria of one state per geopolitical zone, population size, representativeness, relative accessibility, and security. Twelve registration areas (RA) were randomly selected from each selected state, and from each RA, 30 registrants were selected at random from the INEC register of voters. Thus, 2,160 registrants drawn randomly from the voter register were visited and interviewed in the survey. In addition, fifteen non-registrants (those not listed in the INEC register of voters) who were eligible to vote were sampled randomly from each RA. Thus, a total of 3,240 respondents, consisting of 2,160 INEC registrants and 1,080 eligible but not registered citizens, were interviewed in the survey. The participants were asked to answer a three-part questionnaire covering CVR, accreditation, and voting participation topics. The sample selection and the fieldwork exercise were implemented and administered wholly by INEC personnel and coordinated by The Electoral Institute, with technical design and support from UI-LISA. To complement the national survey data, electoral data from the records of INEC and other sources were interrogated to identify factors, issues, and events that might have impacted electoral participation.

1. The Quality of the INEC Register of Voters
2. Only 56 percent of the sampled voters still retained the same information regarding vital electoral statistics as in the register.
3. 12 percent had changed their residence, rendering the residential address in the register invalid.
4. 1 per cent were confirmed to have died when visited at the address provided in the INEC register.
5. 12 percent gave a different age from the one on the INEC register and that the problem was underage.
   1. CVR Exercise
6. Gender is associated with voter registration, with females comprising a relatively higher percentage of non-registrants.
7. The middle-aged respondents (25-44) participated more in voter registration than other age groups.
8. Higher education seems associated with an inclination to register.
9. Religion does not affect attitude toward voter registration.
10. Party-affiliated citizens are more inclined to register. 2019 Voting Exercise
11. No non-accredited registrants participated in the voting process. However, there was evidence that 1% of accredited voters failed to complete the election cycle by not participating in the voting exercise.
12. The four states with the highest percentage of non-participants percentages (Lagos, Rivers, Anambra, and Kano) coincide with those with the worst turnout rates in the 2019 general election.
13. The difference between male and female participation in the election was not statistically significant.
14. The evidence depicts an inverse relationship between non-participation and level of education.
15. The occupation and employment status of registrants is a significant factor associated with non-participation in the voting exercise.
16. An inverse relationship between non-participation and income status.
    1. The Measurement of the Turnout Rate and Quality of Census Data

Voter turnout can be measured in several ways. The Basic Voter Turnout Rate is defined as ; this is the percentage of the total population that acted on its behalf to elect the government. A first-order standardisation of the basic voter turnout is the VAP-based definition ; it is the basic voter turnout rate standardised by the voting age population ratio (or Voting Age Rate) . The VAP-based turnout rate is the ratio of total votes to the theoretically eligible voting population. The registered voter (RV)-based voter turnout rate is a further standardisation of the basic voter turnout rate by the VAP-based voter registration. The registered voters (RV) – based voter turnout rate is defined as ; the is an international norm while the VAP-based  is not very popular, though still used. Ideally, the six measures of electoral participation depicted in Figure 1, particularly the three relating to voter turnout, should be interrelated, moving in the same direction over time.

Table 3:Alternative Measurements of Voter Registration Turnout

Fig. 1: Rates of Voting Participation: 1959 – 2019.

However, this is not the case here. The cumulative verdict of the descriptive statistics of the voter turnout rates and their correlations in Table 4 and the quadrennial increases in Figure 2 is that the voter register has problems. However, it seems that the voter register has improved since 2007 to feature some level of consistency. The undeniable fact is that by whatever criteria, the voter turnout has been declining since 2003.

Table 4: Descriptive Statistics of Voter Registration and Voter Turnout Rates and their Correlations

Key: VAR – Voting Age Rate; BVRR – Basic Voter Registration Rate; SVRR – Standardised Voter Registration Rate; BVTR – Basic Voter Turnout Rate; CVTR – Crude Voter Turnout Rate; SVTR – Standardised Voter Turnout Rate.

Data on the total population, voting age population, total registered voters, and total votes cast at major national elections in Nigeria from 1959-2019 is presented in Figure 3.

Fig. 2: Percentage increase in the measures of voting participation from the previous election.

Paradoxically, in Nigeria, from 1979 to 1999, the registration turnout rate expressing total registered voters as a percentage of the voting-age population returned impossible values (>100 per cent) (see Table 5). These elections represent four of the nine total presidential elections in Nigeria. These figures mean that the total registered voters exceeded the voting age population–a part being greater than the whole. This mathematical impossibility questions the accuracy and quality of at least one of the two candidates of the ratio quotient; it could be inflation of registered voters, an underestimation of the voting age population, or both.

Table 5:Computed Electoral Participation (Registration and Voting) Rates

Fig.3: Total population, voting age population, total registered voters, and total votes cast at major national elections in Nigeria 1959-2019.

The problem of electoral participation in Nigeria could be attributed to actions or inactions on the part of critical stakeholders, and the solution would also come from them; they have to be alive to their respective responsibilities and act collaboratively to make the electoral system effective and efficient. The Independent Nigeria Electoral Commission (INEC) leads the pack; the others include the executive at all levels, the legislative houses at all levels, the politicians, conjugate ministries and departments and agencies, the media, civil societies and the citizenry. For now, it seems that collaboration between critical stakeholders is low; indeed, some of them act at cross purposes with dire consequences for voter registration and voting exercises (Olubusoye O. E., et al., 2020).

My Exposure, Training and Research at the Centre for Econometric and Allied Research (CEAR)

Mr Vice-Chancellor, I will like to discuss my exposure and training at the Centre for Econometric and Allied Research (CEAR) under its distinguished and retired leaders, Prof. Samuel Olofin, Prof. Akin Iwayemi and my Ph.D. Supervisor, Prof. J. O. Iyaniwura. Unfortunately, in this environment, those who did not create anything will permanently and swiftly destroy what others built with great sacrifice and effort. CEAR made me, and I am proud to identify with it. I benefitted immensely from its support and investment in human development, and it provided me with a conducive environment to unlock the extraordinary power of Statistics. CEAR supported my participation in several academic conferences outside Nigeria. I was not the only one; many of my colleagues in other Departments, including Economics, also did. This is made possible through the Centre’s practice of raising funds from research projects and channelling part of honoraria into research and training funds to help young researchers and build capacity. This culture is rare and often attracts resistance from those who prefer sharing everything. CEAR has left a legacy of commitment to investing in young researchers and fostering academic excellence.

The activities at the Centre are in three broad categories: Research, Training and Public Service. In research, the Centre emphasises basic or pure academic research that is interdisciplinary and aimed at contributing to knowledge in areas including but not limited to:

• Econometric modelling for policy analysis and forecasting;
• Applied general equilibrium modelling for policy;
• Input-output modelling and construction of I-O tables;
• Social Accounting Matrices (SAM) construction;
• System simulation studies;
• General mathematical modelling and applications.

In the area of training, the Centre creates an enabling environment for graduate students from any discipline in this University, particularly Economics, Statistics, Mathematics, Agricultural Economics, etc., and others to benefit from research and training capabilities. Prominent people who have benefitted from this scheme include Prof. Chukwuma Soludo, the former Governor of the Central Bank of Nigeria, who was sent from UNN Nsukka; Dr Sarah Alade, a former Deputy Governor of Economic Policy, CBN, who was sent from Obafemi Awolowo University, Ile-Ife, and a good number of others that are too numerous to mention here. Over 200 researchers, primarily postgraduate students and even top-cadre academic staff from various institutions, have benefitted from the Centre’s quarterly Econometric Workshops, which I coordinate with Prof. Afees Salisu.

A compendium of notable research outputs at the centre covering the above areas is contained in (Adenikinju , Busari, & Olofin, Applied Econometrics & Macroeconometric Modelling in Nigeria, 2009). Among others, I would like to share three studies on the Nigerian economy I contributed to.

Modelling of the Inflation Process in Nigeria

In 2004, under the able guidance of Professor Olofin (then Director of CEAR), I developed a proposal with an economist who was also a Fellow at the Centre—Dr Rasheed Oyaromade. The African Economic Research Consortium (AERC) based in Nairobi, Kenya later accepted the funding. Despite contributing significantly to the proposal, the Director insisted I must be the principal investigator simply because I initiated the idea. It was the first successful research proposal after my PhD. A grant of US$14,000 was awarded for the study titled “Modelling the Inflation Process in Nigeria”.

