Consideration of observations:

• Assign names to the group in the dendrogram

• Add value ranges to characterize the levels of the variables represented on the maps

• Avoid abbreviations on the maps

• Improve the map scale for better visibility

• Explain the linear regression model

Technical summary

This project is part of a multidimensional study aimed at analyzing the reasons for underdevelopment in West Africa. It is based on an integrated methodological approach using four complementary modules, combining data collection, statistical analysis, spatial representation and documentary research. This project is part of a multidimensional study aimed at analyzing the reasons for underdevelopment in West Africa. It is based on an integrated methodological approach using four complementary modules, combining data collection, statistical analysis, spatial representation and documentary research.

Initially, primary data were collected using the KoboToolbox platform, through structured surveys administered in the field. These data were then subjected to in-depth statistical processing, in particular by applying Principal Component Analysis (PCA) using R software. This multivariate method made it possible to identify the main explanatory dimensions of underdevelopment in the study area, highlighting the most influential socio-economic variables. The spatial dimension was integrated using Geographic Information Systems (GIS) tools using ArcGIS, which made it possible to map the results, visualize regional disparities and better understand territorial dynamics related to the factors of underdevelopment. In addition, a rigorous documentary review was conducted using Zotero software, facilitating the management and organization of bibliographic references. This approach made it possible to compare field data with existing theoretical knowledge, while ensuring the scientific soundness of the analysis. This project thus illustrates the relevance of an interdisciplinary approach combining digital tools, statistics, cartography and bibliography to critically and methodically address a complex issue such as underdevelopment in West Africa . Spatial and documentary research.

Initially, primary data were collected using the KoboToolbox platform, through structured surveys administered in the field. These data were then subjected to in-depth statistical processing, in particular through the application of Principal Component Analysis (PCA) using R software. This multivariate method made it possible to identify the main explanatory dimensions of underdevelopment in the study area, highlighting the most influential socio-economic variables.

The spatial dimension was integrated using Geographic Information Systems (GIS) tools under ArcGIS, which made it possible to map the results, visualize regional disparities and To better understand territorial dynamics related to the factors of underdevelopment. In addition, a rigorous documentary review was conducted using Zotero software, facilitating the management and organization of bibliographic references. This approach made it possible to compare field data with existing theoretical knowledge, while ensuring the scientific soundness of the analysis. This project thus illustrates the relevance of an interdisciplinary approach combining digital tools, statistics, cartography, and bibliography to critically and methodically address a complex issue such as underdevelopment in West Africa.

Introduction

Underdevelopment in West Africa is a complex challenge that deeply affects the countries of the region, despite their abundant natural resources and cultural wealth. This economic and social reality has immediate and almost daily consequences on the lives of populations, marked by high levels of poverty, deep inequalities, and limited access to basic services such as health, education, and drinking water. Although West African countries have experienced progress in certain sectors, such as economic growth and improved infrastructure, underdevelopment persists, often exacerbated by a combination of internal and external factors (1) .

West Africa remains a region where the causes of underdevelopment are multiple and deeply rooted. Key factors include governance deficits, weak institutions, and ineffective natural resource management. In addition, rapid urban population growth, often without adequate planning, has contributed to unsanitary living conditions, a lack of basic infrastructure, and regional imbalances. Poor natural resource management and unequal access to public services are also key factors hindering progress. Added to this are structural challenges, such as political instability, armed conflict, and food insecurity, which further complicate sustainable development efforts (2) .

In 2021, West Africa continued to face major challenges, despite development policies implemented by international organizations and national governments. These policies were often marked by insufficient financial resources, a lack of coordination between different actors, and an excessive dependence on foreign aid. Furthermore, unplanned urbanization and population growth in cities exacerbated problems of housing, sanitation, and access to health services, creating conditions conducive to the emergence of new health and social challenges.

Principal Component Analysis (PCA) is emerging as a powerful statistical tool for exploring these complex dynamics. By reducing the dimensionality of large datasets while preserving essential information, PCA helps identify the variables and relationships that influence economic and social development in this region. This approach allows for a better understanding of the interactions between different dimensions of underdevelopment – such as poverty, unequal access to public services, political instability and natural resource management – and to identify more targeted strategies to address them (1) . Through this introduction, we will focus on how PCA can be used to analyze the factors of underdevelopment in West Africa, particularly in the areas of agriculture, health, education and governance. By highlighting the strengths and limitations of this method, this study aims to provide a theoretical and practical framework to better understand the development challenges in West Africa and to explore potential ways to overcome them.

• What are the main reasons for underdevelopment in West African countries and how do these factors interact to maintain this situation?

This question invites us to explore the root causes of underdevelopment, taking into account political, economic, social, and environmental dimensions. Analyzing the factors that generate and maintain this situation will allow us to better understand how to break this vicious circle and propose solutions adapted to the specific realities of each country in the region.

General objective:

To analyze the main causes of underdevelopment in West African countries using Principal Component Analysis (PCA) to identify key quantitative variables that influence the economic and social development of the region.

Specific objectives:

Research questions

The main research questions that motivate this study are:

• What are the most relevant variables for explaining the main causes of underdevelopment in West African countries ?

