African Statistical Journal Journal Africain De Statistiques

Éditorial African Statistical Journal Journal africain de statistiques

1. Quantification of Inflation Expectations from Business Tendency Survey Data in Uganda Kenneth Alpha Egesa

2. Cigarette Smoking in Tanzania-Prevalence and Determinants Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu

3. Statistical Literacy for National Development Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

4. Statistical Indicators for Measuring Good Governance in Africa Dahud Kehinde Shangodoyin

Volume 21 – September / septembre 2019 African Development Bank Group Groupe de la Banque africaine de développement

Journal africain de statistiques, numéro 21, septembre 2019 1 Designations employed in this publication do not imply the expression of any opinion on the part of the African Development Bank or the Editorial Board concerning the legal status of any country or territory, or the delimitation of its frontiers. The African Development Bank accepts no responsibility whatsoever for any consequences of its use.

African Development Bank Group Avenue Joseph Anoma 01 BP 1387 Abidjan 01 Côte d’Ivoire Tel: (+225) 20 26 10 20 Email: [email protected] Internet: http://www.afdb.org

Les dénominations employées dans cette publication n’impliquent, de la part de la Banque africaine de développement ou du comité de rédaction, aucune prise de position quant au statut juridique ou au tracé des frontières des pays. La Banque africaine de développement se dégage de toute respon- sabilité de l’utilisation qui pourra être faite de ces données.

Groupe de la Banque africaine de développement Avenue Joseph Anoma 001 BP 1387 Abidjan 01 Côte d’Ivoire Tel: (+225) 20 26 10 20 Courriel: [email protected] Internet: http://www.afdb.org

@AfDB/BAD, 2019– Statistics Department / Département des statistiques African Statistical Journal Journal africain de statistiques

Volume 21 September / septembre 2019 Contents

Editorial ...... 6

Acknowledgments ...... 10

1. Quantification of Inflation Expectations from Business Tendency Survey Data in Uganda Kenneth Alpha Egesa ...... 12

2. Cigarette Smoking in Tanzania-Prevalence and Determinants Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu ...... 31

3. Statistical Literacy for National Development Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya ...... 47

4. Statistical Indicators for Measuring Good Governance in Africa Dahud Kehinde Shangodoyin ...... 67

Call Of Papers ...... 80

Editorial policy ...... 84

Guidelines for manuscript preparation and submission ...... 87

4 The African Statistical Journal, Volume 21, September 2019 Table des matières

Éditorial...... 8

Remerciements ...... 11

1. Quantification des anticipations d’inflation à partir des données de l’en- quête sur les tendances des entreprises en Ouganda Kenneth Alpha Egesa ...... 12

2. Le tabagisme en Tanzanie: Prévalence et déterminants Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu ...... 31

3. Connaissances en statistique pour le développement national Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya ...... 47

4. Indicateurs statistiques pour mesurer la bonne gouvernance en Afrique Dahud Kehinde Shangodoyin ...... 67

Demande de soumissions d’articles ...... 82

Ligne éditoriale ...... 85

Instructions pour la préparation et la soumission des manuscrits ...... 90

Journal africain de statistiques, numéro 21, septembre 2019 5 Editorial

We welcome our readers to Volume 21 of the African Statistical Journal (ASJ). The Journal’s mission remains unchanged since inception: to serve as a forum and common platform, to share ideas on statistical development in Africa, and to stimu- late discussion and dialogue on key emerging issues. The ASJ attempts to reach out not only to practicing statisticians and economists among others in Africa, but also to those beyond the continent who are keen to make incremental value addition to evolving developmental issues affecting the African citizenry.

We begin this volume with an article entitled, “Quantification of Inflation Expecta- tions from Business Tendency – Survey Data in Uganda.” The article uses simple techniques for quantifying qualitative survey data on inflation expectations. The approach focused on firms’ responses regarding future selling prices. The trends in the numerical estimates of inflation expectations (quantitative estimates) were compared to the diffusion index for price expectations (qualitative estimates). The estimates from the two methods used did not differ significantly and it was not possible to draw any conclusion on the superiority of one method over the other.

The second article entitled “Cigarette Smoking in Tanzania – Prevalence and De- terminants,” demonstrates that smoking is on the decline in the developed world, and yet, the opposite is true in Africa. The article amplifies the fact that cigarette smokers are on average poor and unable to seek medical treatment from smoking induced illnesses. The results show that, smokers in Tanzania are more likely to be poor, elderly, less educated, farmers, and on average, male.

The third article entitled “Statistical Literacy for National Development” amplifies the critical role of statistical literacy as a pillar for national development. Using the case of Nigeria, the article advocates for collaborative work between the Nigerian Statistical Association (NSA) and the National Bureau of Statistics (NBS), emphasis is also placed on a national statistical center charged with the responsibility of generating new resources that are consistent with the changing times. As part of the process, a website with requisite resources to foster the culture of statistical literacy is suggested as one of the instrumental variables to promote advancement in national development with clear roles for professional statisticians, and policymakers.

The fourth and final article entitled “Statistical Indicators for Measuring Good Governance in Africa” places fundamental importance on the National Strategies

6 The African Statistical Journal, Volume 21, September 2019 Editorial for the Development of Statistics (NSDS) as a vehicle for realizing measurable governance statistical indicators to inform the evolving development agenda. The article recognizes the ever increasing demand for data, and singles out regional integration as one specific area that calls for comparable and coherent statistics at national, regional, and continental levels. The article bemoans the overreliance on donor funding for surveys in some African countries as it dampens prospects for sustainable statistical capacity across national statistical systems.

We hope you will find the latest volume of the ASJ both informative and stimula- ting. We would like to thank the contributors and the reviewers, as well as all those who have made this volume a reality. We encourage the African Statistical Com- munity to continue using the ASJ as an authoritative forum for sharing knowledge, topical issues worth exploring, including among others, establishing how national statistical systems are responding to data requirements for African Agenda 2063, and UN Agenda 2030, including practical steps to fill the data gaps as a continuous process going forward.

Dr. Charles Leyeka Lufumpa Professor Ben Kiregyera Co-Chair, Editorial Board Co-Chair, Editorial Board Statistics Department International Statistical Consultant, African Development Bank Kampala, Abidjan, Côte d’Ivoire. Uganda. Email: c. [email protected] Email: [email protected]

Journal africain de statistiques, numéro 21, septembre 2019 7 Éditorial

Nous invitons nos lecteurs à découvrir le volume 21 du Journal africain de statistique (ASJ). La mission du Journal n’a pas changé depuis ses débuts: servir de forum et de plate-forme commune, partager des idées sur le développement de la statistique en Afrique et stimuler le débat et le dialogue sur les principaux pro- blèmes émergents. L’ASJ tente non seulement de toucher des statisticiens et des économistes en exercice, notamment en Afrique, mais également avec ceux qui, au-delà du continent, souhaitent apporter une valeur ajoutée aux problèmes de dé- veloppement en constante évolution qui affectent les citoyens africains.

Nous commençons ce volume avec un article intitulé “Quantification des anticipations d’inflation à partir des données de l’enquête sur les tendances des entreprises en Ouganda”. Cet article utilise des techniques simples pour quantifier les données d’enquête qualitatives sur les anticipations d’inflation. L’approche a mis l’accent sur les réponses des entreprises aux prix de vente futurs. Les tendances des estimations numériques des anticipations d’inflation (estimations quantitatives) ont été comparées à l’indice de diffusion des anticipations de prix (estimations qualitatives). Les estimations des deux méthodes utilisées ne différaient pas de manière significative et il n’était pas possible de tirer une conclusion sur la supériorité d’une méthode sur l’autre.

Le deuxième article intitulé “Le tabagisme en Tanzanie - Prévalence et determinants” montre que le tabagisme est en déclin dans les pays développés, alors que l’inverse est vrai en Afrique. L’article amplifie le fait que les fumeurs de cigarettes sont en moyenne pauvres et incapables de demander un traitement médical en rai- son de maladies induites par le tabagisme. Les résultats montrent que, en Tanzanie, les fumeurs sont plus susceptibles d’être des pauvres, des personnes âgées, moins éduquées, des agriculteurs et, en moyenne, des hommes.

Le troisième article intitulé “Connaissances en statistique pour le développement national” amplifie le rôle critique la connaissance en statistique en tant que pilier du développement national. À l’aide du cas du Nigéria, l’article préconise un travail collaboratif entre l’Association nigériane de statistique (NSA) et le Bureau national de la statistique (NBS), l’accent est également mis sur un centre national de statistique chargé de générer de nouvelles ressources cohérentes avec les temps chan- geants. Dans le cadre de ce processus, il est suggéré de créer un site Web doté des

8 The African Statistical Journal, Volume 21, September 2019 Éditorial ressources nécessaires pour promouvoir la culture de la statistique, comme l’une des variables déterminantes pour la promotion du développement national, avec des rôles clairement définis pour les statisticiens professionnels et les décideurs.

Le quatrième et dernier article intitulé “Indicateurs statistiques pour mesurer la bonne gouvernance en Afrique” accorde une importance fondamentale aux Stra- tégie Nationale de Développement de la Statistique (SNDS) en tant que moyen de réaliser des indicateurs statistiques de gouvernance mesurables pour éclairer le programme des nouvelles tendances de développement. L’article reconnaît la demande toujours croissante de données et considère l’intégration régionale comme un domaine spécifique qui requiert des statistiques comparables et cohérentes aux niveaux national, régional et continental. L’article déplore le recours excessif au financement des enquêtes par les donateurs dans certains pays africains, dans la mesure où il réduit les perspectives d’une capacité statistique durable dans l’ensemble des systèmes statistiques nationaux.

Nous espérons que le dernier volume de l’ASJ sera à la fois enrichissant et stimu- lant. Nous voudrions remercier les contributeurs et les critiques, ainsi que tous ceux qui ont fait de ce volume une réalité. Nous encourageons la Communauté africaine de statistique à continuer d’utiliser l’ASJ en tant que forum faisant autorité pour partager des connaissances, des questions d’actualité méritant d’être explorées, notamment pour déterminer comment les systèmes statistiques nationaux répondent aux besoins en données de l’Agenda 2063 pour l’Afrique et de l’Agenda 2030 des Nations Unies, y compris prenant toutes les mesures nécessaires pour combler les lacunes des données en tant que processus viable pour l’avenir.

Dr. Charles Leyeka Lufumpa Professeur Ben Kiregyera Co-président du comité de rédaction Co-président du comité de rédaction Département des statistiques Consultant international de statistique, Banque africaine de développement Kampala Abidjan, Côte d’Ivoire. Ouganda. Email: c. [email protected] Email: [email protected]

Journal africain de statistiques, numéro 21, septembre 2019 9 Acknowledgments

Co-Chairs of the Editorial Board: Dr. Charles Leyeka Lufumpa, Director, Statistics Department, African Develop- ment Bank Group, Abidjan, Côte d’Ivoire.

Professor Ben Kiregyera, International Statistical Consultant, Kampala, Uganda.

Editor in Chief: Ben Mungyereza, Manager, Statistical Capacity Building Division, African Deve- lopment Bank Group, Abidjan, Côte d’Ivoire.

Production Editor: Rees Mpofu, Statistics Department, African Development Bank Group, Abidjan, Côte d’Ivoire.

English Editors: Dr. Tomi Adeaga, Consultant Editor, Vienna, Austria; Dr. Wangui wa Goro, En- glish Editor, Corporate Language Services, African Development Bank, Abidjan, Cote d’Ivoire.

Expert Reviewers: Sanjev Bhonoo, Statistician, Statistics Mauritius, Mauritius.

Simon M. Gaitho, Manager, Consumer Price Index (CPI), Kenya National Bureau of Statistics (KNBS), Nairobi, Kenya.

Asmerom Kidane, Professor of Economics and Statistics, Department of Econo- mics, University of Dar es Salaam, Tanzania.

Godfrey Makware, Manager, Industry, Mining and Energy Statistics, Zimbabwe National Statistics Agency (ZIMSTAT), Harare, Zimbabwe.

Nsubuga Vincent Musoke, Principal Statistician, Uganda Bureau of Statistics, Kampala, Uganda.

10 The African Statistical Journal, Volume 21, September 2019 Remerciements

Copresidents du comité de rédaction : Dr Charles Leyeka Lufumpa, Directeur, Département de la statistique Le Groupe de la Banque africaine de développement, Abidjan, Côte d’Ivoire

Prof. Ben Kiregyera, Consultant international en statistique, Kampala, Ouganda.

Redacteur en chef: Ben Mungyereza, Chargé de mission, Division du renforcement des capacités statistiques, Banque africaine de développement, Abidjan, Côte d’Ivoire.

Éditeur de production: Rees Mpofu, Département des statistiques, Banque africaine de développement, Abidjan, Côte d’Ivoire.

Éditeur anglais: Dr. Tomi Adeaga, Éditeur consultant, Vienne, Autriche

Dr. Wangui wa Goro, Éditeur anglais, Services linguistiques d’entreprise, Banque africaine de développement, Abidjan, Côte d’Ivoire .

Examinateurs experts : Sanjev Bhonoo, Statisticien, Statistics Mauritius, Ile Maurice.

Simon M. Gaitho, Directeur, Indice des prix à la consommation (IPC), Kenya National Bureau of Statistics (KNBS), Nairobi, Kenya .

Asmerom Kidane, Professeur d’économie et de statistique au département d’éco- nomie de l’Université de Dar es-Salaam en Tanzanie.

Godfrey Makware, Directeur des statistiques de l’industrie, mines et de l’énergie, Agence nationale de la statistique du Zimbabwe (ZIMSTAT), Harare, Zimbabwe.

Nsubuga Vincent Musoke, Statisticien principal, Bureau des statistiques de l’Ou- ganda, Kampala, Ouganda.

Journal africain de statistiques, numéro 21, septembre 2019 11 1. Quantification of Inflation Expectations from Business Tendency -Survey Data in Uganda

Kenneth Alpha Egesa 1

Abstract

This article has applied Carlson and Parkin (1975)’s subjective probability method and Pesaran (1984)’s regression method, based on the qualitative balance statistic to quantify qualitative survey data on inflation expectations collected monthly by the Bank of Uganda. The analysis used responses of firms on their expected selling prices in the period ahead. Both methods provided decent estimates of inflation expectations relative to realized inflation. Given the absence of quantitative estimates of inflation expectations, and despite their importance in generating forecasts considered by the monetary policy committee, it is recommended that going forward, both estimates can be used as part of the inputs in the inflation forecasting process.

KEYWORDS : Inflation expectations, Perception surveys, Monetary policy

Résumé

Cet article a appliqué la méthode des probabilités subjectives de Carlson et Parkin (1975) et la méthode de régression de Pesaran (1984), fondées sur la statistique du bilan qualitatif pour quantifier les données d’enquêtes qualitatives sur les anticipations d’inflation collectées mensuellement par la Banque d’Ouganda. L’analyse a utilisé les réponses des entreprises sur leurs prix de vente attendus dans la période à venir. Les deux méthodes ont fourni des estimations correctes des anticipations d’inflation par rapport à l’inflation réalisée. Compte tenu de l’absence d’estimations quantitatives des anticipations d’inflation et de leur importance pour générer les prévisions examinées par le comité de politique monétaire, il est recommandé qu’à l’avenir, les deux estimations puissent être utilisées dans le cadre du processus de prévision de l’inflation.

1 Director Financial Stability Department, Bank of Uganda; [email protected] The author is grateful to various seminar participants at the Bank of Uganda who heard evolving versions of this article. The views expressed in this article are entirely those of the author and are no way reflective of the views of the Bank of Uganda.

12 The African Statistical Journal, Volume 21, September 2019 1.Quantification of Inflation Expectations from Business Tendency -Survey Data in Uganda

1. INTRODUCTION

The Bank of Uganda adopted an inflation targeting lite regime in July 2011 after pursuing a monetary targeting regime for about twenty years. Some of the reasons advanced by the authorities for the change in the monetary policy regime included observed instability of the money demand function, improved responsiveness of the monetary aggregates to central bank interest rates, and the need to separate domestic securities issued for fiscal from those issued for monetary policy purposes. Five years later, real GDP growth has averaged 5.5 percent per annum, while inflation has remained low and stable and was estimated at 5.1 percent in 2015 after peaking at 21.7 percent in 2012, following the presidential elections. In the external sector, the current account balance as a share of GDP has reduced from 10.3 percent in 2011 to 8.5 percent 2015. The shift to inflation targeting entailed an overhaul of the statistical and analytical requirements for policy formulation purposes from a backward-looking approach to a forward-looking one. In particular, from the statistical point of view, much emphasis was placed on the introduction of more timely estimates for both real and external sectors to enable more accurate forecasting of future developments upon which policies would be made. In the real sector, a monthly composite index of economic activity was introduced and consumer and business confidence surveys were introduced to improve analysis on domestic developments.