Inflation, a general increase in the prices of goods and services, has been the subject of extensive research by economists in developed and developing countries. In 1972, Oyejide conducted a pioneering study on the root causes of inflation in Nigeria from a structuralist perspective (Oyejide, 1972). Two years later, the Nigerian Institute of Social and Economic Research (NISER), Ibadan organised a conference in Ibadan to discuss the inflation process in Nigeria. The conference concluded that Nigeria’s inflation could not be solely attributed to monetary or structural factors, but rather a combination of both (Onitiri & Awosika, 1982). Despite Nigeria’s monetary policy focus on price and exchange rate stability over the last forty years, the country has experienced various forms of inflation, ranging from creeping to moderate and from high to galloping. For a visual representation of Nigeria’s inflationary episodes, please refer to Table 6 and Figure 4.

From 1960 to 1972, the average inflation rate was relatively low. The historical average rate during that time was 5.01%. However, from 1973 to 1985, the inflationary pressures were greater than in the previous period, with an average inflation rate of 17.96%. On a comparative scale, the period of 1986-1995 represented a time of even greater inflationary pressures than the preceding periods, with a historical average rate of 31.50%. The inflationary pressures eased relatively from 1996 to 2002, with an average rate of 13.34%.

Table 6: Inflation Episodes in Nigeria (1960-2002)

Source: (Olubusoye & Oyaromade, 2008)

Fig. 4: Rate of Inflation in Nigeria, 1970-2006

Source:(Olubusoye &Oyaromade, 2008)

In our study (Olubusoye & Oyaromade, 2008), an empirical model where inflation is assumed to originate from both the demand and supply sides is specified. Specifically, the supply side is captured by the tradeable sectors, whereas the non-tradeable sectors represent the demand side. The prices of non-traded goods respond to disequilibria in the money market, and the movements in the exchange rates and foreign prices govern the prices of traded goods. The overall price level is a weighted average of the prices of tradeable and non-tradeable goods, with a constant representing the share of tradeable goods in total expenditure.  We specified a general dynamic equation called auto-regressive distributed lag (ADL) of order k as follows:

(Equation 1)

where p=price index; m^s = nominal stock of money; y = real income; p^e= expected inflation; i = interest rate; e = real exchange rate; d = international oil price in domestic currency. The equation was then rewritten to obtain the error correction representation of the form:

(Equation 2)

The parentheses in Equation (2) represent the error correction term. The coefficients of this equation are functions of the coefficients of (1); the two equations are entirely the same. It can be seen from (2) that as long as the variables are cointegrated, all the variables on both sides of (2), including the error correction term, are stationary. If inflation rises above its long-run equilibrium at time t-1, the term in the bracket assumes a positive sign. Because g is negative, its effect at time t is to dampen the inflation towards its steady state path. The following variables are used in the model estimation: cpi = consumer price index; exinf = expected inflation; fisc = fiscal deficits/GDP ratio; gdp = gross domestic product; int = interest rate; mss = money supply; pet = oil prices; rain = average rainfall; rer = real exchange rate. The results of the parsimonious ECM are shown in Table 7. The adjusted R-squared of the estimated model shows that about 90% of the variation in CPI is explained by the combined effects of all the determinants, while the F-statistic shows that the overall regression is significant at a 1% level. Also, the Durbin–Watson statistic value of 2.164 indicates the absence of autocorrelation in the analysis, and the equation standard error of 0.044 signifies that for about two-thirds of the time, the predicted value of CPI would be within 4.4% of the actual value.

Table 7: Results from the Error Correction Model

As shown in Table 7, the lagged values of the CPI marginally influence the volatility of inflation over time with a weak inertia of a mere 10%. Current expectations about future price levels significantly influence the volatility of inflation. Specifically, the coefficient of expected inflation is positive and significant at the 1% critical level. Another interesting result from the analysis is the impact of weather conditions on general prices. As shown in the table, lagged values of rainfall exert a significant negative impact on the level of current inflation. This shows that good weather conditions, as reflected by high rainfall in the previous year, lead to lower food prices (the most essential component of the CPI basket) in the current year.

The real exchange rate exerts a very significant positive influence on the level of CPI, while its lagged values affect inflation negatively. It was also found that the coefficient of output represented by the real GDP is positively signed and is not significant in determining the level of CPI. In addition, petroleum price is found to be significant at only a 10% critical level and is positively signed. The coefficient of the fiscal variable (ratio of fiscal deficits to GDP) is not significantly different from zero, although it is positively signed. Despite the fiscal dominance in Nigeria, this apparent weak impact of the fiscal variable may be attributed to the inclusion of money supply and expected inflation in the analysis, both of which reflect deficit financing. Unexpectedly, however, the coefficient of money supply was found to be negatively signed and significant at the 10% level.

Overall, the empirical analysis revealed that while fiscal deficit and changes in broad money drive inflation in Nigeria in the long run, lagged CPI, inflation expectations, and weather significantly affect inflation in the short run. The exchange rate was a significant determinant of inflation in both the short and the long run. This unique characteristic of the exchange rate in Nigeria may be clearly understood when the activities in the parallel foreign exchange market rate are considered.

Modelling the Nigerian Economy

The concept of evidence-based policymaking is not novel. However, the absence of this critical element in public policy in Nigeria is a significant concern. Decisions at various levels are primarily based on political inclination, sectional interest, back-of-the-envelope approach, rule-of-the-thumb, ideological sentiments, illegal gratifications, religious affiliation, and other parochial interests. Evidence-based policymaking is a sharp contrast to inherent ways of doing things. It involves using statistics to recognise issues of national importance, develop and design policy interventions, make informed policy choices, forecast possible future outcomes, monitor policy implementation, and evaluate policy impact (Olofin, et al., 2011). Without evidence, policymakers often fall back on intuition, ideology, or conventional wisdom—or, at best, theory alone. The resulting policies go astray because of complexities and interdependencies in our society and economy and the unpredictability of people’s reactions to change.

The realisation by the Central Bank of Nigeria (CBN) of the need to imbibe and promote the use of an evidence-based rather than ad-hoc approach in arriving at critical decisions at its monetary policy committee meetings led to a robust collaboration between it and CEAR, contrary to the hitherto practice of engaging foreign consultants. This alliance involves sharing experience and technical expertise with the Bank staff, particularly the Research Department, and providing analytical and empirical evidence to assist in its research, policy-making and regulatory responsibilities. The CEAR-CBN macro-econometric working group includes the Director, Prof. Olofin, Prof. Afees Salisu, Dr Alarudeen Aminu, and my humble self. 

The CEAR-CBN initiative has yielded a positive result by developing an operational model for the Nigerian Economy. It is perhaps the only functional model for Nigeria that is regularly updated to provide a scientific basis for monetary policy decisions. The model is termed “CBN MAC”. There are now two versions—CBN MAC I and II. The CBN MAC I is a medium-scale macro-econometric model of the Nigerian economy that can explain inter-sectoral linkages and relationships among crucial macroeconomic variables. The CBN MAC II model, published by the highly rated Journal ofEconomic Modelling, is a small-scale macro-econometric model (SSM) with considerable theoretical content.

This CBN MAC II model focuses on three specific objectives: examining the interactions among the key macroeconomic variables, evaluating the response of these variables to adjustments in the monetary policy rate, and providing some criteria for judging the various possible policy outcomes. Based on the estimated results and outcomes of the model simulations, some restrictions are imposed on the coefficients to generate more meaningful policy decisions and choices. Also, the model has been subjected to a wide range of sensitivity analyses and found to be relatively robust in tracking developments regarding crucial macroeconomic indicators in the Nigerian economy. This study contains the simulation result conducted on three (3) alternative policy options proposed for the consideration of the monetary policy authority, with the ultimate aim of basing policy choices on a rigorous analytical framework against other rules-of-thumb (Olofin, et al., 2014). The structure of the macro model allows the following:

• Evaluation of the performance of any monetary policy rule, whether backward or forward-looking;
• Examination of the interactions among the main macroeconomic variables, including the monetary policy instrument; and
• Provision of some criteria for judging the various possible policy outcomes.