• Do the statistical distribution of variables and their distribution differ among West African countries?

1. Literary review

West Africa has significant water resources, but suffers from chronic deficits, due to the uneven distribution of rainfall and runoff in time and space, the low mobilization of potential resources and poor management of existing resources. Water infrastructure is still very poorly developed in Africa. To meet the needs of populations in terms of access to water and sanitation, improvement of food security, energy supply, environmental protection, etc., infrastructure development is necessary within a framework of consultation (3) . In Africa, access to water and control of the management of water points are at the heart of major issues and can be sources of conflict. In West Africa, the decline in rainfall levels, environmental degradation and population growth have led to a reduction in water resources. Even in regions where water resources are abundant, the issue of access is not always resolved. Tensions around water points are not only linked to the diversity of ecosystems. They are a function of the availability of hydraulic infrastructure, its state of operation and the nature of its management method. Whatever the environment, there are significant disparities in the spatial distribution of water points, within rural areas, or within cities, or even in the interstices of large cities (4) . West Africa has significant water resources, but suffers from chronic deficits, due to the unequal distribution of rainfall and runoff in time and space, the low mobilization of potential resources and poor management of existing resources. Water resource management covers diverse and complex issues, such as responding to the basic needs of populations, anticipating crises and preserving the resource, participation, cost recovery, etc. The rapid increase in the region’s population, the increase in environmental degradation and pollution, and the threat of a diminishing resource, encourage the implementation of integrated management that takes into account all uses and involves all stakeholders concerned (5)

In seven of the 16 West African countries, more than 5 million people—over 50% of their population—practice open defecation (UNICEF 2018). According to JMP data, open defecation is more prevalent among the poorest populations, with the open defecation rate in the poorest quintile reaching 92% in Benin, 86% in Niger, 84% in Togo, and 75% in Burkina Faso (JMP 2017). Open defecation is also more prevalent in rural than urban areas; in The Gambia, for example, 79.1% of the population practices open defecation in rural areas, compared to 33.3% in urban areas. This is similar in other countries such as Togo (85.3% vs. 38.2%) and Niger (87.3% vs. 37.5%) (JMP 2017). (6) . Open defecation is linked to the lack of adequate sanitation infrastructure, socio-economic factors which lead to many risks to the environment and human health such as diarrhea, typhoid fever, dysentery and other diseases that lead to high infant mortality rates in Africa. A large number of houses do not have a sewerage system, which makes it difficult to dispose of household waste. This also poses an environmental problem because the contents of pits are often discharged into neighborhoods, causing foul odors and exposing populations to unsanitary conditions (7) . West and Central Africa is the only region with a growing number of people practicing open defecation—one of the most unsafe practices, which involves using the bush, stream, local river, or the outdoors as a toilet. Sustainable access to water, sanitation, and hygiene in health centers and schools also remains a challenge. In the region, less than 50% of schools have access to water and less than 40% have access to adequate sanitation. Water, sanitation, and hygiene services in health facilities are also limited. In communities across the region, more than a third of the population still lacks access to safe drinking water, and millions drink untreated and potentially contaminated water that can cause diarrhea, a major cause of child mortality, and cholera (8).

In West Africa, the literacy rate is still low. In Burkina Faso in 2018, only 41% of people over 15 were literate. Without knowing how to read or write, many young people find themselves limited in their professional projects and their daily lives. As part of the Skills for Tomorrow program, Solidarité Laïque implemented a literacy project in Côte d’Ivoire and Burkina Faso, to give these young people the keys to their development. This program was set up to address this decline (9) . In addition to the challenge of millions of children out of school, several West African countries are faced with high illiteracy rates among their populations. In 2020, 60 million out of approximately 208 million Nigerians, or nearly 30% of the population, could neither read nor write in any language, according to the National Commission for Mass Literacy. The literacy rate is equally low in Mali (31%) and Niger (38%). It is 47% in Benin, but 80% in Ghana, a good example in the region (10) .

The Human Development Index (HDI) is a summary measure of average achievement in key dimensions of human development: a long and healthy life, knowledge, and a decent standard of living. It is a standard way to measure well-being. It is used to distinguish whether a country is developed, developing, or underdeveloped, and also to measure the impact of economic policies on quality of life. Countries fall into four broad categories based on their HDI: very high, high, medium, and low human development. Currently, Mauritius is the only African country that falls into the very high human development category (11)

In West African countries, the situation is generally more complex: the dynamism of the informal sector contrasts with the weakness of the modern sector. Moreover, in these countries, smallholders coexist with well-organized networks and often politically well-connected “large informal” businesses. The informal sector accounts for more than 50% of the overall value added of GDP in low-income countries, more than 80% of total employment, and more than 90% of newly created jobs in these countries. Its impact on employment opportunities, productivity, tax revenues, and economic growth is therefore significant. At the same time, the informal sector poses enormous knowledge challenges since, by definition, some, if not most, aspects of the informal sector are poorly documented or undocumented. Understanding the dynamics of the informal sector is crucial for achieving structural transformation in least developed economies, moving them away from subsistence and informal agricultural activities towards more productive activities, growth and better quality jobs (12) . Ineffective governance and widespread corruption hinder development by diverting public resources and weakening state institutions. These practices undermine the confidence of citizens and investors, thus limiting economic and social opportunities (13) . Debt servicing absorbs a significant share of national budgets, reducing the funds available for investment in infrastructure and social services (14) . The concentration of economies on a few export commodities makes countries vulnerable to fluctuations in global prices, limiting their ability to generate stable incomes and invest in other sectors (15) .