While the use of the consumer and business confidence indices have improved economic analysis and aided policy formulation, more remains to be done with some of the elements collected during the survey. An important example is the inflation expectations data collected every month for the three-month period ahead. Much as this information is used in the derivation of the overall confidence indices, there has been no effort made to apply it towards the compilation of a quantitative time series of expected inflation. Inflation expectations are however an important input in -de riving comprehensive inflation forecasts, in addition to the usual factors that drive Uganda’s inflation such as food and industrial products output, exchange rates, interest rates, and fiscal deficit. According to monetary theorists, inflation expectations affect inflation through individuals bargaining over nominal pay increases and firms setting prices (Perera, 2008). For instance, if the view is that inflation will be per- sistently higher, then employees may seek higher nominal wages to maintain their purchasing power resulting in upward pressure on firms’ output prices and consumer prices. Similarly, if firms believe that inflation will be higher in the future, they

Journal africain de statistiques, numéro 21, septembre 2019 13 Kenneth Alpha Egesa

may raise prices, believing that the price increases will not result in a decline in the demand for their output.

The objective of this article therefore is to provide a methodology for compiling inflation expectations that can be adopted, using existing information to support monetary policy formulation. Specifically, this article discusses and applies some of the methods in the literature that can be used to quantify qualitative inflation expectations, and evaluates the accuracy and reliability of the estimates obtained. It is expected that the methods used will be adopted to augment the analytical reports prepared for the monetary policy committee.

The rest of this article is organized as follows: Section II reviews the literature on the use of confidence survey data for computing inflation expectations; Section III presents the data used for the study and a description of the methodology; Section IV presents the inflation expectation estimates, an assessment of their accuracy and reliability, and a discussion of the findings; and Section V provides a summary of the key findings and concludes.

2. LITERATURE REVIEW

2.1. Formation of inflation expectations

Expectations are the unobservable opinions about the future that individuals form in their minds (Gertchev, 2007). Since expectations are central to the analysis done in this article, it is important to review the theoretical constructs surrounding their formation. There are a number of expectations formation hypotheses in the literature although the two most common ones are the adaptive and rational expectations hypotheses. In the adaptive expectations hypothesis, at each new time period, the individual revises his expectation of the future price in view of his current expectation error i.e. the discrepancy between his expectation of the current price and the actual current price. Thus the current expectation of the new period is formed as the sum of the past expectations with the expectation error weighted by a coefficient of revision of expectations also sometimes referred to as the speed of adjustment (Gertchev, 2007). This can be expressed as:

(1)

14 The African Statistical Journal, Volume 21, September 2019 1.Quantification of Inflation Expectations from Business Tendency -Survey Data in Uganda

e Where P t is the actual current price, P t is the future expected price held in the current period (t) and δ is the coefficient of revision of expectations that is normally assumed to lie between 0 and 1. Through the Koyck transformation, it can be shown that expected future price is equivalent to an arithmetic average, with geometrically decreasing weights of the entire history of actual realized prices as follows:

Expanding this j-times yields:

This, in condensed form, becomes:

and since , then as

(2)

Alternatively, adaptive expectations can be defined in respect of price changes as the expected rate of change in prices that is revised per period of time in proportion to the difference between the actual rate of change in prices and the rate of change that was expected (Cagan, 1956).

In the rational expectations hypothesis, the individual uses all of the available per- tinent information when formulating his forecast of prices, interest rates and even government policies (Gertchev, 2007). Thus unlike the adaptive expectations hypothesis, in the rational expectations hypothesis, individuals use the available information on all relevant variables (Mlambo, 2012). Thus, assuming that the variable of interest is Y and the other related variables are X and Z, the hypothesis can be formulated as:

(3)

Where Ut is a random variable and the lagged variables are known at the time of forecasting which is at the end of period t-1, while the random variable is only known

Journal africain de statistiques, numéro 21, septembre 2019 15 Kenneth Alpha Egesa

at the end of period t. There are three main definitions of rational expectations in the literature (Gertchev, 2007 and Mlambo, 2012). In the first, all individuals need to possess the same expectations of the objective distribution, but on the average, the weighted arithmetic mean of the expectations is equal to the prediction of the relevant economic model (Muth, 1961). The second definition imposes that all individuals hold the same subjective probability distributions about future events which additionally coincide with the objective distribution. The third definition assumes that individuals make economically rational expectations in the sense that they search for and process information only to the point where the marginal cost becomes equal to the marginal gain (Feige and Pearce, 1976).

Both the adaptive and rational expectations models have received criticisms in almost equal measure in the literature. The main critique of the adaptive expectations hypothesis is that the formation of the expectations on the basis of only historical data on the variable of interest is less logically satisfactory since such expectations would not fully account for individual’s rationality (Gertchev, 2007). Thus, since adaptive expectations boil down to historical data, in effect, they are considered not to be forward-looking. This argument has however been discounted on the grounds that expectations cannot be divorced from experience. Other criticisms include the observations that errors of expectations may be correlated and the expectations may lag behind actual phenomenon when trends change. In addition, empirically it has been shown that values of coefficients may vary across groups of economic agents, individuals and over time (Mlambo, 2012). The general conclusion is that the adaptive expectations hypothesis has been seen as only partially accounting for individual’s rationality (Gertchev, 2007).

On the other hand, the rational expectations hypothesis has mainly been criticized on the basis of the general understanding that it is an optimal process that eventually eliminates systematic errors which is inconsistent with the view that rational expectations are borne out of uncertainty (Mlambo, 2012). It is argued that systematic errors cannot be eliminated where uncertainty is pervasive. The other criticism is that the rational expectations hypothesis does not contribute much towards economic discourse since it relies on the existence of some other model that does the prediction meaning that economic theory is developed independently of expectations (Gertchev, 2007). Nonetheless, a lot of empirical work informed by the two hypotheses has been done, the criticisms notwithstanding. Overall, more contemporary

16 The African Statistical Journal, Volume 21, September 2019 1.Quantification of Inflation Expectations from Business Tendency -Survey Data in Uganda

empirics have tended to favor the rational expectations hypothesis.

2.2. Estimating inflation expectations

The literature is inundated with several methods for estimation of inflation expectations. However, the various methods can be categorized into three main approaches on the basis of source data comprised of household and firm level surveys, financial markets data and economic indicators (Blanchflower and MacCoille, 2009). Where financial markets data is used, a forward inflation curve is inferred from estimations of nominal and real forward interest curves. The main argument against this approach is that the derived inflation expectations have other components that should not be part of the inflation expectations such as the inflation risk premium, liquidity risk, and other market factors. It has also been argued that this approach requires markets to be very sophisticated to get reliable estimates (Perera, 2008). Economic indicators which can be used to estimate inflation expectations include wage settlement data. However, the use of wage settlement data could also be misleading as it may reflect other factors as well such as labor productivity and the ability of firms to pay (Blanchflower and MacCoille, 2009).

The use of survey data focuses on businesses, the general public, and professional economic forecasters and academics whose opinions are sought on their price expectations over a specified time horizon. In some surveys, questions may simply seek to establish the opinion on the direction of future price movements whereas in others, magnitudes of movements may also be sought. However, survey data tends to exhibit high volatility (Blanchflower and MacCoille, 2009). In addition, survey data also has large dispersions due to the fact that not everyone has the same expectations although this tends to be ignored in their use (Mankiw, Ricardo, and Wolfers, 2004). The use of survey data however is not uniform and different methods have been employed to convert survey data into estimates of inflation expectations.

However, the most common method and one of the oldest in the literature refers to the work of Carlson and Parkin (1975). Carlson and Parkin (1975) estimated quantitative measures of inflation expectations for the United Kingdom from qualitative survey data. In their method which builds on the work of Theil (1958), they assumed that respondents had a common subjective probability distribution over future developments of a variable. The variable would rise or fall if the median of the subjective probability was above or below some indifference interval. The up-

Journal africain de statistiques, numéro 21, septembre 2019 17 Kenneth Alpha Egesa

per and lower boundaries of the indifference intervals were obtained from the respondent’s aggregate responses, and the respective past realizations of the variable under investigation. It was also assumed that the qualitative responses were normally distributed and that expectations were unbiased. Based on the methodology, estimates of expected inflation of the United Kingdom for the period 1961 to 1973 were obtained. The study noted that while the series tracked corresponded well with actual inflation, it was more volatile.

Batchelor and Orr (1988) replicated the method used by Carlson and Parkin (1975) for the UK and extended the estimates to 1985. They also derived additional estimates based on an assumed logistic distribution of the subjective probability of expectations as opposed to the normal distribution assumed by Carlson and Parkin (1975). The empirical evidence from the data supported a logistic distribution. In addition, while Carlson and Parkin (1975) assumed that the just noticeable differences in inflation expectations relative to the subjective probability distribution function were equal across individuals and constant over time, Batchelor and Orr (1988) assumed that they varied. The estimated inflation expectations series obtained were compared with the estimates derived using the Carlson and Parkin methodology for the same period. The Batchelor and Orr estimates for inflation expectations were superior to those obtained, using the Carlson and Parkin methodology in the two respects of lower volatility and better accuracy relative to actual inflation. Other studies have also criticized the Carlson and Parkin methodology for its restrictive assumptions on the subjective probability distribution, and the indifference intervals.

Pesaran (1984) developed a regression methodology as an alternative approach that avoided some of the criticisms of the Carlson and Parkin method on its restrictive assumptions. The method entailed quantification of qualitative responses using a regression model. The model derives estimates of the just noticeable difference based on the relationship between actual inflation and respondents’ perceptions of the past. The estimates of the just noticeable differences in the model are imposed on the qualitative expectations data to estimate inflation expectations. The estimates obtained using this approach are subsequently used as a variable in the specific regression model as opposed to a subjective probability distribution function as in the case of the Carlson and Parkin and the Batchelor and Orr methods (Henzel and Wollmershäuser, 2005). The method also allows for asymmetric indifference intervals, although a key criticism is that the boundaries of the indifference interval are

18 The African Statistical Journal, Volume 21, September 2019 1.Quantification of Inflation Expectations from Business Tendency -Survey Data in Uganda

fixed over time. An attempt was made by Sietz (1988) to address this shortcoming by estimating a regression model that explicitly allows for asymmetric and time varying boundaries. Henzel and Wollmershauser (2005) derived estimates based on the proposed method by Sietz (1988) and found some evidence confirming that the boundaries of the indifference interval varied over time.

In summary, there are different methods that can be applied in the quantification of qualitative responses on inflation expectations. However, the debate on which method produces the best estimates is yet to be concluded. Each method has some criticisms which for the most part, focus on the assumptions made about responses and the distribution function of the subjective probability, and the nature of the boundaries of the indifference interval time (whether they are asymmetric or not, and whether they are constant over time or not). This article focuses on the application of the different methods to Ugandan data, to identify the method that provides the closest estimates to actual inflation.

3. DATA DESCRIPTION AND METHODOLOGY

3.1. Data description

There are two surveys that provide qualitative responses on inflation expectations conducted by the Bank of Uganda: the business tendency and consumer confidence surveys. The business tendency survey is used to provide information used for compiling the business tendency index. The index is essentially a summary of opinions of business leaders regarding the state of the economy and the developments likely to take place in the short term. The survey is conducted every month and targets a panel of 300 businesses. The survey uses the stratified random sampling design covering firms from a list of top 1000 taxpayers (ranked based on Gross Revenues) as sampling units. The top 1000 enterprises are subdivided into 4 strata classified by sector sub-groups, namely; manufacturing, construction, wholesale and retail trade, and other sectors. Firms are classified based on the kind of activity, using International Standard Industrial Classification (ISIC). The sector subgroups are later subdivided into 2 sub-strata classified by size consisting of large and small, and medium enterprises. A sample of large firms is selected using purposive sampling with the intention of making them representative in terms of data collected and the reporting units while the small and medium enterprises are selected using systematic sampling.

Journal africain de statistiques, numéro 21, septembre 2019 19 Kenneth Alpha Egesa

The data is usually collected during the second and third weeks of each month on businesses assessments of the economic environment during the previous and current month and expectations in the next three months. Responses to questions can be negative in the case of an expected deterioration or positive in case of an expected improvement or neutral in case no change is expected in the variable under investigation. In particular, responses to the questions on expectations about future prices include an improvement in prices if expected to increase (positive), deterioration if expected to fall (negative) or no change if expected to stay or remain the same (neutral).

The consumer confidence survey is used to compile the consumer confidence -in dex which helps to understand the consumer’s opinion about the past, present, and future course of the economy. The survey provides useful information about consumer perceptions of the business climate, personal finance, prices, and spending. Telephone interviews are conducted on a sample of 500 employed individuals ran- domly selected each month from a telephone directory. Like the business tendency survey, the data is collected during the first two weeks of the month. However, for consumers, the responses are interpreted as positive if prices are expected to fall, negative if they are expected to increase or neutral if they are expected to stay unchanged.

The price expectations index is computed as a diffusion index defined as the percentage of respondents that answered that they expected a decrease (positive response) plus the percentage of respondents that expected the prices to stay the same with respect to their views on a given indicator. For example, if there were N respondents and P was the fraction of respondents with positive responses while E was the fraction of respondents who expect no change, then the Diffusion Index (DI) would be derived as:

DI=100*(P+E⁄2) (4)

The DI ranges from 0 to 100 with 50 as the mid-point. An index level of 50 indicates that there is an exact balance between those who responded that the prices were expected to increase with those who replied that the prices were expected to reduce. Thus, any level below 50 for businesses implies that prices are expected to fall and above 50 implies prices are expected to increase. For consumers, an index above 50 implies prices are expected to fall and below 50 signals an expected increase.

20 The African Statistical Journal, Volume 21, September 2019 1.Quantification of Inflation Expectations from Business Tendency -Survey Data in Uganda

In this study, the data from the business tendency survey is used for the analysis because the same respondents are surveyed each month. While the consumer confidence survey would have been a better choice for this type of analysis, given that the intention is to establish the expectations of the consumer, the fact that the survey selects different respondents for each survey contradicts some of the general assumptions upon which the quantification methodologies are based. For instance, it is assumed that respondents have some threshold values that are used in the formation of expectations on whether prices will rise, fall or stay the same. Results based on different respondents each round would introduce a variation in the thresholds during each survey round, reflecting location and income category, among others. Nonetheless, the two series are not fundamentally different as shown by the trend analysis in Figure 1.

Figure 1: Diffusion Index (DI) for consumer and business price expectations

Source: Bank of Uganda

The dataset used is comprised of 43 observations of inflation expectations of businesses over a three-month horizon. As seen from Figure 1, businesses expect prices to increase most of the time which results in an index that is above 50 highlighting their optimism with respect to prices. Similarly, consumers expect prices to increase throughout the period which explains why the consumer index is below 50. The trends in the indices are also symmetric as seen from the gap between the two indices. When businesses are overly optimistic about selling prices rising (i.e. when the index is much larger than 50), consumers are also overly pessimistic about

Journal africain de statistiques, numéro 21, septembre 2019 21 Kenneth Alpha Egesa

purchase prices falling (i.e. the index is much lower than 50). The only exception is during the period between the months of June and October 2013.

3.2. Methodology

The estimates of inflation expectations for both businesses and consumers are estimated using the methods used by Carlson and Parkin (1975) and Pesaran (1984), respectively. The Carlson and Parkin (1975) probability approach estimates the expected value of the change in inflation based on the percentage of respondents expecting an increase in inflation UP( t) and a reduction in inflation (DOt) as:

(5)

Where at and bt are the upper and lower boundaries of the expected change in inflation based on individual responses from the survey and rt and ft are the quantiles for time t calculated as:

(6)

(7)

Where is the cumulative distribution function of the standard normal and is the standard deviation of the aggregate distribution of the inflation expectations

Carlson and Parkin (1975) assumed that the upper and lower boundaries were time invariant and symmetric i.e. c = -at= bt and derived c as:

(8)

Substituting for at and bt in Equation (5), the expected change in inflation can be estimated as

(9)

22 The African Statistical Journal, Volume 21, September 2019 1.Quantification of Inflation Expectations from Business Tendency -Survey Data in Uganda

The second methodology applied in the study follows the balance statistic used for summarizing qualitative survey data. The balance statistics summarizes the positive and negative responses. Pesaran (1984) developed a regression approach which makes use of the linear relationship of the actual observed time series and the respondents perceptions as summarized by the balance statistics to establish the relationship for future outcomes (Rosenblatt-Wisch and Scheufele, 2015). In this study, the regression approach is set up as follows:

(10)

However, Et ∆πt+k in Equation (10) is unknown and cannot be estimated. The so- lution to this problem is to replace it with future realizations of expected inflation (Henzel and Wollmershäuser, 2005).

4. RESULTS

The estimates from the subjective probability method show that the upper and lower bound thresholds are 1.12 and -1.12, respectively. This means that the noticeable difference in actual inflation that triggers a response of either an expected increase or decrease in future inflation is 1.12 percent. However, when the regression method is used, and the assumption of asymmetric thresholds held, the estimates indicate 2.12 and -1.53 as the upper and lower boundaries, respectively (see regression estimates in Appendix 1). However, the lower boundary is not statistically significant at conventional levels while the upper boundary is statistically significant at one percent level of significance.

The trends in the numerical estimates of inflation expectations (quantitative estimates) were compared against the diffusion index for price expectations (qualitative estimates). As shown in Figure 2, the two series depict similar turning points; although the dips and rises depicted by the qualitative estimates are less dramatic (see details in Appendix 2). This is not surprising since the qualitative index is mainly intended for providing an indication of the direction of future price developments. The similar trends observed between the qualitative and quantitative measures derived from the qualitative responses highlights the consistence of the quantitative estimates with the survey responses.