The model has the following five equations:

• The IS Equation:

(Equation 3)

Where,is the output gap,  is the nominal interest rate,  is the expected inflation,  is the real interest rate and is the real exchange rate. Note that is computed as  which is the level of real GDP and which is the potential output is measured as the trend level of GDP ((Berg, Karam, & Laxton, 2006).

• The LM Equation:       

(Equation 4)

where  is the nominal money,  is the price level,  is the real output gap,  is the nominal interest rate and is the nominal exchange rate.

• Aggregate Supply Model (Phillips curve):

(Equation 5)

where π_t is the rate of inflation; is the expected inflation in period t+1 given information at time t; π_t-1 is the immediate past value of inflation; y^g_t and y^g_t-1 are the current and immediate past values of the output gap, respectively; Δis the first difference operator and  and  are the real exchange rate and its expected value in period t+1 given information in period t.  Also,  is the error term while χ₀ – χ₃ are the parameter estimates.

• Exchange Rate Equation (Uncovered Interest Parity):

(Equation 6)

where  and are the real exchange rate and its expected value in period t+1 given information in period t.  Also, and are the policy real domestic and foreign interest rates respectively, and real interest rate (r) is defined as while  and are the parameter estimates.

• Interest Rate Equation:

(Equation 7)

where  is the nominal interest rate,  is the output gap,  is the expected inflation,  is the nominal exchange rate and  is the monetary policy rate.

In the results of the estimated coefficients for four estimators adopted in the study, the magnitudes of the coefficients are quite plausible, correctly signed and statistically significant (see Table 8).

For the purpose of policy simulations, three (3) scenarios, focusing on four (4) key macroeconomic variables (inflation, exchange rate, output growth and interest rate) between 2013:Q3 and 2014:Q4 are analysed. These scenarios are:

• Maintaining MPR at 12%;
• Reducing MPR from 12% to 10% in July, 2013;
• Reducing MPR from 12% to 8% in July, 2013.

The results of the policy simulations are as follows:

• Effects of scenarios on inflation (INF):Figure 5a shows that the different policy scenarios do not differ in their impacts on inflation.
• Effects of scenarios on exchange rate (NER):Figure 5b shows that exchange rate appreciation occurs in scenario 1 and depreciation in scenario 3 while scenario 2 leads to a relatively stable movement in exchange rate.
• Effects of scenarios on output growth (YG):Figure 5c shows that the different policy scenarios do not differ in their impacts on output growth, indicating that the most binding constraints on growth, at least in the short run, may not necessarily be related to interest rate adjustments.
• Effects of scenarios on the lending rate (INTR):As shown in Figure 5d, lending rate is highest under scenario 1 followed by scenario 2 and lowest under scenario 3.

Table 8: Estimated Coefficients

In addition, the performance of these scenarios is ranked in terms of their relative dominance. In other words, the scenario that performs relatively best in its impact on a particular variable is assigned the value of ‘1’ and the one that performs least is assigned ‘3’. Dominance as used here implies that relative to others, a scenario can achieve: the lowest inflation rate, least exchange rate depreciation, highest level of output growth and lowest lending rate.

The results in Table 8 show that the monetary authority has to choose between the objectives of maintaining a stable exchange rate and lowering the lending rate. This is because when it reduces MPR, the lending rate falls but the exchange rate depreciates. In other words, maintaining MPR at 12% leads to a high lending rate but achieves a gradual appreciation of the naira against the US dollar. At the other extreme, however, a reduction in MPR from 12% to 8% in July 2013 would lead to the lowest lending rate among the options but the highest depreciation of the exchange rate.

(a) Inflation	(b) Exchange rate

(c) Output growth	(d) Interest rate

Fig. 5: Effects of scenarios on macroeconomic variables

Note: _1, _2, and _3 refer to the results from scenarios 1, 2 and 3, respectively.

Table 9: Ranking of Scenarios

Youth Unemployment in Nigeria: Nature, Causes and Solution

Mr Vice-Chancellor, this is the most recent study of national significance conducted at CEAR, where I served as the Principal Investigator. The research was funded by the Tertiary Education Trust Fund (TETFund). After multiple rejections from various journals, it was a great delight when the article was eventually accepted for publication byQuality and Quantity – a Quartile One journal published by Springer Netherlands. The reason for the rejections was unrelated to the research’s quality or substance. However, instead, it was a country-specific focus, which many journals felt would not generate enough global citations. This is a common experience when studies address local or even national challenges.  

There are strong motivations to study Nigeria’s youth unemployment problem, understand its nature and contributory factors, and proffer viable solutions. First, the youth unemployment rate in Nigeria has risen to a level that is recognised as a significant problem by the government at different levels. According to the World Development Indicator (WDI) data^[6], Nigeria’s youth unemployment increased gradually from 8.4% in 1991 to 9.1% in 2001 and 9.7% in 2011. Between 2014 and 2016, Nigeria witnessed the highest youth unemployment rate in the past three decades, increasing from 12.6% to 20.7%. This makes Nigeria’s youth unemployment level higher than the world rate (12.6% to 13.2%) and the sub-Saharan Africa (SSA) rate (12.2% to 13.8%) during the same period.

Second, the high magnitude of joblessness among youths has social, economic and political consequences on society in terms of high poverty and dependency levels, militancy, and activities of terror groups (Ajaegbu, 2012); (Bagchi & Paul, 2018);(Olukayode, 2017). While the militancy crisis in the Niger Delta region has subsided, the rate of robbery, terrorism and kidnapping has been on the increase in the country. Relatively new and growing economic crimes in Nigeria are internet fraud and banditry. As may be expected, the security challenge in the country has become multi-faceted and is growing increasingly. According to the Global Terrorism Index (GTI) 2018, Nigeria is the third in the rank of countries with high incidents of terrorism, following Iraq and Afghanistan^[7]. This index covers the crisis between Fulani herdsmen and farmers/communities and killings due to Boko Haram, which claimed to have reduced^[8]. In addition, the problem of youth unemployment may have prompted Nigerian youth into cybercrime as Nigeria was ranked third in global internet crimes in 2015 behind the UK and the U.S^[9].

Third and most importantly, some remedial actions have been taken to ameliorate the problem of youth unemployment in Nigeria, with little or no improvement in the level of youth unemployment. The earliest government’s effort to tackle the unemployment problem dates back to the formation of the National Directorate of Employment (NDE) in 1986 (see (Ekong & Ekong, 2016)). Within the last decade, many supply-side programmes have been rolled out at the federal level to combat youth unemployment. These include the Youth Enterprise with Innovation in Nigeria (YouWIN) introduced in 2011, the Subsidy Re-investment and Empowerment Programme (SURE-P) initiated in 2012, the N-Power Scheme of the National Social Insurance Programme (NSIP) launched in 2016, the Bank of Industry’s Youth Entrepreneurship Support (YES) Programme, among others. Youth employment innovations at the state level include the Osun State Youth Empowerment Scheme (O-YES), the Agriculture Youth Empowerment Scheme (A-YES) in Lagos State and the Youth Empowerment Scheme of Oyo State (YES-O). Nonetheless, the rise in the youth unemployment rate suggests that these policies may be inefficient(Olubusoye, Salisu, & Olofin, Youth unemployment in Nigeria: nature, causes and solutions, 2023).

The trend relationship between youth unemployment and some selected factors is examined (see Figures 6-8) to explore the structural and cyclical nature of the youth unemployment problem in Nigeria. The relationships between youth unemployment and output, inflation and trade openness are presented in Figures 6, 7 and 8, respectively. We expect a positive (negative) relationship between youth employment (unemployment) and economic growth if youth unemployment is cyclical. This appears to be the case after 2013 in Nigeria. Specifically, as economic growth declined from 6.31 per cent in 2014 to -1.62 per cent in 2016, youth unemployment increased rapidly from 12.62 per cent to 20.67 percent in the same period (see Figure 6). This suggests that youth unemployment in Nigeria could be cyclical.

Figure 7 suggests that an increase in youth unemployment in Nigeria may be associated with inflation, consequently indicating that the problem may be structural. As the inflation rate increased rapidly from 13 per cent in 1991 to 57 per cent in 1994, youth unemployment increased from 8.35 per cent to 9.13 per cent in the same period. Between 2015 and 2018, the inflation rate increased from 9 per cent to 17.3 per cent, while youth unemployment, on the other hand, increased from 16.3 per cent to 19.7 per cent (see Figure 7). This suggests that an increase in youth unemployment in Nigeria may not be unassociated with an increase in the inflation rate, implying that monetary easing could enhance youth employment in Nigeria.   