2. Presentation of the study area

This study focuses on a selection of ten West African countries: Burkina Faso, Cape Verde, Ivory Coast, Gambia, Guinea, Ghana, Mali, Mauritania, Niger, and Nigeria. These states were selected as units of analysis as part of our research on the explanatory factors of underdevelopment in certain countries of the subregion. The selection was based on two main criteria: data availability and the representativeness of the three major West African linguistic areas, namely French-speaking (Burkina Faso, Ivory Coast, Guinea, Mali, Mauritania, Niger), English-speaking (Gambia, Ghana, Nigeria), and Portuguese-speaking (Cape Verde) countries. This sample provides a relevant framework for a comparative analysis of development dynamics in the West African region. The following map provides a geographical location for the countries concerned.

Figure 1 : Map of the study area

3. Materials and methods used

To carry out our study , we started by collect data through the Our World in Data database , then we processed and analyzed this data . Finally, the results obtained from this analysis were interpreted spatially at the scale of the 10 West African countries using maps.

3.1 Material used

  • R Studio : R Studio is an integrated development environment (IDE) , designed specifically for users of the R programming language. It has given us a complete suite of tools for programming, data analysis, visualization and reporting.

  • ArcGIS (10.4.1 ) :

ArcGIS is a free and open source geographic information system (GIS) software that allows you to create, visualize, edit, analyze, and publish spatial and cartographic data. It has been useful for us in the spatial interpretation of data after analysis .

  • KoboToolbox :

KoboToolbox is an online platform that allows you to collect data simply and efficiently . Whether you are conducting surveys, polls or gathering information, it offers a complete and customizable solution .

  • Zotero :

Zotero is a free and open-source reference management software that helps researchers and students organize and cite their sources. It allows users to collect, manage, and organize bibliographic data, including scholarly articles, books, websites, and other research materials.

  • Office tools: Excel and Word

They were used for formatting, organizing and writing the results .

##3.2 Method

For data collection , we would have opted for a qualitative approach by favoring surveys, which may be an appropriate approach for a population with a low literacy rate . Here are the main elements that we retained concerning the methodology:

• Survey method :

We chose to conduct structured interviews using an interview guide and pre-written questionnaires , administered during door-to-door visits . This method facilitates the collection of information in a systematic and standardized manner .

• Structure of the questionnaire :

The questionnaire developed using KoboCollect software is divided into four parts :

• Information about the investigator, • Location information, • Household information, • Data collection variables.

The data used for processing were extracted from the databases of the following sites :

• https://ourworldindata.org • fao.org • worldbankdata.org

4 Results

As part of our study project, the downloaded data was processed on Excel in order to extract the necessary variables. The variables retained are as follows: (AE) Access to electricity (% of the population); GDP per capita (GDP); Literacy rate (Lit); (GDP) GDP (in constant 2021 PPP Purchasing Power Parity in international dollars); Human Development Index (HDI); Share of the population practicing open defecation (DAL); Share of the population with access to at least basic drinking water services (Wat); (Rech)) Expenditure on research and development (% of GDP); (D_Lit) Public expenditure on education as a percentage of GDP ; (Choma) Total unemployment rate (% of the labor force, modeled estimate from the ILO); (Live) Life expectancy at birth (all sexes combined)

Table of variables

setwd("C:/Users/Tangu/Desktop/Groupe4 RTI/Celia RTI")

data = read.csv(file ="data.csv", header = TRUE, sep = ";", quote = "\"",
                  dec = ",", row.names = 1)
data[,1:10]
##                 AE     PIB   Lit  IDH   DAL   Wat Rech D_lit Choma  Live
## Burkina Faso  18.5 2380.93 34.49 0.45 36.92 48.54 0.20  5.15  4.99 60.45
## Cape Verde    93.7 7207.87 91.00 0.65 10.45 89.18 0.07  7.58 14.66 73.82
## Cote d'Ivoire 69.9 5787.82 50.00 0.53 22.46 72.63 0.07  3.79  2.56 60.14
## Gambia        62.2 2702.32 58.67 0.49  0.32 84.48 0.07  2.89  5.86 64.42
## Ghana         85.4 6412.69 80.38 0.60 17.51 86.32 0.38  3.41  3.29 64.31
## Guinea        44.7 3630.99 45.33 0.47  8.97 68.92 0.04  2.43  5.99 59.35
## Mali          50.6 2348.88 31.00 0.41  5.68 80.75 0.31  3.95  3.53 58.86
## Mauritania    45.4 5963.24 66.96 0.54 29.25 74.30 0.01  1.80 11.01 66.79
## Niger         18.7 1615.69 38.10 0.39 78.14 49.13 0.04  3.82  1.25 59.89
## Nigeria       55.4 5410.69 63.16 0.54 23.37 86.54 0.28  0.36  6.45 53.07