Journal africain de statistiques, numéro 21, septembre 2019 23 Kenneth Alpha Egesa

Figure 2: Diffusion Index (DI) of inflation expectations and inflation expectations estimates

Source: author’s estimates

The estimates of inflation expectations derived from the two methods were also compared against the realized inflation for the respective periods. As shown in Fi- gure 3, the two methods used for quantification of qualitative data on inflation expectations provide very similar results. The general direction is similar, in the sense that estimated inflation expectations and realized inflation fall and rise at the same time. However, the estimates based on the subjective probability method are slightly lower compared to estimates from the regression method, save for the latter part of the period.

Furthermore, the visual inspection of the series shows that actual inflation has higher month-on-month changes relative to the estimates for inflation expectations (see Figure 3). For instance, over the period, realized inflation ranges between -0.5 and 1.5 percent, respectively, while the estimates are range bound between 0 and 0.8, respectively. Similarly, there is no single period during the study period when estimated inflation expectations highlight negative inflation, despite actual inflation realizations being negative in some periods. This indicates some upward bias on inflation expectations of businesses.

24 The African Statistical Journal, Volume 21, September 2019 1.Quantification of Inflation Expectations from Business Tendency -Survey Data in Uganda

Figure 3: Expected and actual inflation

Source: author’s estimates

A comparison was also drawn on the closeness of each of the estimated inflation expectations with realized inflation for the respective period. The inflation expectations based on the subjective probability method have the same root mean square error of 0.49 when compared to the estimates from the regression method relative to realized inflation. However, the estimates from the regression method have an overall higher bias. The average bias amounts to 1.38 percent for the estimates from the regression method compared to 0.04 percent for the estimates derived from the subjective probability method. Other parameters such as correlation coefficients and student t-tests also confirm that the estimates for inflation expectations from the two methods are not significantly different. For instance, the correlation coefficient between the inflation expectations and realized inflation is 0.37 percent for both series.

Journal africain de statistiques, numéro 21, septembre 2019 25 Kenneth Alpha Egesa

5. CONCLUSION

The study set out to apply simple techniques for quantifying qualitative survey data on inflation expectations. The analysis focused on responses of firms on their expected selling prices in the period ahead. Overall, the qualitative survey data was quantified using the subjective probability method of Carlson and Parkin (1975) and the regression technique based on Pesaran (1984)’s balance statistic. Both methods provide decent estimates of inflation expectations relative to realized inflation. The estimates from the two methods used did not differ significantly so it was not possible to draw any conclusion on the superiority of one method over the other. Given the absence of such data on inflation expectations despite their importance in generating forecasts considered during monetary policy formulation by the monetary policy committee, it is recommended that going forward, both estimates may be used as part of the inputs in the forecasting process. An important consideration for future work would be to compare how the estimates perform relative to macroeconomic models.

26 The African Statistical Journal, Volume 21, September 2019 1.Quantification of Inflation Expectations from Business Tendency -Survey Data in Uganda

REFERENCES

Batchelor, R. A., and Orr, A. B. (1988). “Inflation Expectations Revisited.” Economica, New Series, Vol. 55, No. 219 (Aug., 1988): 317-331. https://www.jstor.org/ stable/2554010

Blanchflower, D. G., and MacCoille, C. (2009). “The formation of inflation- expecta tions:An empirical analysis for the UK. National Bureau of Economic Reseach Working Paper Series 15388. http://www.nber.org/papers/w15388

Cagan, P. (1956). “The monetary dynamics of hyperinflation.” In Studies in the Quantity Theory of Money. Friedman, Milton (ed.), The University of Chicago Press, Chicago, pp. 25-117.

Carlson, J. A., and Parkin, M. (1975). “Inflation expectations.” Economica, New Series, Vol. 42,No. 166 (May, 1975): 123-138. https://www.jstor.org/stable/2553588

Chow, G. C. (2011). “Usefulness of adaptive and rational expectations in economics.” Cen- ter for Economic Policy Studies, Princeton University Working Paper No. 221 September 2011.

Debabrata, P. M., and Ray, P. (2010). “Inflation Expectations and Monetary Policy in India: An Empirical Exploration.” IMF Working Papers WP/10/84.

Feige, E. L., and Pearce, D. K. (1976). “Economically Rational Expectations: Are Inno- vations in the Rate of Inflation Independent of Innovations in Measures of Monetary and Fiscal Policy?” The Journal of Political Economy, Vol. 84, No. 3 (Jun., 1976): 499-522. https://www.jstor.org/stable/1829866

Gertchev, Nikolay. (2007). A critique of adaptive and rational expectations. Quarterly Journal of Austrian Economics, 10:313-329. DOI 10.1007/S12113-007-9023-1.

Henzel, S., and Wollmershäuser, T. (2005). “Quantifying Inflation Expectations with the Carlson-Parkin Method - A Survey-based Determination of the Just Noticeable Diffe- rence.” Journal of Business Cycle Measurement and Analysis: Volume 2, No. 3, 2005. https://dx.doi.org/10.1787/jbcma-v2005-art8-en

Kamada, K., Nakajima, J., and Nishiguchi, S. (2015). “Are Household Inflation Expecta- tionsAnchored in Japan?” Bank of Japan Working Paper Series. No.15-E-8 July 2015.

Journal africain de statistiques, numéro 21, septembre 2019 27 Kenneth Alpha Egesa

Mankiw, G. N., Ricardo, R., and Wolfers, J. (2004). “Disagreement about Inflation Ex- pectations.” National Bureau of Economic Research Working Paper Series 9796. http://www.nber.org/papers/w9796

Mlambo, L. (2012). “Adaptive and rational expectations hypotheses: Reviewing the critiques. The International Journal of Economic Behavior- IJEB, 2012, vol. 2, issue 1: 3-15.

Muth, J. F. (1961). “Rational expectations and the theory of price movements.” Econometrica: Journal of the Econometric Society, Vol. 29, No. 3 (Jul., 1961): 315- 335. http://www.jstor.org/stable/1909635

Pesaran, M. H. (1984). “Expectations Formations and Macroeconomic Modelling.” In Contemporary Macroeconomic Modelling. P. Malgrange, and P. Muet (eds.). Oxford: Basil Blackwell. https://doi.org/10.1016/0264-9993(85)90012-4

Perera, A. (2008). “Inflation Expectations and Monetary Policy.”Sri Lanka Econo- mic Association Annual Research Journal 2008. https://ssrn.com/abstract=1586663

Seitz, H. (1988). “The Estimation of Inflation Forecasts from Business Survey Data.” Applied Economics, 20:4, pp. 427-438. DOI: 10.1080/00036848800000055

Rosenblatt-Wisch, R., and Scheufele, R. (2015). “Quantification and characteristics of household inflation expectations in Switzerland.” Applied Economics 2015,vol. 47, issue 26: 2699-2716.

Theil, H. (1958). Economic Forecasts and Policy. Ch. 4, Amsterdam: North-Hol- land.

28 The African Statistical Journal, Volume 21, September 2019 1.Quantification of Inflation Expectations from Business Tendency -Survey Data in Uganda

Appendix 1: Regression estimates

Journal africain de statistiques, numéro 21, septembre 2019 29 Kenneth Alpha Egesa

Appendix 2: Realized inflation, Diffusion Index (DI), inflation expectations, and bias

30 The African Statistical Journal, Volume 21, September 2019 2. Cigarette Smoking in Tanzania-Prevalence and Determinants

Asmerom Kidane 1, Aloyce Hepelwa 2 , Kenneth Mdadila 3 , Salvatory Macha 4 , Anita Lee 5 , Teh Wei Hu 6

Abstract

The study presented here attempts to estimate cigarette smoking prevalence in Tan- zania by taking into account regional, demographic and other socio-economic variations. In this exercise, smoking prevalence is the percent of current daily smokers over the total respondents. The study highlights the fact that smoking prevalence in advanced northern countries is showing a decline while the opposite appears to be the case in African countries. The study also verifies the notion that cigarette smokers in Tanzania and other African countries as being relatively poor; they are unable to afford to be treated for smoking induced illnesses. A sample of 520 adults aged 25 to 90 years were selected for the study which was conducted in 2016. The study areas included two urban areas (Dar es Salaam and Singida), and two rural areas. The overall smoking prevalence rate was estimated to be 11.59 percent. After highlighting some descriptive statistics of the respondents, two interrelated measures of association were estimated, namely Contingency Tables with Chi Squared Statistic, followed by logistic regression on the determinants of smoking prevalence. The results showed that males are 20.76 times more likely to smoke when compared to females, that older people are 2.33 times more likely to smoke when compared to younger respondents, that urban dwellers are 2.50 times more likely to smoke when compared to rural dwellers. Also, literate respondents are 76.35 percent less likely to smoke when compared to non-literate respondents, that government and other full-time employees are 62.57 percent less likely to smoke when compared to traders or farmers. One can safely conclude that smokers in Tanzania are likely to be the poor- the elderly, the less educated, and the farmers. Smokers in Tanzania are also predominantly males.

1 Department of Economics, University of Dar es Salaam, Tanzania; [email protected]. 2 Department of Economics, University of Dar es Salaam, Tanzania; [email protected]. 3 Department of Economics, University of Dar es Salaam, Tanzania; [email protected]. 4 Environment for Development, Department of Economics, University of Dar es Salaam, Tanzania; [email protected]. 5 Public Health Institute Oakland California, USA [email protected]. 6 School of Public Health University of California, Berkeley; [email protected].

Journal africain de statistiques, numéro 21, septembre 2019 31 Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu

KEYWORDS : Smoking, Prevalence, Tanzania, Contingency Tables, Odds Ratio.

Résumé L’étude présentée ici tente d’estimer la prévalence du tabagisme en Tanzanie en tenant compte des variations régionales, démographiques et autres variations socio-économiques. Dans cet exercice, la prévalence du tabagisme est le pourcentage de fumeurs quotidiens actuels par rapport au nombre total de répondants. L’étude met en évidence le fait que la prévalence du tabagisme dans les pays avancés du Nord est en baisse, alors que l’inverse semble être le cas dans les pays africains. L’étude vérifie également la notion selon laquelle les fumeurs de cigarettes en Tan- zanie et dans d’autres pays africains sont relativement pauvres; ils n’ont pas les moyens de se faire soigner pour des maladies liées au tabagisme. Un échantillon de 520 adultes âgés de 25 à 90 ans a été sélectionné pour l’étude menée en 2016. Les zones de l’étude comprenaient deux zones urbaines (Dar es-Salaam et Singida) et deux zones rurales. Le taux de prévalence globale du tabagisme a été estimé à 11,59%. Après avoir mis en évidence certaines statistiques descriptives des répon- dants, deux mesures d’association interdépendantes ont été estimées, à savoir les tableaux de contingence avec la statistique du chi carré, suivies d’une régression logistique des déterminants de la prévalence du tabagisme. Les résultats ont mon- tré que les hommes sont 20,76 fois plus susceptibles de fumer que les femmes, que les personnes âgées sont 2,33 fois plus susceptibles de fumer que les plus jeunes, que les citadins sont 2,50 fois plus susceptibles de fumer que les ruraux. En outre, les répondants alphabétisés ont 76,35% moins de chances de fumer que les non alphabétisés, tandis que le gouvernement et les autres employés à plein temps ont 62,57% moins de risque de fumer par rapport aux commerçants ou aux agriculteurs. On peut sans risque en conclure que les fumeurs en Tanzanie sont probable- ment les pauvres - les personnes âgées, les moins instruits et les agriculteurs. Les fumeurs en Tanzanie sont également principalement des hommes.

Mots-clés: tabagisme, prévalence, Tanzanie, tables de contingence, rapport de cotes.

32 The African Statistical Journal, Volume 21, September 2019 2. Cigarette Smoking in Tanzania-Prevalence and Determinants

1. INTRODUCTION

Tobacco production and cigarette consumption are major health issues in countries throughout the world and Tanzania is no exception. Smoking is the most preven- table cause of death. (WHO 2013B). It is estimated that more than 5 million people may have died due to tobacco-related illnesses in 2012; unless some control measures are introduced, this estimate would be about 8 million by 2030. According to the WHO statistics, the tobacco prevalence rate among Tanzanian adults (25 to 64 years) is estimated at 23 percent. Among youths aged 13 to 15, the prevalence rate is 3.8 percent for males, and 0.4 percent for females. For the Eastern Africa region, the prevalence rate is 29 percent for males and 4 percent for females. Smoking causes a huge economic burden to society; this effect is more pronounced in a developing African country like Tanzania. (WHO URT 2015).

In an attempt to control the diseases and deaths emanating from cigarette smoking, the government of Tanzania introduced effective tobacco control measures. In 2007, the country signed a pact to control tobacco production, and cigarette consumption. However, to date, the government initiative has not been effective in controlling cigarette consumption.

One of the effective ways of monitoring and controlling smoking prevalence is to conduct a periodic national survey on the levels and determinants of cigarette smoking. The aim of this article is thus to estimate the smoking prevalence rate in Tanzania and relate the estimate with various socio-economic characteristics of smokers. This article has been divided into five parts; part two presents literature review on some basic issues related to smoking prevalence, and its negative health consequences. Part three highlights the general and specific objectives of the study, data sources, and methods of analysis. Part four presents the empirical findings, and discusses the results. Part five presents the conclusion, and highlights some policy implications.

2. LITERATURE REVIEW 2.1. Prevalence and Consequences of Cigarette Smoking The negative health consequence of tobacco production in general and cigarette smoking in particular has been well documented. Tobacco is estimated to kill up to one of every two users. (WHO. 2013b) This high mortality rate is preceded by high health expenditure on tobacco induced diseases. Smoking induced diseases

Journal africain de statistiques, numéro 21, septembre 2019 33 Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu

include tuberculosis, lower respiratory infections, and NCD, including cardiovascular diseases, chronic obstructive pulmonary disease, and several types of cancer (Rigotti, 2013). These diseases are believed to cost about half a trillion dollars in economic damages per year (WHO, 2013b). Cognizant of this problem, many developed countries of Europe and North America have campaigned vigorously and invested substantially to reduce tobacco production and cigarette consumption. As a result, the prevalence of cigarette consumption has declined in these high-income countries. In response to this, the tobacco industry has turned to low and middle-income countries, particularly in Africa, Asia, Latin America, and Eastern Europe in search of new customers.

Low-income countries in general and African countries in particular have become major markets for cigarette companies. The epidemic appears to be shifting to the developing world as more than 80 percent of the world smokers live in low and middle income countries (Elsheikh, E., 2009). There is thus the need for an aggres- sive anti-smoking campaign to control, reduce, and eventually stop the prevalence of smoking in Africa. This is a rather arduous task as the public health benefits from reduced cigarette consumption takes many years to manifest themselves. There is also a prolonged time lag between smoking initiation and the onset of tobacco-related diseases. This prolonged time lag has resulted in an increase in smoking prevalence rate (Blecher et al 2013, Baleta 2010).

2.2. Smoking and Poverty

There is a substantial empirical evidence which verifies the notion that smoking is more prevalent among the poorest segments of the population. This is true in both developed and developing countries (Efroymson 2011, Kidane et al 2015A). The poor are already under financial stress. They are always sick, and are unable to afford to visit healthcare services for treatment. Smokers appear to develop many more illnesses than non-smokers resulting in enormous costs, thereby putting financial burdens on any country’s healthcare expenditures. Poor smokers will find it quite difficult to enroll in a health coverage insurance program. In places where individuals purchase health insurance, the costs are proportionately much higher than they are for non-smokers. Smoking-related illnesses take workers out of the workforce; this in turn leads to loss in income, thereby creating a downward pressure on the economy. It is not uncommon for one to observe that smoking is more prevalent among populations with mental-health problems, alcohol, and substance

34 The African Statistical Journal, Volume 21, September 2019 2. Cigarette Smoking in Tanzania-Prevalence and Determinants

abuse. Smoking prevalence is also much higher among criminals, and the homeless.

Studies on the characteristics of smokers within the African setting show that smoking is directly associated with poverty, and that an expense on cigarette is at the expense of basic needs such as food, and other basic necessities. For Tanzania, yearly total expenditure (a proxy for income) among non-smokers is higher than that of smokers (Kidane et al 2015-B). A study on Tanzania compared the percentage of smokers and non-smokers who are below the food poverty line (Kidane et al 2015-A). Among non-smokers, the percentage below the food poverty line is 18.37 percent; the corresponding value for smokers is 23.93 percent.

2.3. Trends of Tobacco Use in Africa

In 2011, WHO estimated that adult tobacco smoking prevalence (men and women) in sub-Saharan Africa ranged from 5 percent in Niger to 34 percent in Sierra Leone (WHO, 2013c).

In nearly all countries, there is a significant gap between rates of usage among males and females; the estimated prevalence among females is less than half of that among males. Prevalence estimates for tobacco smoking in Africa range from 8 to 48 percent in adult men and 0.4 to 20 percent in adult women. Even though women may currently make up a smaller percentage of smokers, as male smoking peaks and declines, female prevalence is expected to continue to rise (Atlas, 2015).