Figure 8 presents trends in the relationship between trade openness and youth unemployment in Nigeria. Between 1991 and 1993, trade openness fell from 37.02 per cent to 33.7 percent, while youth unemployment increased from 8.35 percent to 8.77 percent. Furthermore, when trade openness fell from 53.3 per cent in 2011 to 20.72 per cent in 2016, youth unemployment increased from 9.71 per cent to 20 per cent. When the Nigerian economy rebounded recently, trade openness increased from 20.72 per cent in 2016 to 31.9 per cent in 2018, and youth unemployment fell from 20 per cent to 19.68 per cent. This suggests that a negative relationship between trade openness and youth unemployment may exist, implying that trade liberalisation policies may reduce youth unemployment in Nigeria.

One distinctive contribution of this study is its sectoral analysis. Thus, we further examine how structural economic factors affect youth employment in three major sectors of the economy. The macroeconomic variables are output growth, inflation, interest, and exchange rates. Others include trade openness, value-added tax, capital expenditure and government debt. The trend is presented for the most prominent sectors: agriculture, industry and service.

The graphical views appear to suggest that employment in services and industry (as against employment in the agriculture sector) is more connected with fiscal policy (government expenditure and debt and value-added tax) and monetary policy variables (inflation, exchange rate and interest rate). Inferring for sectoral unemployment (from sectoral employment figures), this is a strong indication that youth unemployment in industry and services sectors may be structural. The implication of this is that proper policies that address the demand side of the labour market; that is, policies that improve the ability of the underlying sectors to generate employment; could prove good to deal with unemployment problems in the sectors. However, such submission cannot be advanced for the agriculture sector. Also strikingly, only in the industry could youth unemployment be seen as cyclical, given the co-movement between output growth and employment (and unemployment by inference) in the sector. The same cannot be adduced for the agriculture and services sectors. Going forward, we explore these indications in greater detail with a formal estimable model to confirm/disprove these preliminary deductions for each sector.

To examine the causes of youth unemployment in Nigeria to be able to define its nature and identify possible solutions, two methodological frameworks, the Vector AutoRegressive (VAR) model and the Panel Autoregressive Distributed Lag (PARDL) model, are employed. The PARDL model is the panel heterogeneous dynamic model. The VAR model was used to examine the youth employment spillovers among the sectors (agricultural, industrial and services sectors) of the economy. The objective was to examine whether unemployed youths can quickly move from one sector (say, the industrial sector) to another (say, the agricultural sector or services sector). The duration of youth unemployment generally, and that of youth unemployment as a result of unsatisfactory working conditions (frictional unemployment), specifically will reduce in a situation where employment spillovers exist from one sector to another. On the other hand, the PARDL model was used to examine the determinants or causes of youth unemployment in Nigeria and in each of the three sectors specifically. In this model, cyclical factors (economic growth) and various structural factors (amenable to changes in fiscal and monetary policies) are considered to identify whether the nature of youth unemployment in Nigeria is cyclical or structural.

The results validate the classical theory of employment given the finding in the short run that labour market flexibility (also, wage flexibility) and lesser labour laws and regulations reduce youth unemployment because of fewer regulations to prevent wages from falling downwards. However, the results show that the increased flexibility causes youth unemployment to increase in the long run. This is attributed to continued deregulation of the labour market or the non-enforcement of extant labour regulations, which allows wage rates to fall below the equilibrium and the minimum wage in many sectors of the economy, including education, health, manufacturing, and others.

Some causes of youth unemployment in Nigeria are identified. First is skill mismatch, as evidence shows in the inability of youths leaving employment in one sector to get absorbed in another sector. This suggests that the nature of youth unemployment in Nigeria could be partly frictional. This indicates that the viewpoint of the Nigerian government about the nature of youth unemployment and the emphasis on developing new skills is not entirely wrong. This result partly supports the finding of some studies on graduate unemployment in Nigeria, which emphasises skill mismatch as the cause of graduate unemployment in Nigeria ((Pitan & Adedeji, Skills Mismatch among University Graduates in the Nigeria Labor Market, 2012), (Pitan & Adedeji, Demographic Characteristics as Determinants of Unemployment among University Graduates in Nigeria, 2016);(Aminu, 2019)). 

In addition, our results show that youth unemployment is not influenced by economic cycles but by structural macroeconomic factors such as government capital expenditure, real exchange rate, interest rate and trade openness, suggesting that youth unemployment is structural. Specifically, in the long run, low capital expenditure was found to cause youth unemployment, as higher government capital expenditure positively influences youth employment. A low real exchange rate causes youth unemployment, as a higher real exchange rate positively influences youth employment. The high-interest rate contributed to high youth unemployment, as the interest rate negatively affects youth employment. Also, low trade openness was found to promote youth unemployment, as higher trade openness promotes youth employment. Our result that high inflation is not among the factors causing youth unemployment in Nigeria supports the finding by (O’Nwachukwu, 2017), who could not establish the inflation rate as a factor responsible for unemployment in Nigeria.

From our sector-by-sector analysis, our results show that high government debts cause youth unemployment in the agricultural and industrial sectors, as government debts negatively affect youth employment in both sectors.  Similarly, high Value Added Tax (VAT) was found to cause youth unemployment in the agricultural and industrial sectors. VAT has a negative significant effect on youth employment in both sectors. In the services sector, youth unemployment is motivated by low government capital expenditure, as low government capital expenditure was found to reduce youth employment in the sector.   

From the results obtained in this study, there is evidenced that youth unemployment in Nigeria is non-cyclical, partly frictional and primarily structural. The fact that the Nigerian government has mainly focused on youth unemployment policies to solve frictional youth unemployment may explain why youth unemployment is increasing despite government efforts. Our result suggests that not providing adequate and sustainable economic infrastructure, high interest rates, closure of borders, high real exchange rate and increased taxation are hostile to the private sector and contribute to the rising youth unemployment in Nigeria. Thus, a demand-side subsidy programme that encourages the private sector to employ youth may be considered the solution to the youth unemployment problem in Nigeria. The demand-side subsidy programme has been recognised as effective in dealing with structural unemployment. A workable framework could be borrowed from South Africa, Spain and India. Much easing monetary and fiscal policies would also be helpful, as this will enhance private sector productivity and promote youth employment in the country.

In addition, the current policy of providing vocational training, encouraging youth entrepreneurship and paying stipends to youth for community services could be augmented by integrating it into a national internship scheme practised in some emerging economies such as Botswana, Kenya, South Africa and Ghana. This provides a more efficient approach to tackling Nigeria’s frictional youth unemployment problem.

Centre for Petroleum Energy Economics and Law (CPEEL)

Mr Vice-Chancellor, Sir, my involvement in the CPEEL project is as far back as its inception in 2011 as an Associate Lecturer. In 2018, I was formally appointed a member of the CPEEL Management Team to contribute to the Centre’s vision of achieving its goal of becoming a leading Centre of Excellence for Energy Law and Energy Studies in Sub-Saharan Africa. It was a great privilege to be part of this winning team led by The Principal Investigator, Prof. Akin Iwayemi and the Co-Principal Investigator, Prof. Adeola Adenikinju. My appointment by the Vice-Chancellor as the Head of the Department of Minerals, Petroleum, Energy Economics and Law (DMPEEL) in 2021 was possible because of the confidence and trust reposed in me by the Founding Fathers (unfortunately, no Founding Mothers!). The multi-disciplinary nature of CPEEL, both in concept and practice, provides an excellent opportunity to unlock the extraordinary power of Statistics further. The idea of bringing together economists, engineers, lawyers, finance experts, scientists and other professionals in diverse fields to address energy issues holistically and collaboratively is enviable. Also, the self-financing approach makes it less reliant on the university for funding. However, the overtly centralised financial system operating in the university is a significant threat to growth, creativity and innovation, and this has continued to stifle the department’s progress. The creation of the Faculty of Multidisciplinary Studies in 2019 led to the emergence of the DMPEEL as an integral and offshoot of CPEEL. It is still difficult to say whether this move was thoughtfully made or done in bad faith.