5 Method of analysis

5.1 Data analysis with Cape Verde

Figure 2:Individual PCA

Figure 3 : Cluster Dendrogram

Figure 4 : PCA Biplot

According to the figures above there is a correlation between the variables and the individuals The study will focus on two areas: •The Dim1 axis (54.3%) explains 54.3% of the total variance. •The Dim2 axis (19.6%) explains 19.6% of the total variance. Together, these two axes explain 73.9% of the variance, which is a good representation of the data. The PCA analysis of figurexx indicates that: • countries close to each other have similar profiles. •Remote countries are very different. •Cape Verde is atypical ( forget) because it is very far from other countries, which indicates a particular profile. PCA analysis of Figure 4 indicates that: •A variable that points in one direction indicates that it is highly correlated with that dimension. In our case variables such as: • GDP, HDI, AE, Wat are strongly correlated with Dim1. •DAL and D_lit are more correlated with Dim2.

Cape Verde is very far from the other countries, meaning it has very different characteristics. It is positioned extremely to the right on Dim1, which indicates that it has very high values of the variables that are correlated with this axis (GDP, HDI, AE, Wat). This suggests that Cape Verde has a higher level of development than the other countries represented.

Countries like Mali, Niger, and Guinea are distant from Cape Verde, indicating that they have very different profiles (likely lower development indices). Cape Verde is an outlier, probably because it has a high GDP and HDI relative to the other countries. Nigeria, Ghana, and Gambia are relatively close, suggesting that they have similar profiles. In summary, Cape Verde stands out strongly and seems to be the country with the best development among those analyzed.

5.2 Data analysis without Cape Verde

Correlation matrix and circle

Figure 5 : Correlation matrices

b. Cercle de correlation

Figure 6 : Correlation circle

This correlation matrix presents the relationships between different variables through a color gradient ranging from red (strong negative correlation) to blue (strong positive correlation). A strong positive correlation, close to 1, indicates that the two variables move in the same direction, while a strong negative correlation, close to -1, means that one variable decreases when the other increases; conversely, a weak or zero correlation, close to 0, shows that there is no significant relationship between the variables.

•   **Strong positive correlations:**

The correlation coefficient of GDP and HDI being 0.94 implies that countries with high GDP generally have high HDI. The correlation coefficient of literacy rate (Lit) and GDP being 0.92 implies that a high literacy rate is strongly related to a higher GDP. Access to Electricity (AE) and HDI with a correlation coefficient of 0.77 explains that access to electricity is a key factor in human development. Access to drinking water (Wat) and HDI with a correlation of 0.63 shows that better accessibility to drinking water is associated with higher human development.

•   **Strong negative correlations :**

DAL and Wat: The correlation coefficient of -0.75 indicates that high DAL index (probably a measure of disease) is associated with reduced access to drinking water. D_lit and Choma: The correlation coefficient of -0.59 indicates a moderate inverse relationship between the literacy rate and the unemployment rate. This means that in countries or regions with higher literacy rates, the unemployment rate tends to be lower. This relationship can be explained by the fact that a high level of literacy improves employability by facilitating access to education and skills needed for the labor market. Literate people have more job opportunities and adapt better to job requirements, thus reducing the unemployment rate.

The socio-economic development variables (HDI, GDP, Lit, AE, Wat) are highly correlated with each other which form a development block. Variables linked to difficult living conditions (DAL, D_lit) are often in opposition to development indicators.

This matrix shows that economic and human development are closely linked to access to infrastructure (energy, water) and education. Conversely, poor sanitation and limited access to resources are associated with lower development.

In summary, policies to improve GDP, education, and basic infrastructure (water, energy) can have an overall positive impact on human development and the reduction of health problems.

Inertia graph Figure 7 : Inertia Graph

The inertia graph is a tool used in Principal Component Analysis (PCA) to determine the optimal number of dimensions to retain.

The first two dimensions (Dim1 and Dim2) together explain 67.1% of the total variance:

• Dim1 49.4% of the variance explains that this first dimension captures almost half of the information present in the data. It is therefore the most important for summarizing the main trends. •Dim2 17.7% of the variance explains that it makes a significant contribution but much less than Dim1, which indicates that it represents a second source of variability in the data.

With 67.1% of variance explained, these two dimensions already provide a good representation of the data, although some information remains unexplained. In practice, a visualization on a plane (Dim1-Dim2) would provide a reliable projection of individuals and variables, which is often sufficient to interpret the underlying structures.

Limiting yourself to Dim1 and Dim2 allows for an efficient data synthesis by retaining the main sources of variation. This allows major trends to be identified without burdening the analysis with less significant dimensions.