Few countries in Africa keep comprehensive data on trends in tobacco use and the subsequent effects on morbidity and mortality, which means that most of the above prevalence measures are only estimates. These countries lack standardized and comparable data disaggregated by gender, age, and risk factors. Compared to other countries, tobacco and cigarette statistics in sub-Saharan tobacco consumption are incomplete and less comparable over time, and space. Estimates are based on small, irregular, non-periodic, non-representative, and thus, less generalizable survey samples.

2.4. Tobacco Production and Cigarette Smoking in Tanzania

Tobacco is one of the cash crops that helps generate foreign exchange earning in Tanzania. The country ranks as a third within Africa (after Malawi and Zimbabwe)

Journal africain de statistiques, numéro 21, septembre 2019 35 Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu

as a major producer and exporter of tobacco leaves. Tobacco is also consumed by Tanzanians with a prevalence rate of 10.8 percent (World Bank, 2013). As is the case in many African countries, Tanzania has a high concentration of the poor living in over-crowded urban areas commonly referred to as shanty towns. Cigarette smoking prevalence in these areas are likely to be high; the cigarettes are likely to be homemade and of “low unfiltered” quality.

Tobacco control in Tanzania dates back to 2003 when the government enacted the Tobacco Products (Regulation) Act (TTCPF, 2003). It is the only legislation on tobacco control in the country. The 2003 Act aims at regulating public smoking, tobacco advertising, promotion, sponsorship and tobacco packaging as well as la- beling (URT, 2003). In 2007, the country ratified the WHO Framework Convention Control (FCTC). Among other steps, WHO FCTC calls on governments which ratified the treaty to (i) adopt tax and price measures to reduce tobacco consumption; (ii) ban tobacco advertising, promotion and sponsorship; (iii) create smoke-free work and public spaces; (iv) put prominent health warning on tobacco packages; and (v) combat illicit trade in tobacco products.

Despite the above mentioned control measures, cigarette smoking prevalence appears to be on the rise.

3. OBJECTIVES OF THE STUDY AND SOURCES OF DATA

3.1. Objectives

The overall objective of this study is to estimate the prevalence of smoking in Tan- zania by taking into account, gender, regional, socio-economic, and related variations. The aim is to quantify and verify the magnitude of smoking prevalence and compare it with past estimates from Tanzania, and other neighboring African countries. The specific objectives are:

• To estimate the prevalence of smoking in Tanzania. • To identify socio-economic and demographic variables that affect the prevalence of cigarette smoking. • To highlight the policy implications of the findings.

36 The African Statistical Journal, Volume 21, September 2019 2. Cigarette Smoking in Tanzania-Prevalence and Determinants

3.2. Source of Data and Method of Analysis

The data for this study is based on a 2016 survey conducted among the Tanzanian population. The sampling method may be classified as Stratified Purposive Sam- pling. A sample of 520 adult respondents were selected; respondents were selected with fair representation by gender, residence, and type of employment. This representation is shown in Table 1

Table 1 Distribution of Respondents by Socio-Economic Status

*Formal and informal traders, **teachers, government, and non-government employees.

The questionnaire is structured on the basis of WHO recommended procedures commonly referred to as GATS (2011). The three basic questions asked were;

i. Current tobacco smoking status (current daily tobacco smokers); ii. Past daily smoking status-for current less than daily smokers (among all adults); and iii. Past smoking status-for current non smokers (among daily smokers). Respondents were thus asked whether they currently smoke on a daily basis, less than daily basis or not at all. The response was 59,9 percent, and 451 respectively. Our prevalence estimates will thus be confined to the current daily tobacco smokers. The method of analysis includes simple and cross classified descriptive statistics, contingency tables with chi squared values, followed by a logistic regression where the estimated coefficients are expressed in terms of odds- ratio.

Journal africain de statistiques, numéro 21, septembre 2019 37 Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu

4. EMPIRICAL FINDINGS 4.1. Some Descriptive Statistics

Before presenting the results, some basic socio-economic, and demographic characteristics of the respondents are highlighted. The results are given in Table 2:

*In Tanzanian shillings (1 USD=2200 TZS)

The table shows that the mean age of male respondents as being 38.17 years (2.84 years older than female respondents). Only 2.88 percent of the respondents are illiterate; the household size is 4.21 which is slightly lower than the national average. The mean number of rooms in a respondent’s house is 2.96. Those who receive clean (running) water and those who receive modern source of energy (electricity) constitute 63.65 percent and 58.27 percent respectively. These results appear to be higher than the national average. The inclusion of relatively large sample from the capital- Dar es Salaam-appears to have inflated the results. As expected, permanent workers who are more educated (teachers, government employees) earn much higher income. Traders (most of whom are small-scale and informal) are ranked second while farmers who are rural dwellers belong to the lowest income group.

38 The African Statistical Journal, Volume 21, September 2019 2. Cigarette Smoking in Tanzania-Prevalence and Determinants

4.2. Smoking Prevalence

Respondents were asked a question on current smoking status. The results show the daily smoking prevalence rate as being 11.57 percent. When compared with other African countries, the Tanzanian estimate appears to be on the lower side. Other estimates range from 23.0 for adult males (25-64); the corresponding value for females is 1.3 (WHO URT Africa Region). A 2015 study of DHS in African countries shows a prevalence rate that ranges from less than 10 percent for Ethiopia to a high of 37% for Sierra Leone (Sreeramareddy et al 2014). These wide variations may be the result of different definitions of the numerator and denominator.

4.3. The Socio-Economic Determinants of Smoking Prevalence

In this study, we apply two complementary statistical methods to identify the various socio-economic determinants of smoking prevalence. The first method is a series of contingency tables where each socio-economic variable is cross classified with smoking prevalence; a chi squared statistic is estimated for each contingency table. The second method is a logistic regression approach where smoking prevalence is regressed on the relevant explanatory variables. The estimated logistic regression coefficients will be expressed in terms of odds-ratio.

4.3.1. Contingency Tables Table 3A. Association between smoking and selected socio-economic variables Variables

Journal africain de statistiques, numéro 21, septembre 2019 39 Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu

*significant at 5%, **significant at 10 percent.

The result in Table 3A shows that almost all smokers are males with a prevalence rate of 16.14 percent; among females, the prevalence is only 0.79 percent When classified by age, younger respondents show a much lower rate (7.55 percent among those aged less than 30 years old). The rate for respondents that are older than 50 years is 19.70 percent. Furthermore, smoking prevalence rate among rural dwellers is higher (14.93 percent) than their urban counterparts which is 9.39 percent.

40 The African Statistical Journal, Volume 21, September 2019 2. Cigarette Smoking in Tanzania-Prevalence and Determinants

With regards to variation by religion, the prevalence rate among Moslems is 16.53 percent, whilst the value for Christians is only 6.99 percent It should be noted that most Tanzanian Moslems reside in the coastal areas while Christian Tanzanians reside in the interior part of country. There appears to be little association between smoking and marital status.

Table 3B. Association between smoking and selected socio-economic variables Variables

*significant at 5%.

Table 3B is a continuation of Table 3A and shows a two way association between smoking prevalence on the one hand and education, income and occupation on the other hand. While smoking prevalence among literate respondents is 12.1 percent, the corresponding value of those who are illiterate is a staggering 40 percent. In- come is classified into four categories by quartiles. The prevalence rate among the lowest income group is 19.61 percent. The corresponding value among the highest

Journal africain de statistiques, numéro 21, septembre 2019 41 Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu

income group (fourth quartile) is 3.86 percent, suggesting that smoking is more prevalent among the poor. Income and occupation are closely associated. Regular employees who have high monthly income have a prevalence rate of 3.31 percent. On the other hand, farmers who have the lowest income have a prevalence rate of 20.83 percent.

4.3.2. Logistic Regression Estimate

In this section, the determinants of cigarette prevalence are estimated on the eight explanatory variables identified in Tables 3A and 3B. The results are given in Table 4:

Table 4 Determinants of Smoking Prevalence: Logistic Regression Estimate (Dependent variable:0=non-smoker, 1=smoker)

*significant at 5%, **mean age and mean income are the cut-off points between young and old, between low and high income earners.

42 The African Statistical Journal, Volume 21, September 2019 2. Cigarette Smoking in Tanzania-Prevalence and Determinants

The results show the estimated logistic regression as having high explanatory and predictive power (Wald chi squared of 68.55 with probability of 0.0000). Five out of eight explanatory variables are statistically significant. For these significant -ex planatory variables, the odds-ratio indicate the following:

i. Males are 20.76 times more likely to smoke when compared to females; ii. Older people are 2.33 times more likely to smoke when compared to younger respondents; iii. Urban dwellers are 2.50 times more likely to smoke when compared to rural dwellers; iv. Literate respondents are 76.35 percent less likely to smoke when compared to non-literate respondents; and v. Government and other permanent respondents are 62.57 percent less likely to smoke when compared to traders or farmers.

Based on the results above, one can safely conclude that smokers are likely to be the poor- the elderly, the less educated, and the farmers.

5. CONCLUSION

The preceding results show that the prevalence of cigarette smoking in Tanzania as being significant. It was noted that Tanzania is one of the major tobacco growing countries in the world. Tobacco leaf buyers grade the tobacco leaves into three categories, as high, medium, and low quality; the high and medium quality are exported while the low quality is used by rural farmers to produce and sell “unfiltered low quality cigarette” for local consumption.

This study also tried to identify smoking prevalence by gender where the prevalence among female is very low while that of males is being relatively high. This may lead to the conclusion that the campaign to control cigarette smoking ought to be a component of a campaign against gender discrimination. Spouses of male smokers may be at a risk of developing smoking related diseases.

This study also showed that individual smokers belong to occupation that yield low income, they appear to be the elderly, and those with little education. In other

Journal africain de statistiques, numéro 21, septembre 2019 43 Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu

words, many cigarette smokers appear to be the poor. Again, the campaign to reduce smoking prevalence ought to be part of a national poverty reduction strategy.

The health care costs of smoking induced morbidity and mortality have already been highlighted. For Tanzania, a study was made to estimate smoking induced cardiovascular disease (Kidane et al 2015C). The study concluded that Tanzanians spent a total (direct plus indirect cost) of 136.1 million dollars annually. This amount is very high for Tanzania whose per capita GDP is a mere 650.00 US dollars.

As part of WHO sponsored tobacco measures, Tanzania has taken several legis- lative and other steps to reduce the magnitude and impact of cigarette smoking. The latest estimate of smoking prevalence presented in this study (11.57 percent) indicates that, to date, the government’s attempt to control smoking leaves a lot to be desired. Treating smoking and its consequences as part of “gender equality” and “poverty reduction strategy” may help reduce the magnitude, and the negative consequences of cigarette smoking.

44 The African Statistical Journal, Volume 21, September 2019 2. Cigarette Smoking in Tanzania-Prevalence and Determinants

REFERENCES

Atlas, (2015). Tobacco and poverty www.tobaccoatlas.org/

Baleta, A. (2010). “Africa’s struggle to be smoke free.” The Lancet 375: 107-108.

Blecher, E. H., and H. Ross (2013). “Tobacco use in Africa: Tobacco control through prevention.” Atlanta, GA: American Cancer Society.

Efroymson, D., H. A. Pham, L. Jones, S. Fitzgerald, and L.T. Thu (2011). To- bacco and Poverty: Evidence from Vietnam. http://tobaccocontrol.bmj. com/ content/20/4/296.abstract

Elsheikh, E., (2009). “Public Health Data: Selected Comparisons.” Global Politics of Health Inequity, Autumn, 1-9.

GATS (2011). “Global Adult Tobacco Survey Collaborative Group.” Tobacco Questions for Surveys. A Subset of Key Questions from the Global Adult Tobacco Survey.” 2nd Edition. Atlanta, Ga. Center for Disease Control and Prevention.

Kidane, A. A. Hepelwa E. Ngeh and T. Hu (2015). “Healthcare Cost of Smoking Induced Cardiovascular Disease in Tanzania.” Journal of Health Science, 3 (2015- C): 126-131.

Kidane, A. A., J. Mduma and T. Hu (2015). “Impact of Smoking on Food Expendi- ture Among Tanzanian Households.”African Statistical Journal Vol. 18 (2015-B). (2015) “Impact of Smoking on Nutrition and the Food Poverty Level in Tanzania.” Journal of Poverty Alleviation and International Development, 6:1(2015-A).

Rigotti, N. A. (2013). “Smoking cessation in patients with respiratory disease: Existing treatments and future directions.” The Lancet Respiratory Medicine 1(3): 241-250.

Sreeramareddy, C.T, P. M. Pradhan and S. Sin (2014). “Prevalence, distribution, and social determinants of tobacco use in 30 sub-Saharan African countries.”BMC Medicine 201412:243 DOI: 10.1186/s12916-014-0243-x.

Journal africain de statistiques, numéro 21, septembre 2019 45 Asmerom Kidane, Aloyce Hepelwa, Kenneth Mdadila, Salvatory Macha, Anita Lee, Teh Wei Hu

Tanzania Tobacco Control Forum (TTCF) (2007). “Statement of the Tanzania To- bacco Control Forum to the Public Hearing on agricultural diversification and alternative crops to tobacco.” Brasilia, 26 February 2007.

WHO (2013). WHO report on the global tobacco epidemic, 2013: “Enforcing bans on tobacco advertising, promotion and sponsorship.” Geneva, Switzerland: WHO (2015b), URT WHO Report on the Global Tobacco Epidemic.

World Bank (2013). CIA world fact book, February 21, 2013. Retrieved from: https://www.cia.gov/library/publications/the-world-factbook/fields/2012html This study is supported by the United States National Institute of Health (NIH) Fogerty International Center and National Cancer Institute (Grant # R01TW009295).

46 The African Statistical Journal, Volume 21, September 2019 3. Statistical Literacy for National Development 3. Statistical Literacy for National Development

Korter Grace 1, Olatunji Lateef 2, Omolehin Joseph 3 , Olubusoye Olusanya 4

Abstract

Statistical literacy is a modern technique of teaching statistics with the goal of empowe- ring citizens with knowledge beyond traditional statistical principles and methods for critical consumption of information. This article advocates for a system-wide organizational culture to support the development of statistical literacy of citizens at all levels. The statistical literacy model proposed involves emphasis on use of data in addition to knowledge component, dispositional component, and statistical knowledge-base. Col- laboration between the Nigerian Statistical Association (NSA) and National Bureau of Statistics in design and trialing of curricula and resource materials for courses in statistical literacy at both school and university levels was seen as a pathway to enhancing statistical literacy. This article proposes the creation of a National Statistical Centre as important for providing a focus for the development of new resources. Hosting the Statistical Literacy Website and expanding NSA range of publications to address varied audience could be useful. Adoption of this modern statistical method will be dependent on the cooperation of stakeholders. The interest and enthusiasm will improve provided statistical activities in the National Statistical System are the responsibility of professional statisticians, and inform policy decisions for economic transformation targeted towards national development.

Key words: Statistical literacy, economic transformation, national statistical system Résumé

L’initiation aux statistiques est une technique moderne d’enseignement des statistiques visant à donner aux citoyens des connaissances allant au-delà des principes et mé- thodes statistiques traditionnels en matière de consommation critique d’informations. Cet article préconise une culture organisationnelle à l’échelle du système afin de soutenir le développement de la connaissance en statistique des citoyens à tous les niveaux.

1 Korter Grace, Department of Mathematics and Statistics, Federal Polytechnic, Offa, Nigeria; grace. [email protected] ALT [email protected] 2 Rector, Federal Polytechnic, Offa, Nigeria; [email protected] 3 Professor, Department of Mathematical Sciences, Federal University Lokoja, Nigeria; joseph.omo- [email protected] ALT [email protected] 4 Nigerian Statistical Association 2nd VP and Department of Statistics, University of Ibadan, Nige- ria; [email protected] ALT. [email protected]

Journal africain de statistiques, numéro 21, septembre 2019 47 Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

Le modèle de connaissance en statistique proposé met l’accent sur l’utilisation de don- nées en outre de la composante connaissances, de la composante dispositionnelle et de la base de connaissances statistiques. La collaboration entre l’Association statistique nigériane (NSA) et le Bureau national de la statistique pour la conception et la mise à l’essai de programmes d’études et de matériel didactique pour des cours d’initiation à la statistique à l’école et à l’université était considérée comme un moyen d’améliorer la connaissance en statistique. Cet article propose la création d’un centre national de statistique comme un élément important pour orienter le développement de nouvelles ressources. Il pourrait être utile d’héberger le site Web sur la connaissance en statistique et d’élargir la gamme des publications de la NSA pour s’adresser à un public varié. L’adoption de cette méthode statistique moderne dépendra de la coopération des parties prenantes. L’intérêt et l’enthousiasme vont améliorer pourvu que les activités statistiques dans le système statistique national qui relèvent de la responsabilité des statisticiens professionnels et éclairer les décisions politiques en matière de transformation économique orientée vers le développement national.