Apart from teaching and supervising masters, M.Phil and Ph.D. students at CPEEL/DMPEEL, I have contributed significantly to knowledge production and harnessing the power of statistics in energy studies. Please let me share two crucial research outputs and one ongoing funded project in this lecture.

Perspectives, Impacts and Policy Responses to COVID-19 Pandemic (Volume I & II)

The Coronavirus (COVID-19) that took over the world with a storm in early 2020 presented an unprecedented shock to global and national economies. Nigeria was not insulated from this shock. The idea of CPEEL documenting the macroeconomic and sectoral impacts of these shocks on Nigeria was mooted by the dynamic and erudite Professor Adeola Adenikinju, the then Director of CPEEL. The CPEEL Management bought the idea promptly and wholly, as evidenced by the email in the textbox (see the Extract below). Leveraging on his connection with the Central Bank of Nigeria, financial support for the book’s publication was secured. Adeola Adeninkiju edited the contributed book, and the co-editors are Akin Iwayemi, Olusanya E. Olubusoye and Olugbenga A. Falode. With 25 chapters, written by 40 authors and a supplementary volume of 7 chapters produced by 18 students of the Centre, it is undoubtedly the single, most comprehensive analytical compendium on the COVID-19 experience in Nigeria (Adenikinju et al., 2020) for more detail.

In the book, we (Olubusoye & Ogbonna, 2020) analysed the impact of the emergence of COVID-19 on the macroeconomic fundamentals in Nigeria and made projections for economic growth. Specifically, we analysed the behaviour of macro aggregates such as oil price, exchangerate, and All Share Index (ASI) since the emergence of COVID-19. Also, we analysedthe assumptions and estimates in the 2020 Appropriation Act and how they were affected by the global pandemic. Finally, we made projections for the quarterly gross domestic product (GDP) growth rate and the annual growth rate for the year 2020. Our results show COVID-19 negatively impacts global oil prices, the Naira/USD exchange rate, Nigeria all share index, inflation and growth rate, with these impacts worsening as the pandemic lingers. The country’s budget estimate is also negatively affected, given the large changes between the budget assumptions and stance during the COVID-19 pandemic. Consequently, more than 50% of the country’s budget would have to be funded by external borrowing, hence increasing the debt burden of the country further. The budget deficit has alarmingly increased by approximately 138% (See Table 10).Economic growth in the first quarter was positive, and its corresponding probability of recording a negative growth minute. The probability of recording negative growth is projected to increase in subsequent quarters, with the third and fourth quarters having a sure probability of 1. Nigeria will likely experience negative growth in the third and fourth quarters of 2020 and may slide into recession by 2020. Our projection of a -3.5% economic contraction aligns with the IMF’s -3.4%. Also, we find crop production in the agriculture sector to significantly impact the variability of the real GDP growth rate; hence, it is a plausible preference as a tool for recovering from the impending recession.

Table 10: 2020 Budget Assumptions and Estimates of Macroeconomic Fundamentals

Source: Compiled by the authors (from Premium Times, Bloomberg and NBS)

Modelling Renewable Energy Consumption in Africa

Based on my experience supervising students’ projects, I have learnt that showing love, patience, understanding, and careful guidance can bring out the best in even the most challenging students. Two exceptional publications, which examined renewable energy consumption patterns in Africa, were produced with my doctoral students in CPEEL and Statistics. These publications were of an extraordinary standard, and they were published in top-rated Elsevier journals in the field of Renewable Energy (CiteScore: 16.1 and Impact Factor: 8.7) and Energy (CiteScore: 14.9 and Impact Factor: 9.0).

The increasing global demand for energy security and sustainable development necessitated the need for a paradigm shift from fossil fuel energy sources to renewable energy sources in Africa. There is a dearth of information on the current pattern of renewable energy consumption and its key drivers in Africa. In the first paper, we employed panel methodology to analyse renewable energy consumption in Africa. The study provided valuable insights into the trends, patterns, and factors influencing the utilisation of renewable energy sources in Africa. The methodology used combines fixed and random effect models and the use of the Hausman test to choose the appropriate model. We identified the determinants of renewable energy consumption in five (5) of the most populous and largest African countries in the five regions of Africa from 1990 to 2015, such as oil rents, coal rents, carbon intensity, natural gas rents, and energy intensity. It was found through the study that using GDP alone to measure a country’s welfare and growth is not sufficient, unlike what most researchers have done. This is because GDP does not always accurately reflect the welfare of a country. The study suggests using energy intensity as a better tool to measure a country’s welfare. Energy intensity measures the efficiency of a country in using energy, which is a good indicator of its overall welfare. (Olanrewaju, Olubusoye, Adenikinju, & Akintande, A panel data analysis of renewable energy consumption in Africa, 2019).

The increasing concern over global warming and energy security has rejuvenated renewable energy as the most vibrant option for sustaining future energy needs. The second study (Akintande O. J., Olubusoye, Adenikinju, & Olanrewaju, 2020) developed a renewable energy consumption model using annual data spanning between 1996 and 2016 in the five most populous African countries (Ethiopia, South Africa, Nigeria, DR Congo, and Egypt). The driving factors investigated were categorised into three broad areas following the existing literature on the subject. These include macroeconomic, socioeconomic, and institutional variables. Altogether, thirty-four predictor variables are analysed. The study employed Bayesian Model Averaging (BMA) procedures to account for the uncertainty associated with model choice and variable selection. The analysis results indicate that population growth, urban population, energy use, electric power consumption, and human capital are the main determinants of renewable energy consumption in the selected countries. Also, an increase in any of these determinants (population growth, urban population, energy demand/use, electricity power demand/consumption) causes an increase in renewable energy consumption.

Intra-Africa Mobility ACADEMY Project Scholarship

The Intra-Africa Mobility ACADEMY Project was the brainwork of Prof. Adeola Adenikinju, and it was a major project at CPEEL that I was privileged to coordinate. ACADEMY refers to African Trans-Regional Cooperation through Academic Mobility. It is an Intra-Africa Academic Mobility Scheme that facilitates the exchange of staff (academic and non-academic) and students between higher education institutions in African countries. Coordinated by the Université de Tlemcen, and financed by the European Commission under the Intra-Africa Academic Mobility Scheme, the project ACADEMY gathers four other African Higher education institutions namely, University of Cape Coast (Ghana), Kenyatta University (Kenya), University of Ibadan (Nigeria), and the University of KwaZulu-Natal (South Africa), as well as a European technical partner, namely Universidad do Porto (Portugal), in addition to two associate partners; Ministère de l’Enseignement Supérieur et de la Recherche Scientifique (Algeria) and the Pan African University for Water and Energy Sciences (Algeria).

The ACADEMY project started in 2018 and admitted the first set of scholars in the 2018/19 academic session.  Amongst others, the project contributed to the promotion of cross-regional continental integration, sustainable and inclusive development. It created an open ^[10]Platform as a key instrument for innovation, knowledge sharing and dissemination of good practices within the participating institutions. The focus of the ACADEMY Project was on the mobility of students, academic and administrative staff between the partner universities, as well as all the other universities across the African continent thereby encouraging creating and sustaining Communities of Practice by students, academic and administrative staff, as well as developing opportunities for mutual learning, intercultural understanding, and exchange of know-how and best practices. The project aimed to address the challenges of the African continent that arise from the rapid increase in the number of students, the need for improving the quality of higher education, the shortage in job creation and brain drain, gender equality, environmental sustainability and poverty.

During my coordination of the project at CPEEL, the African Trans-Regional Cooperation through Academic Mobility (ACADEMY) Programme provided scholarships to 8 Master’s students from UI to undertake 6 months of credit-seeking mobility, 7 Doctorate students from UI to undertake 6-10 months of credit-seeking mobility, and 3 staff members from UI to undertake exchange visits lasting from 1-3 months to one of the five partner universities. Additionally, UI hosted 9 Master’s and 5 Doctorate students, as well as 2 staff members, from the other four partner universities between 2018 and 2023. These students came from Kenyatta University (KU) in Kenya, the University of Cape Coast (UCC) in Ghana, the University of Tlemcen (UT) in Algeria, and the University of KwaZulu-Natal (UKZN). I thank the Office of Deputy Vice-Chancellor (RISP), Office of International Programmes (OIP), and UI Postgraduate College for their support.