Contribution of individuals

• Dimension 1

Figure 8 : contribution of individual to Dim1

The contribution of individual countries to PCA dimension 1 indicates that Niger has the highest contribution exceeding 30%, followed by Ghana and Burkina Faso , which also contribute significantly (over 15%). Nigeria and Mauritania have a moderate contribution. Mali, Gambia, Ivory Coast and Guinea have a much lower contribution. A dotted red line is placed around 10%, probably as a threshold of significance. Niger, Ghana and Burkina Faso are the main contributors to this dimension, while Guinea and Ivory Coast have a marginal influence.

• Dimension 2

Figure 9 : contribution of individual to Dim2

The graph shows the individual contributions of different countries to dimension 2 (Dim-2) •Mauritanian has the highest contribution (over 40%), which means that it strongly influences this dimension. •Mali follows with a significant contribution (around 25%). •Ghana has a moderate contribution (about 10%). •The other countries (Nigeria, Guinea, Niger, Ivory Coast, Gambia, Burkina Faso) have low or even negligible contributions. The strong contribution of Mauritania and Mali to Dim-2 indicates that they are probably the most representative or the most differentiating on this dimension. This may mean that these countries share specific characteristics that strongly influence this dimension, or that they are clearly distinguished from other countries on the variables considered. Countries with weak contributions (below the red line) do not strongly influence this dimension, either because they are similar to other countries or because they do not have distinctive characteristics for this dimension.

Contribution of variables

• Dimension 1

Figure 10 : contribution of variables to Dim1

This figure represents the contribution of different variables to the first dimension (Dim-1) of Principal Component Analysis (PCA). The higher a variable’s contribution, the more influential it is in explaining the variance captured by this first dimension.

•HDI (Human Development Index) • Literacy rate or level of education • Wat (Access to drinking water) • AE (Access to electricity) • GDP (Gross Domestic Product per capita)

These variables are key indicators of a country’s economic and social development. The first dimension of the PCA therefore appears to be strongly associated with the level of human development and basic infrastructure . A country with a high HDI, good access to water and electricity, and a higher GDP will likely have a strong contribution to this dimension.

• D_lit (Education-related expenditure) • DAL (Open Defecation Rate) These variables are related to the health sector and literacy. Although they play a role in explaining Dim-1, they are less influential than the economic development and infrastructure variables. This suggests that health and literacy are important, but secondary, factors in this dimension.

• Choma (Unemployment Rate) • Rech (Investment in Research and Innovation • Live (Life expectancy)

These variables have a weaker influence on the first dimension of the PCA. This may mean that the main dynamics captured by Dim-1 are not directly related to the labor market, research, or longevity, but rather to more fundamental aspects such as economic development and access to essential services.

The first dimension of the ACP appears to be socio-economic development. Countries with a high HDI, good access to water and electricity, and a high GDP are better positioned on this dimension. On the other hand, aspects such as unemployment, research, or life expectancy play a less important role in this classification.

• Dimension 2

Figure 11 : contribution of variables to Dim2

The figure illustrates the contribution of variables to the second dimension (Dim-2) of Principal Component Analysis (PCA). Variables with a high contribution influence this dimension more. - Most contributing variables

The variables with the highest contribution to Dim-2 (above the red line) are:

• Rech (Investment in Research and Innovation)

• Choma (Unemployment Rate)

• Live (Life expectancy)

This dimension appears to capture aspects related to innovation, the labor market and quality of life . A country with high investment in research and innovation tends to rank higher on this dimension. Unemployment , being a key factor, suggests that this dimension also reflects economic and social differences. Life expectancy also plays an important role, which could indicate a relationship between the level of development and the overall well-being of the population.

- Moderately contributing variables

• AE (Access to electricity) • Wat (Access to drinking water) • D_lit (Education-related expenses) • DAL (Open Defecation)

These variables are related to infrastructure and health . Their contribution suggests that Dim-2 also captures disparities in access to essential services and health .

- Low contributing variables

• Literacy rate or level of education

• GDP (Gross Domestic Product per capita)

• HDI (Human Development Index)

Unlike Dim-1, here economic and human development have a lesser impact . This means that Dim-2 could reflect more structural factors related to employment, innovation and quality of life , rather than the level of wealth of a country. The second dimension of the ACP appears to be one of innovation and socio-economic well-being .

Factorial Plan and Quality of Representation :

Figure 12 : PCA variables

This graph is a correlation circle from a Principal Component Analysis (PCA) . It allows you to visualize the links between variables and their contribution to the principal axes ( Dim1 and Dim2 ).

• Dim1 (49.4%) : It represents almost 50% of the total information , which means that it is the main direction of variation of the data.

• Dim2 (17.7%) : It represents 17.7% of the information , i.e. a second important direction, but less influential than Dim1.

- Dim1 (49.4%) – Socio-economic development factor

The variables strongly projected on Dim1 (horizontal axis) are:

✅ Positively correlated (right):

• GDP (Gross Domestic Product)

• HDI (Human Development Index)

• Literacy Rate

• Wat (Access to drinking water)

• AE (Access to electricity)

  Dim1 appears to represent an axis of socio-economic development. Countries or regions well positioned on this dimension have a high standard of living, good access to infrastructure and high literacy.

✅ negatively correlated (left):

• Rech (Research and innovation expenditure)

• DAL (Open Defecation)

• D_lit (Education-related expenses) Innovation and research appear to play opposite roles on this axis, which could indicate that developing countries prioritize basic infrastructure first before investing heavily in Research and Development.