Mots-clés: connaissance en statistique, transformation économique, système statistique national

1. Introduction Statistics is the bane for planning, monitoring, evaluation, sound reasoning, and right decisions. Many aspects of societal or human progress depend on correct analysis of numerical figures. To allow for better governance and a guarantee for sus- tainability, individuals, corporate bodies, and governments need to think quantita- tively. A government and its citizens have proper understanding of the environment or situations in which they find themselves through the knowledge of statistics. The present increase in demand for data across the globe stems from the realization that a society cannot be governed on anecdotes. Statistical ignorance and statistical fallacies are often widespread and quite as dangerous as the logical fallacies that come under the heading of illiteracy (Cockcroft, 1982). As information becomes qualitative, an innumerate citizen today is as vulnerable as the illiterate peasant of Gutenberg’s time (see Steen, 1997). Citizens need to be statistically equipped to react intelligently and effectively to information. Processing information requires critical and sound thinking with a cognitive ability for enquiry capable of enhancement through statistical literacy.

48 The African Statistical Journal, Volume 21, September 2019 3. Statistical Literacy for National Development

Statisticians, educators, and stakeholders need to pay attention to emphasizing the potential power of statistics for enriching the society as required by the necessities of a rapidly changing world. The correlation between statistical activity and governmental functions requires recognition. The objective is to advocate for a system-wide organizational culture to support the development of statistical literacy of citizens at all levels.

This article is organized as follows; section 2 provides the literature review, section 3 amplifies the Implications of Statistical Literacy, whilst sections 4-10 focus on Nigerian National Statistical System; Statistics and Economic Transformation; Sta- tistical Literacy Model; Progress in Statistical Literacy; Challenges in Achieving Statistical Literacy; Pathways to Enhancing Statistical Literacy, and the conclusions respectively. 2. Literature Review

Statistical literacy is a key ability expected of citizens in information-laden societies, often touted as an expected outcome of schooling, and as a necessary component of adults’ numeracy and literacy. It involves understanding and using the basic knowledge and tools of statistics: knowing what basic statistical terms mean, understanding the use of simple statistical symbols, recognizing, and being able to interpret different representations of data (see Garfield, 1999; Snell, 1999; Rumsey, 2002a). Statistical literacy has a natural association with numeracy (Watson 2002). Statistical literacy is the ability to interpret, critically evaluate, and communicate about information and messages. It refers to the aspects necessary to establish an awareness of data that must take place in order to reasonably consume information (see Rumsey, 2002b). The process allows statistical principles and techniques to be applied in contexts associated with other areas of the curriculum and/or areas outside the school experience in the wider society. The concept involves teaching statistics better for a different or additional purpose, using real world and media-based examples.

Some schools and perhaps most post-secondary academic institutions teach statistics to some students as part of mathematics, statistics or science and social studies, yet not in a way that necessarily emphasizes the development of statistical literacy (see Wild, Triggs and Pffankuch, 1997; Hawkins, 1997; Moore and Cobb, 2000;

Journal africain de statistiques, numéro 21, septembre 2019 49 Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

Scheaffer 2001; Gal, 2002a; Wild, 2005). Current knowledge base about statistical literacy of school or university students and of adults in general is patchy (Gal, 2002b). The majority of the current adult population in any country has not had much if any formal exposure to the statistical or mathematical knowledge bases given in known education levels across the world (see Wallman, 1993; Statistics Canada and OECD, 1996; UNESCO, 2000; Ottaviani, 2002). Remarkably, until very recently, even economically advanced societies have prized far less the goal of developing a functionally numerate citizenry than one that is functionally literate. Even in these societies, it will be a huge task to redress existing deficiencies. In developed countries, formal teaching aimed expressly at enhancing statistical literacy is still a fledgling enterprise, offered only in scattered locations and to relatively few people. This is even truer in the case of adult learners than it is for young people (Sowey, 2003).

The need to develop statistical and probabilistic knowledge, and to empower people from all lifestyles to become critical consumers and users of statistical information has been embraced by educators, and policymakers in diverse countries as well as by many professional organizations. The focus is on the school curriculum to develop high-level statistical-questioning skills, the cross-curricular nature of data handling, representation, and interpretation applicable in many contexts (see Hofstetter and Sgroi, 1996; Kinneavy, 1996; Garfield and Gal, 1999; Watson, 2000; Watson, 2002; Gal, 2002a; Best, 2005; Trewin, 2005; Gigerenzer, 2008; Gould, 2010). For example, a National Statement on Mathematics for Australian Schools (Australian Education Council [AEC] 1991, 178) contains a call for students to understand the impact of statistics on daily life.

3. Implications of Statistical Literacy

In England, statistics was first called political arithmetic in the seventeenth century. The Germans gave it the title «Staatenkunde» from which came the word «statistics,» meaning a collection of facts about the state of importance to statesmen. In 1839, a census of agriculture in France included questions on the production of corn per hectare, and the number of livestock. In the same year, a committee of the Statistical Society of London, which in 1885 became the Royal Statistical Society, was appointed to report on the best method of taking the census of 1841. Thus, over a hundred and seventy years ago, the Statistical Society of London was cooperating with the organization of the census; the same is obtainable with the American Sta-

50 The African Statistical Journal, Volume 21, September 2019 3. Statistical Literacy for National Development

tistical Association in the United States. The burst of statistical activity around the 1830’s in the continent of Europe was due to interest in the activities and matters that concerned citizens’ welfare (see William, 1940).

Invariably, the statistical activity led to data generation for social, economic, and environmental indicators that informed policies for national development by statesmen. The basis for measuring and monitoring the developmental goals, targets, and indicators of various national and international programs was firmly established for governance, and accountability of governments. Data collection and use was not only fashionable but also done with great passion. This can be linked to good governance, economic transformation, and improved living standards in the European continent.

In Africa, many programs initiated by statistical offices and statistical training institutions located outside Africa drove statistical development since the 1960s. The program funded by the United Nations Development Programme (UNDP) called Statistical Training Programme for Africa was adopted in 1978 by the Economic Commission for Africa (ECA) Conference of Ministers. The aim of the program was to ensure that the African region had a consistent supply of qualified statistical staff. The second program was the Addis Ababa Plan of Action for Statistical Deve- lopment in Africa adopted by ECA Conference of Ministers in May 1990. Among others, the objectives of the Plan of Action were to; achieve national self-sufficien- cy in statistical production; ensure autonomy of the National Statistical System, and improve the coordination of all statistical development programs at both national and international levels.

After fifty years of concerted effort in improving statistical activity, it remains a mirage in Africa, as citizens have no feel for numbers and governments take no recourse to statistical data. For example, the Federal Government of Nigeria inau- gurated an Economic Recovery and Growth Plan and projected that Nigeria will make significant progress to achieve structural economic change with a more diver- sified and inclusive economy in five key areas by 2020. This happened without a basic monitoring and evaluation framework in place advising the citizens over the period of the strategy. This indicated absence of statistical literacy in the Nigerian government, and people.

Statistical literacy is required as a form of intervention to undo misinterpretations,

Journal africain de statistiques, numéro 21, septembre 2019 51 Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

misconceptions, and misleading information in a society. It exposes citizens to data usage as part of monitoring and evaluation promoting the culture of building a framework for data collection. This helps individuals to understand the statistical requirements for specific tasks. Ultimately, statistical literacy allows citizens to use numbers intelligently in decision-making in today’s information age.

Active participation in the society is made possible through statistical literacy (Na- tional Council of Teachers of Mathematics, 2000). Statistical literacy enables individuals, corporate bodies, stakeholders, policymakers, and governments to appreciate that programs should not only have achievable goals but also quantifiable targets supported by data, with a reporting framework for monitoring and evaluation purposes.

4. Nigerian National Statistical System

The objectives of the Nigerian National Statistical System (NSS) as stipulated in the Statistics Act of 2007 are, to; raise public awareness about the importance and role of statistical information to society. It attempts to collect, process, analyze, and disseminate quality statistical data; promote the use of best practice and international standards in statistical production, management, and dissemination. The NSS attempts to promote the use of statistical data and information at individual, institutional, Local Government Area, State, National and International levels, especially for evidence-based policy design, and decision-making with a view to build sustainable capacity for the production and use of statistical data, and information in the country for planning purposes.

The current national statistical system comprises of data producers, data users, data suppliers, research, and training institutions. The institutions involved in the data production and compilation include NBS, Central Bank of Nigeria (CBN), National Population Commission, Department of Planning, Research and Statistics of Minis- tries and Parastatals, the State Statistical Agencies, Budget and Planning of Local Government Councils, among others. The data users are the group that utilizes statistical products and services. The members are quite diverse and ever increasing. Some of them include policy and decision-makers in government ministries and institutions, politicians (e.g. members of the national and state assemblies, political parties, etc.), researchers, academicians, Non-Governmental Organizations (NGOs), donor community, international organizations, the media, and the public.

52 The African Statistical Journal, Volume 21, September 2019 3. Statistical Literacy for National Development

Data suppliers provide data to collecting agencies. These are mainly individuals, groups, households, and establishments. The research and training institutions include the Nigerian Institute of Social and Economic Research, Centre for Econo- metric and Allied Research, and Federal Institute of Industrial Research, Oshodi. The training institutions also include, the Federal School of Statistics and the De- partment of Statistics in Nigerian tertiary institutions, particularly, the Department of Statistics in the Nigerian Premier University, the University of Ibadan, and the Department of Mathematics/ Statistics in the Federal Polytechnic, Offa.

The four components of the Nigerian National Statistical System are interconnec- ted and the link is statistical data and information. Statistical literacy will enhance effective functionality and data quality of the system to produce a multiplier effect on development in the society.

5. Statistics and Economic Transformation

Our economy’s complexity, growth, and rapid structural changes require that public and private leaders have unbiased, relevant information on which to base their decisions. Data on real Gross Domestic Product (GDP), Consumer Price Index (CPI), and the trade deficit, for example, guide government spending, budget projections, and the allocation of federal funds. They are also essential inputs to monetary, fiscal, trade, and regulatory policy. Economic data, such as measures of price change, have a significant influence on interest rates and cost of living adjustments that -af fect every Nigerian who runs a business and saves for retirement. Similarly, timely, comparable data on the characteristics of the population are crucial in monitoring and responding to societal changes. Population figures from the census are used for reapportionment and allocation of hundreds of millions of naira every year (see Wallman, 2010a).

Business firms engaged in interrelated set of activities in producing goods, buying materials, employing personnel, and making profits should keep consistent production and financial records. For example, data on manufacturing shipments, inven- tories and retail sales, employment, earnings, prices and profits are used to make judgments on current economic conditions and prospects. They are used in com- bination to compile Gross National Product (GNP) accounts, estimate productivity figures and to deflate sales and earnings figures (see Leontief, 1971). Employment figures and changing prices are used in public and private decisions on monetary

Journal africain de statistiques, numéro 21, septembre 2019 53 Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

and business policies. In countries like the US, the stock market rises and falls based on employment numbers (Wallman, 2010b).

Transportation statistics gives details on roads usage and where new roads may be needed. Trends and phenomena of rates, population growth, incidence of diseases, industrial production, and employment trends contribute to people’s choice in chance-based situations (e.g., buying lottery tickets or insurance advice). Salaries and wages can support informed participation in public debate in many workplaces, given statistical information about the quality of processes and a good understanding of data about the status of the organization (see Bowen and Lawler, 1992).

Statistics produced for the purpose of monitoring the economy are also increasingly used to direct payment flows to individuals and to political units (Burton, 1973). This creates a powerful interest on the part of data producers/consumers shaping the way statistics are defined and collected. The increased use of statistics for directing money flows underscores the crucial importance of preserving the independence of the National Statistical System and insulating it from the pressure of special interest groups.

6. Statistical literacy model

A brief detour to describe the cognitive and dispositional abilities required for statistical literacy as proposed by Gal (2002b) is as follows:

The statistical literacy model includes the knowledge-base and other processes that should be available to adults and by implication, to learners graduating from colle- ges enabling them to comprehend, interpret, critically evaluate, and react to statistical information encountered in different contexts. The model assumes that people’s statistical literacy involves knowledge component that comprise of five cognitive elements: literacy skills, statistical knowledge, mathematical knowledge, context knowledge, and critical questions. In addition, a dispositional component made up of two elements, critical stance and beliefs and attitudes is expected. As with people’s overall numeracy (Gal, 2000), the components and elements in the proposed model should not be viewed as fixed and separate entities but as a context-dependent, dynamic set of knowledge and disposition that together enable statistically literate behavior. Five key parts of statistical knowledge-based requirements for statistical literacy proposed include understanding why data are needed and how

54 The African Statistical Journal, Volume 21, September 2019 3. Statistical Literacy for National Development

data can be produced; familiarity with basic terms and ideas related to descriptive statistics; familiarity with basic terms and ideas related to graphical and tabular dis- plays; understanding basic notions of probability, and appreciating how statistical conclusions or inferences are reached.

Scheaffer, Watkins, and Landwehr (1998), opined that an obvious prerequisite for comprehending and interpreting statistical messages is knowledge of basic statistical and probabilistic concepts and procedures, as well as related mathematical concepts and issues. The statistical knowledge-based proposals include making sense of numbers, understanding variables; interpreting tables and graphs. It has also proposed some aspects of planning a survey or experiment, like what constitutes a good sample, or methods of data collection and questionnaire design. This also includes data analysis processes, such as detecting patterns in univariate or two-way frequency data, or summarizing key features with summary statistics; re- lationships between probability and statistics, such as in determining characteristics of random samples, background for significance testing and inferential reasoning, such as confidence intervals or testing hypotheses.

However, the Nigerian situation reflects lack of regard for data, thus the statistical literacy model proposed by Gal (2002b) and Scheaffer, Watkins, and Landwehr (1998) is recommended, as it emphasizes the use of data as a first step (see Figure 1).

Figure 1: Figure 1: A model of statistical literacy

Source: Compiled by authors

Journal africain de statistiques, numéro 21, septembre 2019 55 Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

7. Progress in Statistical Literacy

There is a shift of emphasis in statistics instruction from procedural understanding of statistical techniques, formulas, computations, and procedures to developing conceptual understanding and statistical reasoning and thinking. Researchers began to take an interest in studying how children understand basic concepts related to data analysis, and how to develop good statistical reasoning and understanding as part of instruction in elementary and secondary mathematics classes.

The issues of developing teacher knowledge of statistics as well as methods of helping teachers understand the big ideas of statistics can be found in the joint IASE-ICMI study. TEAM project (Franklin and Mewborn, 2006) attempted to bring mathematics educators and statisticians together, creating new ways to pre- pare future K-12 teachers of statistics, making sure the students have a course in statistics as part of their requirements. The students were taught in methods that emphasize conceptual understanding, data exploration, and use of appropriate technology. There is a growing network of researchers interested in studying the development of students’ statistical literacy, reasoning, and thinking. The topics on these research studies conducted by members of this community reflect the shift in emphasis in statistics instruction, from focusing on procedural understanding (Joan and Ben-Zvi, 2007).

Several countries and organizations have introduced programs to improve school-level education on data-analysis and probability, sometimes called data handling, sto- chastics, or chance (see Australian Education Council, 1991; National Council of Teachers of Mathematics, 2000; Plante and Reid, 2011). The American Statistical Association and the Royal Statistical Society are two leaders, among professional organizations worldwide, in the design and trialing of curricula and resource materials for courses in statistical literacy at both school and university levels. There are also, of course, many individual initiatives, piloted by innovative statistics educators around the world (see Snell, 2002; Sowey, 2003; Olubusoye, 2014; Olubusoye, 2017). More voices should be raised to emphasize the importance of developing statistical literacy skills applicable in many contexts.

5 See http://www.ugr.es/~icmi/iase_study

56 The African Statistical Journal, Volume 21, September 2019 3. Statistical Literacy for National Development

8. Challenges in Achieving Statistical Literacy

Linking of statistical literacy with other terminologies used in educational circles to attract attention to our course may constitute a challenge. Positioning of statistical literacy within the school curriculum to achieve maximum exposure particularly where many of the staff will have phobias related to past experiences with mathematics and statistics will be a big challenge. The application of what we will be learning from research about statistical literacy into school students’ understanding of statistical literacy may also constitute a barrier (see Watson 2002).

Statistical literacy has a very low profile in the school curriculum, thus making a broad connection with other areas of study needs to be carefully made so that statistical literacy will not be considered the domain in schools. The concept en- compasses a multifaceted nature of many social problems that call upon knowledge from so many subject areas, such that teachers in most subjects would suggest that it was another subject teacher’s responsibility to provide knowledge in a particular area. This will however, make the approach more interesting and most challenging for statistics teachers and lecturers.

Statistical literacy is the most nebulous and abstract of all statistical topics which do not appear as a standard topic in introductory statistics syllabus. This results in the inconsistent treatment and level of attention paid to statistical literacy as an introductory course in statistics. How do we determine the extent to which statistical literacy skills are needed by students in school, in everyday life and in workplace? Other worries include attitude and motivation toward statistical literacy and how we modify our courses and teaching methods to improve students’ statistical IQ level? How do we help, encourage, develop and implement this significant behavioral change? (see Rumsey 2002a).