During my coordination experience, I got the chance to showcase the Power of Statistics. With the assistance of Mr Olalekan Akintande (now Dr), we prepared a statistical report that was submitted to the European Union (EU) highlighting the impact of the project on the international student population in the partner universities. One of the most significant findings was that the University of Ibadan (UI) had the least number of international students enrolled in Masters and Doctorate programmes, with only 0.17% as of June 2023, as compared to the other four partner universities. This finding raises questions about the UI campaign for internationalisation (see Table 11).

Internationalisation of student enrolment in postgraduate studies is essential for several reasons, including diversity of perspectives, enhanced learning experience, global networking, cultural exchange, intercultural competence, contribution to the research ecosystem, diplomacy, interconnectedness, and significant economic contributions to the host country. International students often provide financial support to the host country’s economy through tuition fees, accommodation expenses, and other related expenditures. This financial support helps fund educational institutions, contributes to local economies, and supports the growth of educational infrastructure.

Table 11:Total Postgraduate and International Students Enrolments

Consultancy Services

Mr Vice-Chancellor, I have provided various statistical services at continental, national, state, and university levels. I would like to share one such service available to the public. While working at Addis Ababa University, I contributed to developing performance indicators for the Second Decade of Education for Africa (2006 – 2008) at the continental level. The report (Olubusoye & Khellaf, Olubusoye, O.E. and Khellaf, B. ) Performance Indicators for the Second Decade of Education for Africa, 2008) was a consultancy project carried out for the African Union (AU) through the Department of Human Resources, Science, and Technology. The Education Management Information System (EMIS) was the critical element of the second decade of education in Africa. This was because the evaluation of the first decade of education in Africa had revealed a significant lack of data on education. Additionally, the continent came to the realisation that one cannot plan for what one cannot measure. Measuring progress, success, and achievement is essential to ensuring continuous improvement and intervention during implementation.

The summary of the breakdown of the matrix of activities for each focus area in the AU Plan of Action (POA) is shown in Table 12.

Table 12: Summary Statistics of the Matrix of Activities in the AU Plan of Action

Source:(Olubusoye & Khellaf, 2008)

As shown in Table 12, there are 29 thematic or priority areas, 52 objectives, 90 actions or interventions, 68 strategies and 153 indicators in the AU POA. In our work, we proposed ninety (90) performance indicators based on the fundamental goals, priority areas and objectives of the plan. For each indicator, a definition, purpose, formula, data required, data source and, where applicable, the limitations were stated.

For this lecture, I will like to share with you some of the indicators proposed for higher education. As specified in the AU POA, the goal of higher education is stated as follows:

Complete revitalisation of higher education in Africa, with the emergence of strong and vibrant institutions profoundly engaged in fundamental and development-oriented research, teaching, community outreach and enrichment services to the lower levels of education; and functioning in an environment of academic freedom and institutional autonomy, within an overall framework of public accountability.

In line with this goal, we categorised our indicators into outcome, process and output indicators. These are presented in Tables 13 and 14, respectively.I strongly recommend that the University of Ibadan Management and the National Universities Commission (NUC) adopt some of these indicators to measure academic standards. The Academic Planning Office should be saddled with the task of maintaining a comprehensive database of requisite data needed for computing the indicators. Let it be part of the university’s annual report.

Table 13: Performance Indicators for Monitoring Higher Education (Outcome Indicators)

Indicator	Purpose	Formula
1. Secondary to Higher Education Transition Rate (SHETR)	To measure the capacity of higher education institutions (HEIs) to give access to secondary school graduates	No. of secondary school graduates who gained admission to HEIs No. of secondary school graduates
2. Higher Education Student Completion Rate (HESCR)	To measure an institution’s efficiency and quality	No of the students who completed the HE programme with or without repetition No of the students who enrolled for the programme
3. Government Higher Education Funding Index (GHEFI)	To measure the level of government financial support for higher education	Total government expenditure on HE GDP
4. Higher Education Funding Index (HEFI)	To measure the level of support for higher education	Total expenditure (government & others) on HE GDP
5. Higher Education Internationalization Index (HEII)	To measure the competitiveness and credibility of higher education institutions and their programmes	No. of foreign students and staff in HEIs No of staff in HEIs
6. Higher Education Continental Integration Index (HECII)	To measure the mobility trend of students and academics within the continent.	No. of foreign students and staff from Africa in HEIs Total number of staff in HEIs
7. Research Degrees Completion Rate (RDCR)	To measure the potential capacity for research and knowledge production.	No. of Ph.D. degree graduates No. of candidates enrolled for PhD

8. Academic Staff Quality Index (ASQI)	To measure the quality of teaching staff in higher education institutions	No. of staff with a Ph.D. degree Total number of staff
9. Higher Education Institutions Research Publication Index (HEIRPI)	To measure knowledge production and quality of higher education institution	Total number of peer-reviewed publications by all academic staff Total number of academic staff
10. Higher Education Unit Cost (HEUC)	To measure the average expenditure per student in higher education	Total expenditure on HEIs Total number of students
11. Research Funding Index (RFI)	To measure the trend of research funding to universities and research institutions annually	Total amount allocated for research Total university expenditure
12. HEI Research Consultancies Index (HRCI)	To measure the involvement of higher education institutions in knowledge production and development	Total number of consultancies undertaken by the universities Total number of universities
13. HE Science, Mathematics and Technology Gender Index (HEMTGI)	To measure gender gap in science, mathematics and technology in higher education institutions	No. of Ph.D. degree graduates No. of candidates enrolled for Ph.D.
14. Graduate Employment Rate (GEmR)	To measure the relevance of the higher education programmes to national developmental needs	No. of HEIs graduates with employment Total number of graduates
15. HE Academic Attrition Rate (HEAAR)	To measure the environment for sustainable academic activities	No. of academic staff who left for any reason Total number of academic staff
16. HE Gross Enrolment Ratio (HEGER)	To know the capacity of higher education in each country	No. of students enrolled in HEIs regardless of age Population of the age group for HE
17. Higher Education Participation Rate (HEPR)	To measure the access given to youths in higher education institutions	Population of the age group 16-25 enrolled in HEIs Population of the age group 16-25

Table 14: Performance Indicators for Monitoring Higher Education (Process and Output Indicators)

Indicator	Purpose	Formula
1. HE Policies Index (HEPI)	To monitor how countries promote policies that facilitate the revitalisation and delivery of quality HE	No. of countries with approved HE documents Total number of countries
2. HE Patent/Copyright Index (HEPCI)	To measure HE contribution to high-level knowledge production	No. of approved patents and copyrights No. of applications for patents and copyrights
3. HE Research Funding Index (HERFI)	To know the number of research funding agencies available for scholars and researchers in HEIs	Number of research agencies in each country
4. HEI Collaboration Index (HEICI)	To know the extent of networking and collaboration among HEIs	No. of agreements signed among HEIs No. HEIs involved
5.HEIs Joint Programme Index (HEIJPI)	To know the existence of networking and collaboration among HEIs	Number of joint programmes delivered or executed with other higher education institutions
6. African Academic Journals Index (AAJI)	To measure knowledge production and research output dissemination	Number of African journals published in a country
7. HE Harmonization Index (HEHI)	To measure the level of Harmonization of programmes among the HEIs	No. of institutions participating in credit transfer No. of HEIs
8. HE Quality Assurance Index	To show how quality is monitored by the higher education institutions	No. of HEIs with quality assurance unit Total number of HEs
9. HE ICT Index	To gauge the extent of usage and promotion of ICT and e-learning	No. of HEIs using ICT and e-learning Total number of HEIs
10. HEIs Governance Index (HEIGI)	To show the level of governance, leadership and management in HEIs	No. of HEIs with current strategic plans and annual reports Total number of HEIs
11. HE Contribution to Development Index (HECDI)	To show the contributions of HEIs to raising quality and efficiency in the lower levels of education	Number of research publications and policies oriented reports by higher education institutions focusing on primary and secondary education issues

Mentorship and Doctorate Supervision

Mr Vice-Chancellor, if you asked me what gives me the greatest joy in this university, my answer would be quick and precise – the Ph.D. students I have supervised. I derive immense satisfaction from contributing to developing human capital in Nigeria and fostering the growth of skilled professionals capable of harnessing the power of statistics. My goal is to inspire others to follow suit and create a sustainable talent pipeline to drive innovation and data analytics progress toward solving societal problems. I firmly believe that by investing in human resources, we can accelerate economic growth, foster social development, and address critical societal challenges. I am committed to leveraging my expertise and experience to lead the charge in this pursuit and make a meaningful impact without making noise.