Dim2 (17.7%) – Factor of innovation and socio-economic well-being.

✅ negatively correlated (low):

• Research and Innovation Spending

Countries that invest heavily in **research and innovation** have different socio-economic dynamics, and this could be in contrast to countries where the labor market is more unstable.

This means that countries or regions well positioned on Dim1 are economically advanced, while those that stand out on Dim2 can be characterized by issues related to quality of life and employment.

Figure 13 : PCA Biplot

This biplot is derived from a Principal Component Analysis (PCA) and allows you to simultaneously visualize:

• The variables (blue arrows) that influence the main axes. • Individuals (countries in red) positioned according to their values on these dimensions.

Dim1 (49.4%) : It captures almost half (49.4%) of the variance of the data and Dim2 (17.7%) : It explains an additional part of the variance (17.7%). These axes make it possible to classify countries according to structural factors.

Right (positive on Dim1): Countries with good socio-economic indicators

• Variables: GDP, HDI, Literacy (Lit), Access to drinking water (Wat), Access to electricity (AE).

• Countries concerned: Nigeria, Ghana, Ivory Coast.

• Interpretation: These countries have relatively high indicators of economic and infrastructure development.

Left (negative on Dim1): Countries with weaker indicators

• Variables: DAL (years of life lost), D_lit (availability of hospital beds).

• Countries concerned: Niger, Burkina Faso, Mali.

• Interpretation: These countries are characterized by low socio-economic development, limited infrastructure and reduced access to essential services.

Top (positive on Dim2): Countries with issues related to unemployment and life expectancy

• Variables: Unemployment (Choma), Life expectancy (Live).

• Countries concerned: Mauritania, Gambia.

• Interpretation: These countries may have high unemployment rates and longer life expectancy.

Bottom (negative on Dim2): Countries with high investment in research and weak medical infrastructure

• Variables: Search (Rech), Bed availability (D_lit).

• Countries concerned: Mali, Niger.

• Interpretation: These countries may have little access to care but also limited effort in research and innovation.

Nigeria and Ghana: Strong economic development, access to infrastructure, but with issues of wealth distribution.

Burkina Faso and Niger: Weak economic development and worrying health indicators.

Mauritania: Unique position, combining high unemployment and high life expectancy.

Guinea and Ivory Coast: Intermediate countries, close to the origin of the axes, without marked distinction in dimensions.

#6 Classification

Figure 14 : Cluster Dendrogram

Figure 15 : Factorial plan

Figure 16 : Hierarchical tree on the plan

This dendrogram is derived from an ascending hierarchical classification (CHA) and allows countries to be grouped according to their similarities on several socio-economic variables.

The tree reveals three main groups:

Cluster 1 (Moderately developed countries,in red) includes Mali, Ivory Coast, Guinea and Gambia

Cluster 2 (Underdeveloped countries,in green) includes Mauritania, Nigeria and Ghana .

Cluster 3 (Emerging countries,without specific color) includes Niger and Burkina Faso The height of mergers indicates the similarity between countries:

o Niger and Burkina Faso are very close , which suggests strong common characteristics.

o The red cluster (Mali, Ivory Coast, Guinea, Gambia) is merged at an intermediate height , indicating moderate similarities.

o The final clustering occurs at a high height , suggesting that the three major groups identified are quite distinct from each other.

The division of the tree into three groups appears consistent , indicating that these countries have significant commonalities within each cluster.

#7 Linear regression

Figure 17 : Regression linear

The graph shows that the model predictions follow a linear trend with respect to the actual values. However, there are significant deviations for some points, indicating an imperfect fit. Blue line: Regression line that represents the relationship estimated by the model. Black dots: Actual observations compared to predictions. The explanatory variables are: GDP, HDI and Wt. The model of the equation is: Y=-76.31+6.05〖10〗^(-5) PIB+120.98IDH+0.92Wat

•R² (R-squared): 0.8228

•This means that 82.28% of the variance in the actual values is explained by the model. This is a good fit. •Adjusted R²: 0.7165

•Adjusted to take into account the number of explanatory variables. It is slightly lower, which is normal. •F-statistic: 7.741, p = 0.02515

•The overall model is significant at the 5% level, meaning that there is a linear relationship between the independent variables and the dependent variable. 3. Interpretation of coefficients:

•GDP (Gross Domestic Product): Very small and insignificant effect on the dependent variable. This suggests that GDP does not have a major impact on the prediction.

•HDI (Human Development Index): Positive effect, but not significant. This indicates a possible relationship, but not strong enough to be proven with this data.

•Wat: Positive and significant effect at the 5% threshold, suggesting that an increase in this variable is associated with an increase in the target variable. The model appears to be generally effective with a good R^2 . However, only Wat has a significant effect. It would be useful to explore other explanatory variables or improve data quality to obtain more robust results.