Creating a measurable change in statistical literacy for the general population is a complex task. Attention is required to the need to develop multiple knowledge elements and dispositions to the nature of the diverse players that can contribute

6 SRTL - The International Statistical Reasoning, Thinking, and Literacy Research Forums, http:// srtl.stat.auckland.ac.nz/ 7 See also Jane Watson’s web site called «Chance and Data in the News, and http://www.themercury. com.au/nie/mathguys/mercindx.htm

Journal africain de statistiques, numéro 21, septembre 2019 57 Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

to students’ and peoples’ statistical literacy, and to the unique characteristics and needs of different target groups: specifically, current adults outside the purview of any formal education system and future adults served by regular school systems within the formal educational system. Another concern is how to better address statistical literacy as part of current instructional efforts in statistics in terms of curriculum design, teaching methods and learning resources given that the infrastructure for instruction already exist and some teachers and students are already engaged in teaching and learning statistics (Cerrito, 1999; Gal, 2002a). While, at the same time, many students are not currently receiving instruction in statistics (Schmidt, Mcknight, Valverde, Hovang, and Wiley, 1997; Moore and Cobb, 2000)

School textbooks are often conservative and lag behind innovative curriculum frameworks, and teachers may have relatively little freedom to choose what textbook to use, depending on the degree of centralization of textbook production and adoption. Textbooks that support instructions for statistical literacy may be unacceptable to teachers who view the goals of their introductory statistics classes as emphasizing knowledge of formal aspects of statistics procedures and rules for inference. Yet, calculation skills do not necessarily lead to intuitive understanding (Weldon, 2002).

9. Pathways to Enhancing Statistical Literacy

9.1. Collaboration between the Nigerian Statistical Association (NSA) and National Bureau of Statistics (NBS)

The Statistical Society of Australia and the Australian Bureau of Statistics have been working on a strategy to ensure that Australian school children acquire a suffi- cient understanding and appreciation of how data can be acquired and used so they can make informed judgments in their daily lives, as children and then as adults, similar collaboration is proposed between the Nigerian Statistical Association (NSA) and National Bureau of Statistics (NBS).

The challenges of statistical literacy are to be addressed by statisticians in the field, and educators in the classroom. Thus, collaboration between the NSA and NBS will allow for enhancement of statistical literacy in the National Statistical System. The media could be used by the two bodies to educate the public regarding the need to use data, be critical of information, and how to critically evaluate and determine the

58 The African Statistical Journal, Volume 21, September 2019 3. Statistical Literacy for National Development

quality of information. This could encourage political and commercial decisions that affect her citizens to be based on data.

Collaborative efforts should be made to integrate statistical literacy into the school curriculum at primary and secondary levels, and in tertiary institutions. Similar to the American Statistical Association and the Royal Statistical Society, the NSA should take the lead in the design and trialing of curricula and resource materials for courses in statistical literacy at both school and university levels. Statistical literacy education for adults could follow different paths, and examined quite independently from the schools’ perspective.

It is imperative to appreciate the audience’s perspective and set the parameters for statistical literacy in non-threatening ways through grassroots efforts to a joint effort between teachers and school administrators. They must be met at their own levels with examples from their disciplines, and emphasize the responsibility of teachers in other school curriculum areas.

9.2. National Statistical Centre

A National Statistics Education Centre was seen as important for providing a focus for the development of new resources as well as a clearinghouse for existing resources to teach statistical literacy in Australia.

In Nigeria, there is a need for a centralized resource of materials, and instructor training, and support to provide an increasing momentum toward providing relevant context statistics literacy training. A National Statistical Centre is therefore proposed. This will help statistics educators not to re-invest the wheel or spend too much time individually on finding relevant contexts for students. The Natio- nal Statistical Centre will provide statisticians, statistics educators, researchers, and policymakers a point to work together to champion statistical literacy as a valued educational and civic goal. The center will enhance the statistics knowledge-base and professional development options available to educators working with pupils, secondary school students, and undergraduate and post graduate students. To ease transition, statistics educators need to be given enough resources such as access to rich and field-tested teaching materials, and assessment tools. In addition, the center should be responsible for emphasizing the following issues to educators at all levels:

Journal africain de statistiques, numéro 21, septembre 2019 59 Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

Linking statistical literacy with current events covered in the news media as an effective way of motivating interest in statistical literacy rather than using artificial setting in textbooks. For example, statistical literacy is more than being able to calculate arithmetic mean correctly without a surfeit of examples from other parts of the curriculum where applying the arithmetic mean is an essential part of decision-making.

Students should be made to experience relevant contexts and formulate questions of interest as an inherent part of statistical literacy. All examples, homework, and even test questions should be presented within a relevant context. The need for usable robust knowledge and incorporating the generation of questions should form arcane scholarly activity in investigation which provides much of the excitement.

Training in statistics could be poorer and much less valuable than it should be if the present dispensation is continued (see Moore 1998; Wild, Pfannkuch, Regan and Horton, 2011; Smith, Molinaro, Lee and Guzman-Alvarez, 2014). Teachers who see their main role as dealing with principles of statistics or with rules for inference must know that more work needs to be done in precisely identifying the issues, problems, or dilemma before the need for descriptive and inferential statistics is addressed (see Mccall, 1998; Batanero, 2002).

Ultimately, the goal of the center will be to achieve improved data use, knowledge component skills, dispositional components, and statistical knowledge-base in Ni- geria.

9.3. Statistical Literacy Website and Publications

Judging by the current trend of communication technology and availability of smart phones, television, IPADS, and internet facilities, statistical literacy website could be used to promote statistical literacy by volunteers from members of the NSA, statistical educators and stakeholders. Appropriate safeguards should be in place to protect volunteers against the risks of public liability and the possibility of conflict of interest with the requirements of their employers. Face-to-face consultation could be offered when and where appropriate. In addition, the NSA needs to expand its range of publications to address the need of a more varied audience, a good start is the use of magazines to acquire general readership.

60 The African Statistical Journal, Volume 21, September 2019 3. Statistical Literacy for National Development

10. Conclusion

Development in the direction of appreciating the place of data will accelerate the effectiveness of the approach for statistical literacy in Nigeria. Adoption ofthis modern statistical method will be dependent on the cooperation of stakeholders. The interest and enthusiasm will continue to develop when statistical activities in the National Statistical System are the responsibility of professional statisticians, and inform policy decisions for economic transformation targeted towards national development. Further works on how to teach statistical literacy should be the area of future research.

Journal africain de statistiques, numéro 21, septembre 2019 61 Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

REFERENCES

Australian Education Council (AEC) (1991). “A National Statement on Mathema- tics for Australian Schools.” Carlton, Victoria, Australia: AEC.

Batanero, Carmen (2002). “Discussion: The Role of Models in Understanding and Improving Statistical Literacy.” International Statistical Review / Revue Internatio- nale de Statistique 70 (1): 37-40.

Best, Joel (2005). “Lies, Calculations and Constructions: Beyond How to Lie with Statistics.’’Statistical Science 20 (3): 210-214. Bowen, D.E. and Lawler, E.E. (1992). “The Empowerment of Service Workers:What, why, how, and when.” Sloan Management Review: 31-39.

Burton, G. Malkiel (1978). “Problems with the Federal Economic Statistical Sys- tem and Some Alternatives for Improvement.”The American Statistician 32 (3): 81-88.

Cerrito, Patricia B. (1999). “Teaching Statistical Literacy.” College Teaching 47 (1): 9-13.

Cockcroft, W.H. (1982). “Mathematics counts: Report of the Committee of lnquiry into the Teaching of Mathematics in Schools.” London: HMSO.

Franklin, C.A. and Mewborn, D.S. (2006). “The Statistical Education of Grades Pre-K-2 Teachers: A Shared Responsibility.” In Thinking and Reasoning with Data and Chance: Sixty-eighth Yearbook, G.F. Burnii, Reston, VA (eds.): National Coun- cil of Teachers of Mathematics (NCTM).

Gal, Iddo (2000). “The Numeracy Challenge.” In Adult Numeracy Development: Theory, Research and Practice, Iddo Gal (ed.): NJ: Hampton Press. (2002a). “Res- ponse: Developing Statistical Literacy: Towards Implementing Change.”Internatio- nal Statistical Review / Revue Internationale de Statistique 70 (1): 46-51. (2002b). “Adults’ Statistical Literacy: Meanings, Components, Responsibilities.”Internatio- nal Statistical Review / Revue Internationale de Statistique 70 (1): 1-25.

62 The African Statistical Journal, Volume 21, September 2019 3. Statistical Literacy for National Development

Garfield, Joan. (1999). “Thinking about Statistical Reasoning, Thinking, and Lite- racy.” Paper presented at First International Research Forum on Statistical Reaso- ning, Thinking and Literacy (STRL-1), Kibbutz Be ‘eri, Israel.

Garfield, Joan. and Dani Ben-Zvi (2007). “How Students Learn Statistics Revisited: A Current Review of Research on Teaching and Learning Statistics.”International StatisticalReview / Revue Internationale de Statistique 75 (3): 372-396.

Garfield, J .B. and Gal, I. (1999). “Teaching and Assessing Statistical Reasoning.” In Developing Mathematical Reasoning in Grades K-12. 1999 Yearbook, L.V. Stiff (ed.): 207-219. Reston, VA: National Council of Teachers of Mathematics.

Gigerenzer, Gerd (2008). “Sound Reasoning Requires Statistical Understanding.” Science News 174 (11): 32.

Gould, Robert (2010). “Statistics and the Modern Student.” International Statisti- cal Review / Revue Internationale de Statistique 78 (2): 297-315.

Hawkins, Anne (1997). “New Pedagogy and New Content: The Case of Statistics: Discussion.” International Statistical Review / Revue Internationale de Statistique 65 (2): 141-146.

Hofstetter, Elaine B. and Laura A. Sgroi (1996). “Data with Snap, Crackle, and Pop.” Mathematics Teaching in the Middle School 1(March-April 1996): 760-64.

Kinneavy, Kevin (1996). «The Pond: Doing Research Together.» Mathematics Tea- ching in the Middle School 1 (March-April 1996): 696-702.

Leontief, Wassily (1971), «Theoretical Assumptions and Non Observed Facts.” American Economic Review: 61: 1-7.

McCall, C. (1998). “Commentary: Important Topics in Statistics.” Amstat News 252: 76.

Moore, D.S. (1998). “Statistics among the Liberal Arts.” Journal of the American Statistical Association 93 (444): 1253-1259.

Journal africain de statistiques, numéro 21, septembre 2019 63 Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

Moore, D.S. and Cobb, G.W. (2000). “Statistics and Mathematics: Tension and Cooperation.” American Mathematical Monthly 107 (7): 615-630.

National Council of Teachers of Mathematics (2000). “Principles and Standards for School Mathematics.” Reston, VA: National Council of Teachers of Mathematics.

Olusanya, Olubusoye E. (2014). “Statistical Literacy and Empirical Modelling for National Transformation.” Faculty lecture, University of Ibadan, Ibadan, Nigeria. Available at https://www.researchgate.net/publication/267324628(2017). “Statisti- cian Canvasses Evaluation Mechanism in FG’s Economic Recovery Plan.”http:// thenationonlineng.net/statistician-canvasses-evaluation-mechanism-in-fgs-economic-recovery-plan/

Ottaviani, Gabriella M. (2002). “Discussion: Statistics, from a Tool for State and Society to a Tool for All Citizens.» International Statistical Review / Revue Inter- nationale de Statistique 70 (1): 30-32.

Plante, Jean-François and Nancy Reid (2011). “Statistics in the News.” The Ameri- can Statistician 65 (2): 80-88.

Rumsey, D.J. (2002a). “Statistical Literacy as a Goal for Introductory Statistics Courses.” Journal of Statistics Education [Online] 10 (3): www.amstat.org/publications/jse/v10n3/rumsey2.html (2002b). “Discussion: Statistical Literacy: Impli- cations for Teaching, Research, and Practice.” International Statistical Review / Revue Internationale de Statistique 70 (1): 32-36.

Scheaffer, R. L. (2001). «Quantitative Literacy and Statistics.» AmStat News, No- vember; also online at http://www.amstat.org/publications/amstat_news/ 2001/ presl l.html

Scheaffer, R.L., Watkins, A.E. and Landwehr, J.M. (1998). “What every High School Graduate Should Know about Statistics.” In Reflections on Statistics: Lear- ning, Teaching and Assessment in Grades K-12, S.P. Lajoie (ed.):3-31. NJ: Law- rence Erlbaum Associates.

Schmidt, W.H., McKnight, C.C., Valverde, G.A., Houang, R.T. and Wiley, D.E. (1997). “Many Visions, Many Aims.” (Volume 1). A Cross-national Investigation

64 The African Statistical Journal, Volume 21, September 2019 3. Statistical Literacy for National Development

of Curricular Intentions in School Mathematics. Dordrecht: Kluwer. Schoenfeld, A.H. (2001). “Reflections on an Impoverished Education.” In Mathe- matics and Democracy: The Case for Quantitative Literacy, L.A. Steen (ed.). USA: National Council on Education and the Disciplines.

Smith, Amy, Marco Molinaro, Alisa Lee, and Alberto Guzman-Alvarez (2014). “Thinking with Data: Students use Authentic Data to Form Evidence-Based Conclu- sions and Think like a Scientist.” The Science Teacher 81 (8): 58-63.

Snell, L. (1999). “Using Chance Media to Promote Statistical Literacy.” Paper Pre- sented at the 1999 Joint Statistical Meetings, Dallas, TX.

Snell, J. Laurie (2002). “Discussion: But How Do You Teach It?” International Statistical Review / Revue Internationale de Statistique 70 (1): 45-46.

Sowey, Eric R. (2003). “The Getting of Wisdom: Educating Statisticians to En- hance Their Clients Numeracy.” The American Statistician 57 (2): 89-93.

Statistics Canada and Organisation for Economic Cooperation and Development (OECD) (1996). “Literacy, Economy, Society: First Results from the International Adult Literacy Survey.” Ottawa.

Steen, L.A. (Ed.) (1997). “Why Numbers Count: Quantitative Literacy for Tomor- row’s America.” New York: The College Board.

Trewin, Dennis (2005). “Improving Statistical Literacy in Schools in Australia.” In- ternational Statistical Review / Revue Internationale de Statistique73 (2): 211-212.

UNESCO (2000). “World Education Report: The Right to Education-Towards Edu- cation for all throughout Life.” Paris.

Wallman, Katherine (1993). “Enhancing Statistical Literacy: Enriching Our So- ciety.” Journal of the American Statistical Association 88 (421): 1-8.(2010a). “Fe- deral Statistics: Understanding a Crucial Resource.” The Annals of the American Academy of Political and Social Science 631: 22-24. (2010b). “Top Statistician Explains What all those Numbers Mean.” Science News 178 (10): 32.

Journal africain de statistiques, numéro 21, septembre 2019 65 Korter Grace, Olatunji Lateef, Omolehin Joseph, Olubusoye Olusanya

Watson, Jane M. (2000). “Statistics in Context.” Mathematics Teacher 93: 54-58. (2002). “Discussion: Statistical Literacy before Adulthood.” International Statisti- cal Review / Revue Internationale de Statistique 70 (1): 26-30.

Weldon, K. L. (2002). “Adults’ Statistical Literacy: Meanings, Components, Res- ponsibilities: Discussion.” International Statistical Review / Revue Internationale de Statistique 70 (1): 43-44.

Wild, Chris (2005). “Education Is Everybody’s Responsibility.” International Sta- tistical Review / Revue Internationale de Statistique 73 (2): 213-214.(1994). “Em- bracing the «Wider View» of Statistics.” The American Statistician 48(2): 163-171.

Wild, C., Triggs, C. and Pfannkuch, M. (1997). “Assessment on a Budget: using Traditional Methods Assessment Challenge in Statistics Education.” Iddo Gal and J.B. Garfield (eds.): 205-220. The Netherlands: Amsterdam, International Statisti- cal Institute/IOS Press.

Wild C. J., M. Pfannkuch, M. Regan and N. J. Horton (2011). “Towards more Ac- cessible Conceptions of Statistical Inferences [with Discussion].” Journal of the Royal Statistical Society. Series A (Statistics in Society), 174 (2): 247-295.

William, Fielding Ogburn (1940). “Statistical Trends.” Journal of the American Statistical Association 35 (209): 252-260.

66 The African Statistical Journal, Volume 21, September 2019 4. Statistical Indicators for Measuring Good Governance in Africa 4. Statistical Indicators for Measuring Good Gover- nance in Africa

Dahud Kehinde Shangodoyin 1

Abstract

This article focuses on the measure of good governance with measurable indicators. The importance of statistics has been growing rapidly with the greater integration and inter-dependence of Africa economies, which is evident from burgeoning data demands at national, regional, and continental levels. Both micro and macro statistics have immense significance for understanding the socio-economic reality and thereby appropriate policy formulation.

The National Strategy for the Development of Statistics (NSDS) in Africa would be robust with the inclusion of simple methodology that can measure well-being of people holistically. The primary objectives in this article will be to develop measurable indicators of good governance at all levels, to examine how applicable and acceptable the developed well-being indicators are in evaluating and monitoring government performance, and how these indicators can be integrated in the monitoring and evaluation system within the framework of NSDS. Parameters that could be used for measuring well-being have been identified, and a simple measure of good governance on the provision of minimum basic needs is proposed, using available statistics collected by Na- tional Statistical Agency.

Keywords: Governance, minimum basic needs, national strategy for development of statistics, well-being indicators.

Résumé

Cet article se concentre sur la mesure de la bonne gouvernance avec des indicateurs mesurables. L’importance des statistiques a rapidement augmenté avec l’intégration et l’interdépendance accrues des économies africaines, comme en témoigne la demande croissante en données aux niveaux national, régional et continental. Les statistiques micro et macro ont une importance capitale pour comprendre la réalité socio-écono- mique et par conséquent pour formuler une politique appropriée.