I feel highly privileged to have successfully guided 17 Ph.D. students to completion of their programme; they include 7 females and 10 males and spread across two departments/faculties in the university: 12 from the Department of Statistics, Faculty of Science and 5 from the Department of Minerals, Petroleum, Energy Economics and Law (DMPEEL), Faculty of Multidisciplinary Studies (see Table 15). They are all incredible, resourceful, teachable, and impactful. Three of them are colleagues in the Department of Statistics, and two have already reached the professorial rank. Dr Olalekan Akintande, one of them, was awarded the prestigious Microsoft Research PhD Fellowship for 2021-2022 for his thesis titled “Confirmatory Cross-Validation as A Revisit of Machine Intelligence Pre-In-Post Processing Design and Fairness Quantification Metric”. I have many good stories to share about each one of them, but I would like to speak about one.

Dr Olufemi Adesiyan is someone I found to be humble and determined. We were classmates and completed our B.Sc. at the University of Ilorin in 1991, followed by our M.Sc. at the University of Ibadan. We both began teaching at the Department of Statistics, The Polytechnic Ibadan. It takes a great deal of humility for someone to submit to a contemporary as a supervisor to earn a doctorate degree. I greatly respect Dr Olufemi Adesiyan for his exemplary attitude and for following instructions and directives without prejudice.

Please refer to the table below for details on my Ph.D. students, including their names, gender, the department/center from which they graduated, the year they completed their PhD, and their current designation. Apart from those I supervised or co-supervised, I have mentored countless others. In this category, I cannot but mention Dr OlaOluwa Simon Yaya. I boldly say he is one of the promising and erudite scholars in the Department of Statistics.

Table 15:Ph.D. Students Supervised by Prof. O. E. Olubusoye

Ongoing Research Project: Modelling Grid Electricity Demand Using Artificial Intelligence

During my tenure as the Head of DMPEEL, several research groups were created, including the one tagged CRG-Energy Statistics and Modelling. This group comprised mostly master’s and doctorate students from both Statistics and DMPEEL. The members include me as the Principal Investigator (PI), Lucy Nwobi (Co-PI), Mr Abayomi Daramola, Kayode Ajulo, Precious, Muhammed EMMANUEL, Omosalewa Tinu ADEYANJU, Boluwatife Joshua OYEBADE. I provided the needed leadership and guidance in responding to the Call for Proposals by the Responsible Artificial Intelligence Network for Climate Action in Africa (RAINCA)^[11]. At the end of the rigorous selection process, RAINCA awarded Grant Number – 109705-001/002 for our project on Modeling Grid Electricity Demand Using Artificial Intelligence.The fund for the project is administered by the West African Science Service Centre on Climate Change and Adapted Land Use (WASCAL) based in Accra Ghana. The project entails building a database needed for profiling grid-electricity demand in Nigeria and developing a suitable machine-learning algorithm that can help predict future demand based on macroeconomic aggregates, climatological variables, grid infrastructure, and other potential drivers.

Profiling grid-electricity demand for a country like Nigeria has several important benefits. Understanding the demand patterns can help in the effective planning and management of the electricity grid, ensuring a reliable and stable supply of electricity to meet the needs of the population and support economic growth. By analysing the demand patterns, policymakers and energy planners can determine where and when to invest in new power generation capacity, transmission lines, and distribution systems. This enables them to efficiently allocate resources and improve the overall infrastructure of the electricity grid. Profiling grid-electricity demand facilitates the implementation of demand response programs. These programs incentivise consumers to adjust their electricity usage during peak demand periods by offering lower tariffs or other benefits. By managing demand in this way, Nigeria can better balance supply and demand, reduce strain on the grid, and potentially avoid the need for costly infrastructure upgrades.

Two capacity-building workshops on machine learning techniques were organised through the project. Additionally, two master’s dissertations at DMPEEL, a male and a female, were funded to ensure gender balance. Two team members were also supported through the grant to participate in two international conferences. The first was the AfricAI Conference 2023, and the second was a Proposal Development Training organised by RAINCA, both in Kigali, Rwanda.

Recommendations

For University Management and the National Universities Commission (NUC)

• Universities should prioritise and actively promote Statistics by funding and creating an enabling environment for the Department of Statistics to provide free and mandatory statistical assistance to the university community.
• The University of Ibadan Management and the National Universities Commission (NUC) should use relevant performance indicators (refer to Tables 13 and 14 in the inaugural lecture) to measure academic standards.
• The University of Ibadan Management should give special attention to internationalisation and walk the talk. A premier university should truly reflect a global community of intellectuals.

For the Government

• Declaration of one day every year to celebrate Nigerian Statistics Day (NSD) to stimulate statistical consciousness among citizens. I suggest 21^st December of every year for this celebration. The celebration will bring together key stakeholders in the Nigeria project to demand evidence-driven policies and massive investment in data production, accessibility, and dissemination. This will bring rapid and sustainable national development.
• The government should adopt a demand-side subsidy (public-private partnership) that encourages the private sector to gainfully engage the teaming youth population. This would simultaneously promote productivity and reduce unemployment.

For Everyone

• Strive for Statistical Alertness by becoming Statistically Literate.

Conclusion

In this lecture, I emphasised the importance of data and how statistical methods help to generate evidence from it. I aimed to establish the connection between data, statistics, and science and their crucial roles in advancing knowledge and understanding in various fields. Data refers to raw or unprocessed information that is collected through observations, measurements, experiments, surveys, or other means. It can take many forms, such as numerical values, textual information, images, or audio recordings. In scientific research, data serves as the foundation for analysis and investigation. Statistics involves the collection, analysis, interpretation, presentation, and organization of data. It provides methods and tools to summarize, describe, and draw meaningful insights from data. Statistical techniques help scientists identify patterns, trends, relationships, and correlations within datasets, enabling them to make informed conclusions and decisions. Statistics possesses extraordinary power due to its ability to extract meaningful insights from complex data, providing a quantitative basis for decision-making. Statistics forms the foundation of machine learning (ML) and artificial intelligence (AI) by providing mathematical and probabilistic frameworks for understanding, modelling, and predicting data. Through statistics, we have harnessed the potential to make life-saving medical breakthroughs, understand and mitigate climate change’s impacts, and enable machines to understand language and images. We have bridged the gap between data and decision, uncertainty and clarity, and the present and future.

In my presentation to you, Mr Vice-Chancellor, I provided an overview of my teaching, research, and statistical collaborations that span a wide range of departments, centres, and faculties at the esteemed University of Ibadan. I highlighted my statistical activities that extend to national institutions such as INEC and CBN, as well as continental bodies like the African Union. Working alongside domain experts in fields such as Economics, Energy, and Political Science, as well as esteemed colleagues in Statistics, I demonstrated how statistics can be utilised to produce evidence for policy-making and societal transformation. Moreover, I discussed my contributions to providing quality leadership and mentorship to the next generation of data scientists within my department and beyond the university. I shared how we established an interdisciplinary statistical laboratory in the Department of Statistics, which has served as an engine of transformation in statistical capacity building for UI researchers and students from tertiary institutions across the country.

In conclusion, I submit that to attain genuine national development and progress, political leaders must harness the potential of statistics. By utilising statistical data to inform policies and agendas, Oyo State and other Nigerian states can experience a remarkable transformation. Similarly, for universities to excel in research and fulfil their commitment to knowledge and human resource production, Vice-Chancellors and other key administrators must unlock the exceptional power of statistics. We all must acknowledge the significance of data and equip ourselves with the necessary knowledge and skills to effectively unlock the power of statistics.

Acknowledgements

Throughout my academic and research pursuits, I have received an immense amount of support from a vast array of individuals. While it would be impossible to name everyone who has contributed to my success, I would like to express my gratitude to those who have played a significant role in my journey.