#8 Representation of study variables on the map according to their level of importance for different individuals

the variables used in our study on (10) countries are:

Figure 18 : Access electricity

Figure 19 : Literacy rate

Figure 20 : Open defecation

Figure 21 : GDP percapita

Figure 22 : Human development indices

Figure 23 : Public depth in education as percentage of GDP

Figure 24 : Deepness in research and development

Figure 25 : Population with access to at least basic drinking water services

Figure 26 : Total unemployment rate

Figure 27 : Life expectancy at birth

Figure 28: Cluster dendrogramme

#9 Discussion

The positive correlations observed between certain key variables provide a better understanding of the structural factors of underdevelopment in West Africa. By analyzing these links, we can identify the levers of action needed to stimulate human and economic development in the region.

The variables that best explain the reasons for underdevelopment in some West African countries shows that with the correlation coefficient of 0.94 between the Gross Domestic Product (GDP) and the Human Development Index (HDI) shows that countries with a high GDP generally have a higher HDI. This is explained by the fact that the wealth produced in a country allows for investment in social infrastructure, such as education, health and access to basic services, which are essential components of human development. According to Sachs and Warner (1997) , countries rich in natural resources but with poor economic governance tend to remain underdeveloped despite high levels of GDP (16) .

The correlation of 0.92 between literacy rate and GDP indicates that countries with high literacy rate tend to have higher GDP. This relationship highlights the crucial role of education in economic growth. According to Hanushek and Woessmann (2008) , an improvement in the quality of education and an increase in the skills of workers are determining factors for sustained economic growth (17) .

The correlation coefficient of 0.77 between access to electricity and the HDI highlights the importance of electrification in improving people’s living standards. Access to electricity is a key factor in development. However, in West Africa, more than 50% of the population does not have access to electricity , according to the World Bank (2021) (18) . This situation severely limits the potential for economic and human development.

With a correlation coefficient of 0.63 between access to drinking water and HDI, it is evident that improving drinking water infrastructure contributes to human development . However, nearly 30% of the population in West Africa still does not have access to an improved drinking water source , according to UNICEF (2022) (19) .

The strong positive correlations between GDP, HDI, literacy, access to electricity and access to drinking water demonstrate that underdevelopment in West Africa is largely due to structural inadequacies in education, access to basic services and wealth distribution .

Underdevelopment in West African countries is a multidimensional phenomenon marked by inadequacies in infrastructure, education, and access to basic services. Analyzing the negative correlations between open defecation (OD) and access to drinking water (Wat), as well as the literacy rate (D_lit) and the unemployment rate (Choma), highlights key factors that hinder socioeconomic development in the region.

A correlation coefficient of -0.75 indicates that a high rate of open defecation is strongly associated with low access to safe drinking water . This reveals a deficit in sanitation infrastructure and water resource management. In many West African countries, particularly in rural areas, access to sanitation infrastructure is limited. According to UNICEF ( 2021 ), nearly 30% of the West African population still practices open defecation , due to the lack of latrines and appropriate sanitation facilities. This situation is directly linked to a lack of access to safe drinking water , which prevents the maintenance of adequate hygiene conditions.

A correlation coefficient of -0.59 indicates that a higher literacy rate is associated with a lower unemployment rate . This means that education plays a key role in job placement and reducing unemployment . In developing economies, literacy and training are key drivers of employment . Education provides the skills needed to access economic opportunities, while low literacy limits career prospects. P

roposed solutions: The strong positive correlations between GDP, HDI, literacy, access to electricity and access to drinking water demonstrate that underdevelopment in West Africa is largely due to structural inadequacies in education, access to basic services and wealth distribution . To overcome this situation, several actions are necessary:

-Invest massively in education and vocational training Education is a fundamental driver of economic and social development. A literate and skilled population is better able to generate wealth and improve its standard of living.

• Increase public spending on education (reach at least 6% of GDP as recommended by UNESCO) .

• Develop vocational and technical training programs adapted to the needs of the labor market.

• Encourage girls’ education and combat educational inequalities.

• Strengthen the quality of teaching by training teachers and integrating new technologies into classrooms.

• Develop renewable energies (solar, hydroelectricity, biomass) to ensure sustainable and affordable electrification.

• Encourage public-private partnerships (PPPs) to accelerate rural electrification projects.

• Promote the implementation of mini-electrical grids and off-grid solutions to reach remote areas.

• Invest in modern drinking water and sanitation infrastructure.

• sustainable water resource management programs to meet the challenges of climate change.

• Promote awareness campaigns on hygiene and the rational use of water.

• Develop the manufacturing sector and agro-industry to process raw materials locally.

• Supporting the growth of small and medium-sized enterprises (SMEs) through better access to financing.

• Encourage technology incubators and startups to capture digital innovation.

Conclusion

The persistent underdevelopment of West African countries is the result of a complex interweaving of economic, social, political, institutional, and environmental factors. Through the analysis of empirical data from reliable sources such as the World Bank, the FAO, and Our World in Data, it is clear that variables such as poor access to electricity, inadequate basic infrastructure (drinking water, sanitation), low literacy rates, limited public spending in key sectors such as education, health, and research, as well as a high unemployment rate, contribute significantly to maintaining precarious living conditions for a large part of the population.