1President of the African Statistical Association, [email protected] Prof D. K. Shangodoyin, is the President of African Statistical Association. Corresponding email address: [email protected]

Journal africain de statistiques, numéro 21, septembre 2019 67 Dahud Kehinde Shangodoyin

La Stratégie Nationale de Développement de la Statistique (SNDS) en Afrique serait robuste avec l’inclusion d’une méthodologie simple permettant de mesurer le bien-être des personnes de manière globale. Les principaux objectifs de cet article seront de développer des indicateurs mesurables de la bonne gouvernance à tous les niveaux, d’examiner dans quelle mesure les indicateurs de bien-être développés sont appli- cables et acceptables dans l’évaluation et le suivi de la performance du gouvernement, et comment ces indicateurs peuvent être intégrés pour la suivi et gestion du système d’évaluation dans le cadre de la SNDS. Les paramètres pouvant être utilisés pour mesurer le bien-être ont été identifiés, et une simple mesure de la bonne gouvernance en matière de fourniture de besoins de base minimum est proposée, à l’aide des statistiques disponibles rassemblées par l’Agence nationale de la statistique.

Mots-clés: gouvernance, besoin minimum de base, stratégie nationale de développe- ment des statistiques, indicateurs de bien-être.

1. Introduction

In the 18th century, governments at all levels had to come up with developmental plans that gave their respective populations better life and reduced poverty; thus, the strong need for methodological measures of programs’ progress became necessary.

Statistical science activities began with Roman and Greece cities which tried to organize censuses and surveys mainly carried out to determine the number of their respective populations, and generally, for planning purposes; since then, statistics became continuously useful in different domains to be considered as the milestone for an informed decision-making in both private and public institutions.

Statistics play a key role in planning, monitoring and evaluation of socio-economic policies (Shangodoyin and Lasisi, 2011); this is because availability of reliable and accurate data during the implementation process ensures the effective control on delivery of various public and private services, and thus results in good governance. How can governance be defined to the understanding of ordinary average person?

Governance could be defined as the style in which power and authority are exer- cised in the management of a country’s socio-economic, natural, and physical re-

68 The African Statistical Journal, Volume 21, September 2019 4. Statistical Indicators for Measuring Good Governance in Africa

sources. But, governance is not just about how government conducts business in its own sphere; it also entails how government connects and cooperates with civil society and how well government facilitates and encourages participation of people in the delivery of goods and services as well as in monitoring and evaluation of government performance (Landell-Mills and Serageldin, 1992).

The key to improve the efficiency and effectiveness of public and private sector management is to make available, reliable and timely official statistics that give necessary impetus and technical support to the state agencies, institutional drivers, and civil society to act in the collective interest at the least cost to society. For instance, incomplete capturing of Civil Registration and Vital Statistics (CRVS) leads to the scandal of invincibility, where many people are born and their existence is not documented throughout their lifetime.

According to United Nations Statistics Division (UNSD, 2016), one major challenge that countries in Africa face is the weak coordination or cooperation among the different stakeholders in CRVS. The existence and functions of a coordination mechanism among different agencies is often not clearly stated in the civil registration law or regulation. Descriptions on how information on the registered vital events should be transferred to either upper-level administrative offices or to a different agency are mostly missing from the legislation. Out of seventeen Engli- sh-speaking countries in Africa surveyed in the UNSD report, only four countries have created a coordination agency or committee (Egypt, Lesotho, Mozambique, and Zimbabwe). In other countries, for example the Gambia and Swaziland, the system is fragmented with limited coordination; sectors concerned do not or rarely coordinate their activities such as data collection, data management, and dissemination or data accessibility.

Even when there are specifications in the civil registration regulation on the transfer of information and coordination, such as in Lesotho, the law is often not strictly followed by the executing agencies. For example, the transmission of data from the civil registration authority to the national statistical office is arranged on an ad-hoc basis and not systematically as would be suggested later in this article.

The CRVS system in Africa still has profound weaknesses such as inadequate legal

Journal africain de statistiques, numéro 21, septembre 2019 69 Dahud Kehinde Shangodoyin

institutional frameworks as well as operational inefficiencies, together with poor funding constraints. A complete CRVS is a prerequisite for effective governance and any impediments to it must be strongly combated. Statistics generated through civil registration should contribute to the formulation of evidence-based policies across all sectors. This article calls for a review of literature on socio-economic and environmental development, and its causality on minimum basic needs; hence the need for this qualitative study.

The perspectives of socio-economic and environmental development have been changing over the years in a rapid fashion to accommodate themselves to changing needs and perceptions of the people and the economies; the proposals for solving the socio-economic developmental problems have been overwhelmingly large and varied in nature (Kohli, 1987). The strategies of development may be divided into two approaches such as growth-oriented and basic needs with only one fundamental difference between them. The former approach is based on income while the latter is based on supply of basic services.

In general, strategies on development will provide useful results if adequate, timely, and reliable statistics are readily available. Statistics as a body of empirical evidence is required by governments and their development partners not only to monitor and evaluate progress on developmental projects and give good governance, but it is also useful to prioritize the allocation of scarce human and natural resources across the competing local councils, districts/provinces, states and federal Ministries, Departments and Agencies (MDAs).

2. A Review of Strategies Adopted in Measuring Well-being in Some African Countries

Below is a succinct review of the strategies used in developing statistical indicators for measuring well-being in three fast moving economies in Africa; Nigeria, Kenya, and South Africa.

The Nigeria Living Standard Survey Report 2010 and the Poverty profile of Nige- ria (1980 – 2010) were produced to assist various levels of government to evaluate and monitor their social, and economic programs. The World Bank, the UK De-

70 The African Statistical Journal, Volume 21, September 2019 4. Statistical Indicators for Measuring Good Governance in Africa

partment for International Development (DfID), and other development partners assisted Nigeria in improving the availability of statistical data on which effective poverty monitoring and evaluation depend on. Also in Kenya, the key objectives of the Kenya Integrated Household Budget Survey (KIHBS) are to provide measures of living standards and updated poverty, and inequality that reflect the well-being of Kenyans. The Government collaborated with various development partners, namely, the Department for International Development (DfID), the United States Agency for International Development (USAID), the European Union (EU), the Danish International Development Agency (DANIDA), and the United Nations De- velopment Programme (UNDP).

Statistics show that South Africa has run the General Household Survey (GHS) on a regular basis since 2002, covering a variety of multidimensional poverty measures. In particular, the GHS covers six broad areas: education, health, activities related to work and unemployment, non-remunerated trips undertaken by the household, housing, and household access to services and facilities. The survey was instituted because the Government of South Africa needs to determine on a regular basis, the level of development in the country and the performance of development programs and projects. StatsSA has been working with international partners to generate area statistics on poverty; it partnered with the Centre for the Analysis of South African Social Policy (CASASP), and the Human Sciences Research Council (HSRC) to develop Provincial Indices of Multiple Deprivation (PIMD) for South Africa.

Surveys conducted by these statistical agencies had financial contributions from development partners; this was perhaps because of shortfall in budgetary allocation by various governments.

The three countries’ strategies on well-being measurement discussed have not covered extensively all the factors enumerated below. The NSDS in African countries should include the use of simple methodology discussed above that can measure the well-being of citizens including all areas of social concerns. The primary objective should be to develop measurable indicators of good governance at all levels, to examine how applicable and acceptable the developed well-being indicators are in evaluating and monitoring government performance.

Journal africain de statistiques, numéro 21, septembre 2019 71 Dahud Kehinde Shangodoyin

It is of great concern that without the financial assistance from development partners, most African countries are not able to provide data for measuring well-being indicators. The African leadership should provide adequate funding for data collection and analysis for comprehensive measurement of citizens’ well-being.

3. Statistical Indicators for Gauging Governance in Africa

The importance of statistics has been growing rapidly with the greater integration and inter-dependence of African economies, which is evident from burgeoning data demands at national, regional, and continental levels. Both micro and macro statistics have immense significance in understanding the socio-economic reality and thereby leads to appropriate policy formulation.

Observations show that the quality of governance and not the type of political regime makes the difference in the economic performance of countries (Root, 1995). The project performance of countries largely depends on four dimensions of governance: capacity and efficiency of public sector, accountability, legal framework for development, transparency and statistical information (World Bank, 1992). For instance, statistical information on infrastructure, health, education, and socio-economic conditions of society are essential for planning and policy formation at local, state/district, and federal levels.

The Minimum Basic Needs (MBN) approach has been considered by international agencies like the World Bank and the IMF as the most effective instrument for the betterment of the Third World. MBN can be an ideal instrument for giving necessary impetus to work on socio-economic indicators of the developing countries.

3.1. Formulation of Minimum Basic Needs Indicators

Several sets of well-being indicators have been made to assess the accomplishment of governments in promoting human and social development. Conceptually, the MBN approach is considered as the most comprehensive and integrated approach to development. According to Ghai (1980), a government program may be defined as a basic needs activity if it incorporates some or all of the following features:

72 The African Statistical Journal, Volume 21, September 2019 4. Statistical Indicators for Measuring Good Governance in Africa

(i) it raises incomes of the ‘poverty groups’ to specified levels over a given period through creation of employment, redistribution of assets, and measures to enhance productivity; (ii) it directly contributes to the achievement of the targets established in respect of core basic needs like nutrition, health, education, housing, and safe drinking water supply; (iii) it increases production of other basic goods and services purchased by low-income groups from, their disposable incomes; and (iv) it promotes decentralization of power, people’s participation in political decision making, and self-reliance.

The following social concerns are peculiar to all African countries: health and nutrition, learning, income and consumption, employment, non-human productive resources, housing, utilities, the environment, public safety, justice, political values, and social-mobility. For each concern, a limited number of measurable indicators can be drawn up to gauge social concerns.

The indicators chosen should be reliable, replicable, and simple to interpret in order to be easily understood by the end-users and the public. In addition, there should be a preference in favor of indicators that touch on all social concerns. In other words, the indicators should reflect the entire areas of social-concerns in a single-sweep rather than the indicators computed on one-sided social-concerns. It will be of interest that present welfare should consist not only of the welfare of people currently alive but also of the welfare of future generations.

In Africa, the need to develop a single-swipe index on social concerns is generally perceived to be important due to political and technical factors involved in various aspects of governance. Construction of a composite measure of human development which expresses various components in a single magnitude might be difficult but several successful efforts have been made to alleviate these challenges. For instance, the Human Development Index (HDI) is based on three indicators: longevity, educational attainment and standard of living (UNDP 1996). The HDI has been wi- dely accepted and used for international comparison amongst countries. However, lack of opportunity signifies that access to the means of developing or maintaining an essential human capability is not being adequately provided.

Journal africain de statistiques, numéro 21, septembre 2019 73 Dahud Kehinde Shangodoyin

The rationale of the MBN approach is that the direct provision of health services, education, housing, sanitation, water supply, and adequate nutrition is likely to reduce absolute poverty more immediately than alternative strategies which merely rely on income (Burki and Haq, 1981).

The arguments in support of MBN approach are as follows:

(i) growth strategies fail to benefit the poor because of uneven distribution of economic opportunities; (ii) the productivity and incomes of the poor depend on the direct provision of health and education in the first place; (iii) it may take a long time to increase the incomes of the poor so that they can afford basic needs; (iv) the poor and illiterate tend not to spend their income wisely; (v) facilities like that of basic services can only be provided publicly; and (vi) it is difficult to help all the poor in a uniform way in the absence of basic needs provision.

Thus, indicators on the availability of social services which are the direct means to ensuring a decent standard of living would be of better use than proxy indicators or indirect measures like income as in the HDI. Even what constitutes income in developed countries is different from what is obtainable in developing countries.

Given that developing countries dominate Africa, it is important for policymakers to determine how sensitive the indicators are to changes in macro policies, i.e., the impact of macro policies on vulnerable groups. It is essential to focus on indicators that are sensitive to policy changes on social concerns. The Minimum Basic Needs (MBN) indicators would be suitable for this purpose. It will promote the construction of selective policies to target the basic needs of the whole population directly, rather than focusing on an indirect approach to satisfying basic human needs.

The MBN indicators should cover three important areas of social concern: survival, security, and enabling needs. Indicators under each area are enumerated in the table below.

74 The African Statistical Journal, Volume 21, September 2019 4. Statistical Indicators for Measuring Good Governance in Africa

Journal africain de statistiques, numéro 21, septembre 2019 75 Dahud Kehinde Shangodoyin

For the table above, most surveys conducted by the national statistical office can provide the ingredients needed for computing indicators for these KPAs but there could be a challenge on how to measure the components of enabling needs-participation.

The National Statistical Offices across the African continent should endeavor to -de velop the MBN index based on social concerns’ indicators (to cover all factors dis- played on the table above) to measure their populations’ level of well-being against the HDI mostly used for international comparison.

4. Conclusion

Observations show that all the surveys conducted by most statistical agencies had financial contributions from development partners; this is because of shortfall in budgetary allocation by national governments.

The three countries’ strategies on well-being indicators discussed have not covered extensively all factors enumerated in the table above. The NSDS in African countries should include the use of simple methodology that can measure the well-being of citizens in a comprehensive manner, including all areas of social concern. The primary objectives should be to develop measurable indicators of good governance at all levels, and to examine how applicable and acceptable the developed well- being indicators are in evaluating and monitoring government performance. Of importance is also to incorporate the same indicators in the NSDS framework.

It is of great concern that without the contribution of development partners, most African countries are not able to provide data for measuring well-being indicators. The African leadership should provide adequate budgetary allocation for data collection and analysis for comprehensive measurement of their citizens’ well-being.

76 The African Statistical Journal, Volume 21, September 2019 4. Statistical Indicators for Measuring Good Governance in Africa

REFERENCES

Alderman H., B. Miriam, D. Gabriel, M. Nthabiseng and Ö. Berk (2002) “How Low Can You Go? CombiningCensus and Survey Data for Poverty Mapping in South Africa.” Journal of African Economies, 11(2): 169–200.

Booysen, F (2002): “An overview and evaluation of composite indices of development.” Social Indicators Research 59(2): 115–151.

European Centre for Disease Prevention and Control (2017):http://ecdc.europa.eu/ en/healthtopics/ebola_marburg_fevers/pages/ebola-outbreak-west-africa.aspx Eu- rostat (2016): Africa-EU key statistical indicators. Statistics Explained. http://epp. eurostat.ec.europa.eu/statisticsexplained

Ghai, D P (1980): “The basic needs approach to development. Some Issues regarding concepts and methodology.” International Labour Office. ISBN 10-9221018016.

Gough, I. and Thomas, T (1994): “Need Satisfaction and Welfare Outcomes: Theo- ry and Explanations,” Social Policy and Administration, 28(1): 33–56.

Greer, J. and E. Thorbecke (1986): “A Methodology for Measuring Food Poverty Applied to Kenya.” Journal of Development Economics, 24: 59-74.

Herrenkohl, R., Judson, G and Heffner, J. (1999): “Defining and measuring -em ployee empowerment.” Journal of Applied Behavioral Science, 35: 373-385.

Hilbert JA (2014): “A systemic study of biofuels in complex agriculture markets.” In: Proceedings of the 22ndEuropean Biomass Conference and Exhibition (EUBCE) Hamburg, pp. 158–164.

International Labor Organization (1981): “Basic needs in an economy underpres- sure; findings and recommendations of an ILO/JASPA basic needs mission to Zam- bia.” Addis Ababa XLIII 201P ISBN 92-2-102683-3.

International Labor Organization (1981): “First thing first; meeting the basic needs of the people of Nigeria.”Addis Ababa X 256P ISBN 92-2-102682-5.

Journal africain de statistiques, numéro 21, septembre 2019 77 Dahud Kehinde Shangodoyin

International Labor Organization (1982): “Basic need in danger; a basic need oriented development strategy for Tanzania.” Addis Ababa XLI 416P ISBN 92-2- 103256-6.

International Labor Organization (1986): “Challenge of employment and basic needs in Africa,” Nairobi. Oxford University Press. ISBN 0-19-572559-x.

Javed Burki, S. and Ul Haq, M (1981): “Meeting Basic Needs: An Overview.”Wor- ld Development, 9: 167-182.

Jurado, Elsa P (1996): “Indicators of Political Opportunity and Political Welfare- Measuring Philippine Development: Report of the Social Indicators Project.” De- velopment Academy of the Philippines.

Kenya National Bureau of Statistics (2007): “Basic Report on Well-being in Kenya.” KIHBS 2005/06. ISBN-9966-767-08-8.

Kohli, A (1987): “The State and Poverty in India.” CUP.

Landell-Mills, Pierre and Ismail Serageldin (1992): “Governance and the External Factor.” In Proceedings of the World Bank Annual Conference on Development Economics 1991. Washington, D.C.: World Bank.

Mangahas, Mahar (ed.) (1976): “The Measurement of Philippine National Wel- fare.” Measuring Philippine Development: Report of the Social Indicators Project. Development Academy of the Philippines.

Nigeria Bureau of Statistics (2010): “Nigeria Poverty Profile 2010.”