I will like to start my acknowledgement with my immediate and extended family. According to Romans 11:16, “If the root is holy, so are the branches.”I owe my beautiful, dutiful, virtuous, lovely wife and soulmate, Mrs Oluranti Grace Olubusoye, a huge gratitude. She has been extremely supportive, and prayerful in this journey. I cannot imagine going this far without her in my life. In the same vein, I thank my loving daughters, Tolu and Tutu for their understanding and cooperation. I thank the Kekereye Dynasty and the Aro Sons and Daughters from Awo-Ekiti under the High Chief Engr Abayomi Olubusoye for providing the humble beginning. I thank my father in the Lord, Pastor Joseph Olubusoye, for his love and care over the years.

I thank all my teachers and schoolmates from primary school to university. I recognise the support of all my schoolmates at Eyemote Comprehensive High School (ECHOS), Class 79 – 84. One of my favourite teachers in the school was Dr Joseph Owoeye. He has been tracking and monitoring my progress since then. “For even if you had ten thousand others to teach you about Christ, you have only one spiritual father. For I became your father in Christ Jesus when I preached the Good News to you.” (1 Corinthians 4:15, NLT). Just like the people of Corinthians, I have gotten many teachers but one academic father and Ph.D. supervisor, Prof. J.O. Iyaniwura. He supervised over a dozen Ph.Ds including my own successfully and I learnt a lot of lessons from him on Ph.D. supervision. His unwavering support and guidance throughout my academic career are immeasurable and have been instrumental in shaping my research interests and scholarship. I am also grateful to all my teachers and mentors at the University of Ibadan, amongst them are Prof. S.O. Olofin, Prof. Biyi Afonja, Dr (Mrs) T. L. Johnson, Prof. T.A. Bamiduro, Prof. Akin Iwayemi, Prof. A. Adenikinju, Prof. A.A. Sodipo, and Prof. Dahud K. Shangodoyin, Hon. Commissioner for Works and Transport, Oyo State, late Prof. G. N. Amahia whose advice and encouragement have been invaluable in my academic pursuits. I cannot forget all my erudite teachers and mentors, now retired, at the University of Ilorin: Professors O. S. Adegboye, E. T. Jolayemi, B. A. Oyejola, Peter Osainaye, I. O.  Oshungade and R. A. Ipinyomi. I specially recognise Baba Ajayi O. Oladejo, Former Director General of the Federal Office of Statistics (FOS), for continuously showering so much affection on me.

I am equally grateful to my dedicated retired teachers and mentors at the Department of Statistics, University of Ilorin. They are Professors O. S. Adegboye, E. T. Jolayemi, B. A. Oyejola, Peter Osainaye, I. O. Oshungade, and R. A. Ipinyomi.

Also, I express my appreciation to my Head of Department, Prof. O. I. Osowole, for his leadership and support and to all my colleagues and students in the Department of Statistics, whose collaboration and camaraderie have made my academic journey even more rewarding. I cannot but mention Prof. O. I. Shittu, Prof. Abosede A. Adepoju, and Prof. Angela Chukwu. I thank those I worked with on many projects and training at CEAR. I fondly remember Dr Alarudeen Aminu, Prof. Afees Adebare Salisu, Dr Afolabi Olowookere, Mr Gbenga Oni and others. Our partners at the Central Bank of Nigeria (CBN) are equally appreciated. I must mention Dr Mohammed Musa Tumala, Director of Statistics in the Bank, Dr Michael A. Adebiyi and Dr A. O. Adenuga, both from the Bank’s Research Department.

I thank my partners at the National Bureau of Statistics (NBS). Foremost is Prince Adeyemi Adeniran, the Statistician General (SG) of the Federation and Mr. Biyi Fafunmi, Director ICT Department.

I am also indebted to the Head of the Department of Minerals, Petroleum Energy Economics and Law (DMPEEL), Prof. Olugbenga Falode, as well as to all the staff and students in that department, for their support and encouragement during my headship of DMPEEL. I cannot but mention Mrs Mojisola Ogunmoye, the CPEEL’s Accountant and Mr. Daramola, the Programme Officer, for the extraordinary support you gave me during the period I served at CPEEL/DMPEEL. So many other people, such as Prof. Simisola O. Akintola, Prof. John Oluwole Akintayo, Prof. P. C. Obutte, Prof. Bayo Adaralegbe, Prof. Adegbenga Adekoya, Director of Innovation Lab for Policy Leadership in Agriculture and Food Security (PiLAF) and others, too numerous to mention, made the job easy for me. Prof. R. A. Oderinde, the U.I. Marc Arthur Liaison Officer, has been extremely wonderful as a mentor, leader, and motivator. It was highly pleasant working with you. Thank you, Sirs.

I owe immense gratitude to the LISA 2020 Global Network, particularly, Prof. Eric Vance, the Global Director, LISA 2020 Global Network and the entire Excos at the global level, including Prof, Olawale Awe, the Vice President for Engagement and Public Relations. I cannot forget the contributions of Ian Crandell, Tonya Pruitt, Kim Love, Matthew Druckenmiller, and a host of other prominent network members for contributing to the nurturing of UI-LISA. I thank all my colleagues and LISA Collaborators. I thank Prof. O. I. Osowole, Prof. A. A. Adepoju, Dr Oluwayemisi Alaba, Dr C. G. Udomboso, Dr Oluwaseun Otekunrin, Dr O. S. Yaya, Dr O. B. Akanbi, Dr S. O. Oyamakin, Dr A. T. Adeniran, Dr O. J. Akintande, Dr T. P. Ogundumade and all the UI-LISA. Dr Serifat Folorunsho served as the Administrative Officer for UI-LISA from its inception, and she worked passionately for UI-LISA’s sustainability. She is hugely committed and strong-willed.  I thank the Southwest Region Stat Labs under the leadership of Dr Ezra Gayawan, FUTA LISA.

I thank all my spiritual fathers and mothers from Christ Apostolic Church (CAC), Oke-Ife, Agbowo and all its branches in Alaja, Alakia, and Aho. I acknowledge the support of the Youth Associations, the Sunday School Teachers, the Choir members, the Board of Elders and the Ministers. I am constrained by time and space to list everyone. The grace of our Jesus Christ be with you all. Amen. (2 Thessalonians 3:18).

Through statistical collaborations, I have been given the honour of working in two faculties, Science and Multidisciplinary Studies. I appreciate the support provided by the latter, first under Prof. Isaac Albert and now under Prof. Wole Olatokun. However, I am eternally grateful to the Dean of Science, Prof. O. O. Sonibare, for choosing me to deliver this lecture. I thank all Committees at the Department of Statistics and Faculty of Science for their contributions. Many people assisted me in editing the manuscript for this lecture. I thank Prof. T. A. Bamiduro, Prof. Abosede A. Adepoju, Dr Oluwayemisi Alaba, Dr O. S. Yaya, Dr O. J. Akintande for reading through and making helpful suggestions. I thank the Director of the University of Ibadan Publishing House, Mr O. M. Oladejo, for his painstaking and thorough editing of the manuscript. The person behind the PowerPoint preparation is Dr Ahamuefula Ephraim Ogbonna. I thank you all for your sacrifices.

Finally, remember that unlocking the extraordinary power of statistics can work wonders with data. Thank you, Mr Vice-Chancellor and distinguished ladies and gentlemen, for your attention.

• Editor’s Note: Olubusoye , a Professor of Econometrics Faculty of Science University of Ibadan, and delivered this inaugural lecture at the University of Ibadan On Thursday, December 21, 2023  

---

[1]https://www.brandtimes.com.ng/data-is-life-campaign-airtel-repositions-as-nigerias-network-of-choice/

[2]https://www.post2020hlp.org/the-report/

[3]https://www.timeanddate.com/holidays/un/world-statistics-day

[4]https://www.uneca.org/stories/african-statistics-day-2023#:~:text=African%20Statistics%20Day%20is%20an,of%20social%20and%20economic%20life.

[5]https://www.lisa2020.org/labs

[6] Unemployment, youth total (% of total labor force ages 15-24) (modeled ILO estimate) – Retrieved July 28, 2019.

[7] See https://reliefweb.int/report/world/global-terrorism-index-2018

[8] See https://www.bbc.com/pidgin/tori-46459641

[9] See Premium Times, https://www.premiumtimesng.com/news/top-news/241160-nigeria-ranks-3rd-global-internet-crimes-behind-uk-u-s-ncc.html

[10]https://myacademy.africa/

[11]https://rain-ca.org/about/