From a structural perspective, the unequal distribution of wealth, excessive dependence on raw materials, political instability, poor governance, and recurring conflicts further weaken the prospects for sustainable development. Added to this are the exacerbating effects of climate change, particularly on agricultural systems, and rapid population growth, which puts continued pressure on natural resources and basic social services.

Given this reality, it is essential to rethink development policies with a holistic, inclusive, and sustainable approach. It is crucial to strengthen institutional capacities and invest in education, health, renewable energy, and research, while promoting better governance, transparency, and citizen participation. Development can only be sustainable if it places people at the center of priorities, guaranteeing access to fundamental rights for all.

Thus, although the challenges are numerous, West African countries possess considerable potential. A strategic mobilization of internal and external resources, accompanied by strong political will and collective commitment, can pave the way to a more prosperous and equitable future for the region.

Bibliographic references

  1. D MS. Poverty and Underdevelopment in West African Countries: A Multivariate Analysis. African Journal of Economics and Politics. [Internet]. 2019. Available at: https://www.jstor.org/stable/90005659 12.

  2. L M. Economic policies and the fight against underdevelopment in West Africa. Journal of African Economic Policy. https://academic.oup.com/jae/article/18/3/ /5556530 567;

  3. THE COMMISSION, A COMISSÃO, LA COMMISSION. WATER RESOURCES POLICY FOR WEST AFRICA [Internet]. Available at: https://faolex.fao.org/docs/pdf/Eco181511e.pdf

  4. Catherine Baron and Alain Bonnassieu. The Challenges of Access to Water in West Africa: Diversity of Governance Modes and Conflicts of Use [Internet]. Report No.: 156. Available at: https://shs.cairn.info/article/MED_156_0017

  5. WEST AFRICAN WATER RESOURCES POLICY. Available at: https://afwasakm.afwasa.org/fr/produit/politique-des-ressources-en-eau-de-lafrique-de-louest/

  6. Issa Wone, Assane Seck University of Ziguinchor. Leadership of local authorities in sanitation and hygiene in West Africa. Rapid scoping desk study.

  7. Abdou Salam Fall (IFAN UCAD), Aminata Tooli Fall (IFAN UCAD) and Rokhaya Cissé (IFAN, UCAD), Laurent Vidal (IRD). Sanitation and Hygiene in West and Central Africa [Internet]. Available at: https://horizon.documentation.ird.fr/exl-doc/pleins_textes/divers18-01/010071753.pdf

  8. UNICEF. In West and Central Africa, we help provide water, sanitation, and hygiene skills to the poorest communities, schools, and health facilities—including in emergency situations. Available at: https://www.unicef.org/wca/fr/what-we-do/wash

  9. Secular Solidarity. In West Africa, literacy as a professional lever. Available at: https://www.solidarite-laique.org/informe/en-afrique-de-louest-lalphabetisation-comme-levier- professionnel/#:~:text=En%20Afrique%20de%20l%27Ouest%2C%20le%20taux%20d%27alphab%C3%A9tisation

10 RFI. Literacy a priority in West Africa. Available at: https://www.rfi.fr/fr/podcasts/%C3%A7a-fait-d%C3%A9bat-avec-wathi/20240908-l-alphab%C3%A9tisation-une-priorit%C3%A9-en-afrique-de-l-ouest

  1. gifex. African countries by Human Development Index. Available at: https://gifex.com/fr/fichier/pays-africains-par-indice-de-developpement-humain/

  2. Omar Thiam. The Informal Sector in West Africa: Issues and Perspectives [Internet]. Available at: https://shs.cairn.info/article/MSS_025_0118

13 Mayekoo. The Main Causes of Poverty in Africa [Internet]. Available at: https://mayekoo.com/fr/blogs/news/causes-of-poverty-in-africa

  1. The world. Senegal in search of a “new political model” with an unprecedented distribution of power. Published November 15, 2024 at 9:00 a.m. [Internet]. Available at: https://www.lemonde.fr/idees/article/2024/11/15/le-senegal-en-quete-d-un-nouveau-modele-politique-avec-une-repartition-inedite-du-pouvoir_6394788_3232.html

  2. Julien Bouissou. In Africa, the “economy of discontent” is gaining ground. Available at: https://www.lemonde.fr/economie/article/2024/11/20/en-afrique-l-economie-du-mecontentement-gagne-du-terrain_6404737_3234.html

  3. Sachs, JD, & Warner, AM Natural Resource Abundance and Economic Growth. Harvard University. 1997;(14).

  4. Hanushek, EA, & Woessmann, L. The Role of Cognitive Skills in Economic Development. Journal of Economic Literature. 2008;(15).

  5. World Bank. Africa Energy Outlook: Electrification Challenges and Opportunities. 2021;(16).

  6. UNICEF. State of the World’s Children Report: Access to Safe Drinking Water. 2022;(17).

Lien du Guide d'entretien
<https://kf.kobotoolbox.org/#/forms/aZvrBCZqJoSD3KyRLTTTUR> Lien du
Questionnaire
<https://kf.kobotoolbox.org/#/forms/a2Q5LPZQmSyNyrviN5JpVG>