Paqueo, Vicente (1976): “Social Indicators for Health and Nutrition-Measuring Philippine Development.” Development Academy of the Philippines. Root, Hilton L. (1995): “Managing Development through Institution Building.” Economics and Development Resource Center, Asian Development Bank. Occa- sional Papers No. 12.

78 The African Statistical Journal, Volume 21, September 2019 4. Statistical Indicators for Measuring Good Governance in Africa

Shangodoyin, D.K. and Lasisi, T, (2011): “The role of statistics in national development with reference to Botswana and Nigeria statistical systems.” Journal of Sus- tainable Development, Vol. 4, No. 3. 131-135.Stewart, F (2006): “The Basic Needs Approach.” In D. A. Clark (ed.), The Elgar Companion to Development Studies. Cheltenham: Edward Elgar.

Streeten, P (1982): “Basic needs and the new international economic order.” Mondes en Développment 10 (39):317-331.

Vanderboom, C.E., Vincent, A., Luedtke, C.A., Rhudy, L.M., and Bowles, K.H. (2013). “Feasibility of interactive technology for symptom monitoring in patients with fibromyalgia.” Pain Management Nursing: Official Journal of the American Society of Pain Management Nurses.

Vittachi, V. T. (1976): “A Necessary Utopia, New internationalist.”

United Nations Development Programme (1996): “Human Development Report.” New York, Oxford University Press.

United Nations Statistics Division (2016): “Demographic Statistics.” CRVS Tech- nical Report Series, Vol. 3. Weaver, B., Lindsay, B., and Gitelman, B. (2012). “Communication technology and social media: Opportunities and implications for healthcare systems.” The Online Journal of Issues in Nursing, 17(3). Vol.17 No.03.

World Bank (1992): “Governance and Development.” The World Bank: Washing- ton, D.C.

Journal africain de statistiques, numéro 21, septembre 2019 79 African Statistical Journal Call for Papers

The African Statistical Journal (ASJ) is currently accepting manuscripts for publication in French or/and English. The ASJ was established to promote the understanding of statistical development in the African region. It focuses on issues related to official statistics as well as application of statistical method-ologies to solve practical problems of general interest to applied statisticians.

In addition to individual academic and practicing statisticians, the Jour-nal should be of great interest to a number of institutions in the region including National Sta- tistical Offices, Central Banks, research and train-ing institutions and sub-regional economic groupings, and international development agencies.

The Journal serves as a research outlet and information sharing publication among statisticians and users of statistical information mainly in the Africa region. It pu- blishes, among other things: •articles of an expository or review nature that demonstrate the vital role of statistics to society rather than present technical materials,

•articles on statistical methodologies with special emphasis on applications,

•articles about good practices and lessons learned in statistical development in the region, •opinions on issues of general interest to the statistical community and users of statistical information in the African region, •notices and announcements on upcoming events, conferences, calls for papers, and recent statistical developments and anything that may be of interest to the statistical community in the region.

All manuscripts are reviewed and evaluated on content, language, and presenta- tion. The ASJ is fully committed to providing free access to all articles as soon as they are published. We ask you to support this initiative by publishing your papers

80 in this journal. Prospective authors should send their manuscript(s) toASJ-Statis- [email protected]

The ASJ is also looking for qualified reviewers. Please contact us if you are interested in serving as a reviewer.

For instructions for authors and other details, please visit our website – http:// www.afdb.org/en/knowledge/publications/african-statistical-journal/

81 Journal africain de statistiques Demande de soumission d’articles

Le journal africain de statistiques (JSA) accepte actuellement des manuscrits pour la publication en anglais ou en français. Le JSA a été établi pour favoriser la com- préhension du développement statistique dans la région africaine. Il se concentre sur des questions liées aux statistiques officielles aussi bien que l’application des méthodologies statistiques pour résoudre des problèmes pratiques d’intérêt général pour les praticiens de la statistique.

En plus des universitaires et des statisticiens de métier, le Journal devrait revêtir un grand intérêt pour les institutions de la région, notamment les offices nationaux de statistiques, les banques centrales, les instituts de recherche et les organisations économiques sous-régionaux et les agences internationales de développement.

•des articles sur le plaidoyer en matière de statistique qui démontrent le rôle essentiel des statistiques dans la société, plutôt que de présenter le matériel technique, •des articles sur les méthodologies statistiques, avec un accent particulier sur les applications, •des articles sur les meilleures pratiques et les leçons tirées sur le développement de la statistique dans la région, •des avis sur des questions d’intérêt général pour la communauté statistique et les utilisateurs de l’information statistique dans la région africaine, •des informations et des annonces sur les prochains événements, les conférences, les appels à contribution pour des papiers, et •les développements statistiques récents et tout autre aspect susceptible d’intéresser la communauté statistique dans la région.

82 Tous les manuscrits sont passés en revue et évalués sur le contenu, la langue et la présentation. Le JSA s’engage entièrement à fournir le libre accès à tous les articles dès qu’ils sont publiés. Nous vous demandons de soutenir cette initiative en publiant vos articles dans ce journal. Les auteurs éventuels devraient envoyer leur manuscrit(s) à [email protected]

Le JSA cherche également les critiques qualifiés. Veuillez nous contacter si vous êtes intéressé à contribuer en tant que critique.

Veuillez visiter notre site Web http://www.afdb.org/en/knowledge/ publications/ african-statistical-journal/ pour les instructions aux auteurs et autres détails.

83 Editorial policy

The African Statistical Journal (ASJ) was established to promote the un-derstan- ding of statistical development in the African region. It focuses on issues related to official statistics as well as application of statistical methodologies to solve practical problems of general interest to applied statisticians. Of particular interest will be the exposition of: how statistics can help to illuminate development and public policy issues like poverty, gender, environment, energy, HIV/AIDS, etc.; development of statistical literacy; tracking national and regional development agendas; development of statistical capacities and effective national statistical systems; and the de-velopment of sectoral statistics, e.g. educational statistics, health statistics, agricultural statistics, etc.

In addition to individual academic and practicing statisticians, the Journal should be of great interest to a number of institutions in the region including National Statisti- cal Offices, central banks, research and training institutions, sub-regional economic groupings, and international development agencies.

The Journal serves as a research outlet and information sharing publica-tion among statisticians and users of statistical information mainly in the African region. It pu- blishes, among other things, articles of an expository or review nature that demonstrate the vital role of statistics to society rather than present technical materials, articles on statistical methodologies with a special emphasis on applications, articles about good practices and lessons learned in statistical development in the region, opinions on issues of general interest to the statistical community and users of statistical information in the African region, notices and announcements on upcoming events, con-ferences, calls for papers, and recent statistical developments and anything that may be of interest to the statistical community in the region.

The papers, which need not contain original material, should be of general interest to a wide section of professional statisticians in the region.

All manuscripts are peer reviewed and evaluated on content, language and presen- tation.

84 Ligne éditoriale

Le Journal statistique africain a été établi pour favoriser la compréhension du déve- loppement statistique dans la région africaine. Il se concentre sur des questions liées aux statistiques officielles aussi bien que l’application des méthodologies- statis tiques pour résoudre des problèmes pratiques d’intérêt général pour les statisticiens de métier. L’intérêt particulier est de montrer comment les statistiques peuvent ai- der à mettre en exergue les problèmes de développement et de politique publique tels que la pauvreté, le genre, l’environnement, l’énergie, le VIH/ SIDA, etc.; le développement de la culture statistique ; la prise en compte des questions de déve- loppement régional et national; le développement des capacités statistiques et des systèmes statis-tiques nationaux efficaces; et le développement des statistiques sec- torielles comme les statistiques d’éducation, de santé, des statistiques agricoles, etc.

Le Journal constitue un document de recherche et d’information entre les statisticiens et les utilisateurs de l’information statistique, principalement dans la région africaine. Il publie entre autres: des articles sur le plaidoyer en matière de statistique qui démontrent le rôle essentiel des statistiques dans la société plutôt que la présentation des outils techniques, des articles sur les méthodologies statistiques, avec un accent particulier sur les applications, des articles sur les meilleures pratiques et les leçons tirées de la région, des avis sur des questions d’intérêt général pour la communauté statistique et les utilisateurs de l’information statistique dans la région africaine, des informations et des annonces sur les prochains événements, les conférences, les appels à contribution pour des papiers, et les développements statistiques récents et tout autre aspect susceptible d’intéresser la communauté statis-tique dans la région.

85 Les articles, qui n’ont pas besoin de contenir du matériel original, devraient intéres- ser une grande partie des statisticiens professionnels dans la région.

Tous les manuscrits seront passés en revue et évalués sur le contenu, la langue et la présentation.

86 Guidelines for manuscript preparation and submission

Submissions Manuscripts in English or French should be sent by email to ASJ-Statistics@ afdb. org

Title The title should be brief and specific. The title page should include the title, the author’s name, affiliation and address. The affiliation and address should be given as a footnote on the title page. If the manuscript is co-authored, the same information should be given for the co-author(s).

Abstract, Key Words, and Acknowledgments A short abstract of about 150 words must be included at the beginning of the manuscript, together with up to 6 key words used in the manuscript. These key words should not repeat words used in the title. Acknowledg-ments, if any, should inserted as a new section at the end of the paper and before the References.

Sections and Numbering Major headings in the text should be numbered (e.g. 1. INTRODUCTION). Numbe- red subheadings and sub-subheadings (e.g. 1.1 The establishment of the NSDS and 1.1.1 Bodies comprising the NSDS). Main body text in the form of paragraphs should not be numbered.

Formatting Please use minimal formatting as this will facilitate harmonization of all the papers. As your default, keep to “normal” (12 pt. Times New Roman) for main text with a single line space between paragraphs. Do not apply “body text” as an inbuilt style. The levels of heading need to be easily identifiable. We recommend all capitals bold for the first level of heading in the main text (e.g. “1. INTRODUCTION”); thereafter bold upper and lower case for subheadings (e.g. “1.1 The establishment of the NSDS”) and italic not bold (e.g. 1.1.1. Creating a culture of cooperation) for sub-subheadings. Please refer to the latest volume of the AJS as a guide.

87 House Style The Bank’s house style is U.S. rather than British spellings (e.g. “organi-zation” not “organisation”; “program” rather than “programme”, “ana-lyze” rather than “analyse” etc.). Use percent rather than per cent or % although the percentage sign should be used in figures and tables and double rather than single quotation marks. Dates should be U.S. style (e.g. December 11, 1985 not 11 December 1985).

Tables and Figures Tables and figures should be numbered and given a title. These should be referred to in the text by number (e.g. “See Table 1”), not by page or indica-tions such as “below” or “above”.

Equations Any equations in the paper should be numbered. The numbers should be placed to the right of the equation.

References A list of references should be given at the end of the paper (to precede the Annexes, if included). The references should be arranged alphabetically by surname/name of organization. Where there is more than one publication listed for an author, order these chronologically (starting with the earliest). The references should give the author’s name, year of publication, title of the essay/book, name of journal if applicable. Use a, b, c, etc. to separate publications by the same author in the same year. Titles of journals and books should be in italic; titles of working papers and unpublished reports should be set in double quotation marks and not italicized.

Examples: Fantom, N. and N. Watanabe (2008). “Improving the World Bank’s Da-tabase of Sta- tistical Capacity,” African Statistical Newsletter, 2 (3): 21–22.

Herzog, A. R. and L. Dielman (1985). “Age Differences in Response Ac-curacy for Factual Survey Questions,” Journal of Gerontology, 40: 350–367.

Kish, L. (1988a). “Multipurpose Sample Designs,” Survey Methodology, 14 (3): 19–32.

88 Kish, L. (1988b). “A Taxonomy of Elusive Populations,” in Proceedings of the Annual Meeting of the American Statistical Association. January 1988.

Parpart, J. L., M. P. Connelly, and V. E. Barriteau (eds.) (2000). Theoretical Pers- pectives on Gender and Development. Ottawa: International Development Research Center.

World Bank (2006). Statistical Capacity Improvement in IDA Countries – Progress Report. Washington, DC: The World Bank.

Cross References In the main body of the article, cross-references should be Harvard-style, e.g. (Kish, 1988a; Herzog and Dielman, 1985: 351). For cross-references to three or more authors, only the first surname should be given, followed by et al., although the names of all the authors must be provided in the References entry itself. Abbreviations ibid. and op. cit. should be avoided.

89 Instructions pour la préparation et la soumission de manuscrits

Soumission Les manuscrits en anglais ou en français doivent être envoyés à : [email protected]

Titre Le titre devrait être bref et détaillé. La page de titre doit inclure le titre du papier, le nom de l’auteur, l’affiliation et l’adresse. L’affiliation et l’adresse doivent figurer comme note de bas de page. Si le manuscrit est produit par des coauteurs, la même information doit être donnée pour les coauteurs.

Résumé, mots clés et remerciements Un résumé court d’environ 150 mots doit être inclus au début du manus-crit ainsi qu’environ 6 mots clés utilisés dans le manuscrit. Les mots clés ne doivent pas répéter les mots utilisés dans le titre. Les remerciements, s’il y en a, doivent être insérés à la fin de l’article, avant les références bibliographiques.

Section et numérotation Les principaux titres doivent être numérotés (par exemple “1. INTRO-DUCTION“). Les sous-titres et sous sous-titres numérotés (par exemple “1.1 L’élaboration de SNDS” et “1.1.1 Créant une culture de coopération”) peuvent être employés. Le corps principal du texte sous forme de paragraphes ne devrait pas être numéroté.

Formatage Veuillez utiliser le formatage minimal car ceci facilitera l’harmonisation de tous les articles. Garder par défaut le format “normal” (12 pt. Times New Roman) pour le texte principal avec l’espace d’une seule ligne entre les paragraphes. Ne pas appliquer le “corps de texte “ en tant que modèle intégré. Les niveaux du titre doivent être fa- cilement identifiables. Nous recommandons les majuscules en gras pour le premier niveau titre dans le texte principal (par exemple “1. INTRODUCTION“) ; ensuite les lettres minuscules en gras pour les sous-sections (par exemple “1.1 l’élaboration de la SNDS”) et ensuite l’italique pour les sous sous-titres (par exemple “1.1.1 Créant une culture de coopération”). Veuillez vous référer au dernier volume du JSA comme guide.

90 Tables et Figures Les tableaux et les graphiques doivent être numérotés et comporter un titre. Ceux-ci devraient être mentionnés (par exemple “voir Tableau 1” ) dans le texte par le nombre correspondant, et non par une indication de page ou par d’autres indications telles que “ci-dessous” ou “au-dessus de”.

Équations Toutes les équations dans le papier doivent être numérotées. Les nombres doivent être placés à la droite de l’équation.

Références bibliographiques Une liste de références doit être fournie à la fin de l’article (avant les annexes, le cas échéant). Les références doivent être classées par ordre alphabétique selon le nom de l’auteur ou de l’organisation. Là où il y’a plus d’une publi-cation listée pour un auteur, elles doivent être classées chronologiquement (en commençant par les plus récents). Les références doivent donner le nom de l’auteur et l’année de publication, le titre du livre, le nom du journal le cas échéant. Utiliser a, b, c, etc. pour séparer les publications du même auteur au cours der la même année. Les titres des journaux et des livres devraient être en italique ; les titres des documents de travail et des rapports non publiés devraient être placés dans de doubles guillemets et ne pas être imprimés en italique.

Exemples : Fantom, N. and N. Watanabe (2008). “Improving the World Bank’s Data-base of Sta- tistical Capacity,” African Statistical Newsletter, 2 (3): 21–22.

Herzog, A. R. and L. Dielman (1985). “Age Differences in Response Accu-racy for Factual Survey Questions,” Journal of Gerontology, 40: 350–367.

Kish, L. (1988a). “Multipurpose Sample Designs,” Survey Methodology, 14 (3): 19–32.

Kish, L. (1988b). “A Taxonomy of Elusive Populations,” in Proceedings of the Annual Meeting of the American Statistical Association. January 1988.

Parpart, J. L., M. P. Connelly, and V. E. Barriteau (eds.) (2000). Theoretical Pers- pectives on Gender and Development. Ottawa: International Development Research Center.

91 World Bank (2006). Statistical Capacity Improvement in IDA Countries – Progress Report. Washington DC: The World Bank.

Renvois Dans le corps principal de l’article, les renvois devraient suivre le modèle de Harvard, par exemple (Kish, 1988a ; Herzog et Dielman, 1985 : 351). Pour des renvois à trois auteurs ou plus, seulement le premier nom de famille devrait être donné, suivi par et al., bien que les noms de tous les auteurs doivent être fournis dans la Bibliographie elle- même. Les abréviations ibid. et op. cit. ne devraient pas être employées dans le texte ou dans les notes de bas de page.

92 Éditorial

VELOPM INE DE DEV DE EN A E N T IC LO A F R P C U F P I A N E R D M F E A U E

Q N

N T

T F N O E N D M E S P A P FR O IC EL AIN DE DEV

African Development Bank Group / Groupe de la Banque africaine de développement Avenue Joseph Anoma 01 BP 1387 Abidjan 01 Côte d’Ivoire Tel: (+225) 20 26 10 20 Internet: http://www.afdb.org Email: [email protected]

Journal africain de statistiques, numéro 21, septembre 2019 93